Dell POWERVAULT MD3620F User Manual

Dell PowerVault MD3600f and  
MD3620f Storage Arrays  
Owner’s Manual  
Regulatory Model: E03J Series and E04J Series  
Regulatory Type: E03J001 and E04J001  
Contents  
1
Introduction  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
About This Document  
19  
Inside the Box of the Dell PowerVault MD3600f  
Series Storage Array.  
Dell PowerVault MD3600f  
Series Storage Array .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20  
.
.
.
.
.
.
.
.
20  
Dell PowerVault Modular  
Disk Storage Manager .  
.
.
.
.
.
.
.
.
.
.
.
20  
Other Information You May Need .  
2
Planning: About Your Storage Array  
Overview .  
.
.
.
.
.
.
.
.
Hardware Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
24  
Front-Panel Features and Indicators  
Hard–Drive Indicator Patterns.  
Power Supply and Cooling Fan Features .  
Power Indicator Codes .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
28  
29  
30  
.
.
.
.
.
.
.
.
.
3
Planning: RAID Controller Modules  
.
.
.
.
.
.
.
.
31  
RAID Controller Modules  
.
.
.
.
.
.
.
.
.
.
.
.
31  
Contents  
3
RAID Controller Module Connectors and Features .  
.
.
32  
RAID Controller Module—Additional Features .  
.
.
.
.
.
.
.
.
35  
Battery Backup Unit  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
System Password Reset  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
35  
36  
Cache Mirroring.  
.
.
.
.
.
.
.
.
.
.
.
Write–Back Cache  
Write–Through Cache  
SFP Transceivers, Fiber Optic, and SAS Cables.  
Interoperability of 2 Gbps, 4 Gbps,  
.
.
.
.
.
.
.
.
36  
and 8 Gbps Devices  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
38  
4
Array Terms and Concepts  
.
.
.
.
.
.
.
.
.
.
.
Physical Disks, Virtual Disks, and Disk Groups .  
.
.
.
.
.
.
.
.
.
.
.
.
39  
40  
40  
Physical Disks.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Physical Disk States  
Self-Monitoring Analysis and  
Reporting Technology.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
41  
42  
Virtual Disks and Disk Groups  
Virtual Disk States  
.
.
.
.
.
.
RAID Levels.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
43  
RAID Level Usage.  
43  
Segment Size.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
45  
45  
Virtual Disk Operations  
4
Contents  
Virtual Disk Initialization  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
45  
Background Initialization .  
Foreground Initialization .  
Consistency Check .  
.
.
.
.
.
.
.
.
.
Media Verification  
Cycle Time .  
.
.
.
.
.
Disk Group Operations.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
48  
49  
RAID Level Migration .  
Segment Size Migration  
Virtual Disk Capacity Expansion .  
Disk Group Expansion  
Disk Group Defragmentation .  
Disk Group Operations Limit  
.
.
RAID Background Operations Priority  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Disk Migration  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Disk Roaming .  
Advanced Features  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
53  
55  
55  
56  
57  
Host Server–to–Virtual Disk Mapping .  
Host Types  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Snapshot Virtual Disks .  
.
.
.
.
.
Snapshot Repository Virtual Disk.  
Virtual Disk Copy .  
.
.
.
.
.
.
.
.
Virtual Disk Recovery.  
.
.
.
.
.
.
Using Snapshot and Disk Copy Together.  
Multi–Path Software.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
57  
Preferred and Alternate  
Controllers and Paths.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
57  
58  
Virtual Disk Ownership .  
Contents  
5
Load Balancing.  
.
.
.
.
.
.
.
.
.
.
59  
Monitoring MD3600f Series System Performance  
5
6
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
User Interface  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
63  
64  
Enterprise Management Window  
Array Management Window .  
Configuration: About  
Your Storage Array  
.
.
.
.
.
.
.
.
.
.
Storage Arrays .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Locating Storage Arrays  
Setting a Password .  
.
.
Adding/Editing a Comment to an  
Existing Storage Array  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
76  
.
Enabling Premium Features.  
Failover Alert Display .  
Changing the Cache Settings  
on the Storage Array .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
77  
77  
Changing Expansion Enclosure ID Number  
Changing the Enclosure Order  
in the Physical Pane  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
78  
Configuring Alert Notifications  
78  
6
Contents  
Configuring E–mail Alerts.  
.
.
.
.
.
.
.
.
.
.
.
.
.
81  
Configuring SNMP Alerts.  
Battery Settings .  
.
.
.
.
.
.
.
.
.
Module Clocks  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
84  
7
8
.
.
.
.
.
.
.
.
.
Enabling or Disabling the Event Monitor .  
.
.
.
.
.
.
.
86  
Windows .  
Linux  
.
.
.
.
.
.
.
.
.
.
.
.
.
Configuration: About Your Host  
.
.
.
.
.
.
.
.
87  
Configuring Host Access  
Using the Mappings Tab .  
Defining a Host .  
.
.
.
.
Removing Host Access  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Managing Host Groups  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
92  
92  
93  
93  
Creating a Host Group  
Moving a Host to a Different Host Group.  
Removing a Host Group.  
Host Topology.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Starting or Stopping the Host Context Agent.  
I/O Data Path Protection .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
94  
95  
Managing Host Port Identifiers .  
Contents  
7
9
Configuration: Disk Groups  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Creating Disk Groups and Virtual Disks .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
102  
102  
Creating Disk Groups .  
Locating a Disk Group  
Creating Virtual Disks .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Changing the Virtual Disk  
Modification Priority  
.
.
Changing the I/O Type.  
.
Choosing an Appropriate Physical Disk Type .  
.
.
.
.
Physical Disk Security With Self Encrypting Disk  
.
.
.
.
.
.
.
.
.
.
Creating a Security Key.  
Changing a Security Key .  
Saving a Security Key.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Unlocking Secure Physical Disks .  
Erasing Secure Physical Disks .  
Configuring Hot Spare Physical Disks.  
Hot Spares and Rebuild.  
Global Hot Spares.  
.
.
.
Hot Spare Drive Protection .  
Enclosure Loss Protection .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
121  
Host-to-Virtual Disk Mapping .  
.
.
.
.
.
.
.
.
.
124  
Creating Host-to-Virtual Disk Mappings .  
124  
Modifying and Removing  
Host-to-Virtual Disk Mapping.  
Changing Controller Ownership  
of the Virtual Disk .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
126  
127  
.
.
.
.
.
.
.
8
Contents  
Removing Host-to-Virtual Disk Mapping .  
.
.
.
.
.
128  
Ownership of a Disk Group .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
128  
Restricted Mappings.  
Changing the RAID Controller Module  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Changing the RAID Level of a Disk Group.  
Storage Partitioning .  
.
.
Disk Group and Virtual Disk Expansion .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
137  
Disk Group Expansion  
.
.
.
.
.
.
.
.
.
.
Using Free Capacity  
Using Unconfigured Capacity  
Disk Group Migration .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
137  
138  
Export Disk Group.  
Exporting a Disk Group .  
Import Disk Group .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
138  
138  
Storage Array Media Scan  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
139  
10 Configuration: Premium Feature—  
Snapshot Virtual Disks  
.
.
.
.
.
.
.
.
.
.
.
.
.
143  
Creating a Snapshot Virtual Disk  
Using the Simple Path .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
144  
About the Simple Path  
.
.
.
145  
Contents  
9
Preparing Host Servers to Create the  
Creating a Snapshot Virtual Disk  
Using the Advanced Path  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
148  
Preparing Host Servers to Create the  
Snapshot Using the Advanced Path  
.
.
.
.
.
.
.
149  
Creating the Snapshot  
Using the Advanced Path.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
151  
153  
157  
158  
Specifying Snapshot Virtual Disk Names .  
Snapshot Repository Capacity.  
Re-creating Snapshot Virtual Disks .  
.
.
.
.
.
.
.
.
.
Preparing Host Servers to Re-create a  
Snapshot Virtual Disk .  
.
.
.
.
.
.
Re-creating a Snapshot Virtual Disk  
.
.
.
160  
11 Configuration: Premium Feature—  
Virtual Disk Copy  
.
.
.
.
.
Creating a Virtual Disk Copy for  
an MSCS Shared Disk .  
.
.
.
.
.
.
.
.
.
.
.
164  
Virtual Disk Read/Write Permissions .  
Virtual Disk Copy Restrictions .  
.
.
.
.
Creating a Virtual Disk Copy.  
Before You Begin .  
Virtual Disk Copy and  
Modification Operations  
Create Copy Wizard.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
165  
.
.
.
165  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
166  
166  
.
.
10  
Contents  
Failed Virtual Disk Copy.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
166  
Failed RAID Controller Module .  
Copy Manager .  
.
.
.
.
.
Copying the Virtual Disk .  
During Virtual Disk Copy  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
169  
169  
170  
Setting Copy Priority.  
.
.
.
Stopping a Virtual Disk Copy  
Recopying a Virtual Disk  
.
.
.
.
.
.
.
.
.
.
.
.
171  
172  
Recopying the Virtual Disk .  
Removing Copy Pairs  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
173  
12 Configuration: Premium Feature—  
Performance Tier  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
175  
177  
Remote Replication  
.
.
.
.
.
.
.
.
.
.
.
.
Switchable Host Access Configuration  
Required With Remote Replication.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
177  
178  
178  
Activating Remote Replication .  
.
.
.
Replication Repository Virtual Disks .  
RAID Levels for Replication  
Repository Virtual Disks  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
179  
Contents  
11  
Primary and Secondary Virtual Disk Pairs .  
.
.
.
.
.
.
.
.
.
179  
180  
With Other Features  
Storage Partitioning With  
Remote Replication .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Snapshot Virtual Disk With  
Remote Replication .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Switch Cascading.  
.
.
.
Setting Up and Managing  
Feature and Creating  
Replication Virtual Disks  
.
.
.
.
.
.
.
Repository Virtual Disks From  
.
.
.
.
.
.
.
.
.
Activating and Creating Replication  
Repository Virtual Disks From an  
Existing Disk Group .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Setting Synchronization Priority and  
Synchronization Method .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
188  
189  
Completing the Remote Replication  
RAID Controller Module  
Ownership/Preferred Path .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
190  
190  
Viewing Information About a Remote  
Replication or Replication  
Repository Virtual Disk .  
.
.
.
.
.
.
.
12  
Contents  
Viewing the Properties Pane .  
Viewing Logical Elements of the  
Secondary Virtual Disk .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
192  
.
.
.
.
.
.
.
.
.
.
Virtual Disk Status Icons .  
.
.
.
.
.
.
.
.
.
.
Changing Write Mode and Consistency  
Group Membership .  
.
.
.
.
.
.
.
Resynchronizing Virtual Disks  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Unsynchronized Virtual Disks  
Automatically Resynchronizing Virtual Disks  
Manually Resynchronizing Virtual Disks .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Suspending a Remote Replication .  
Resuming a Remote Replication .  
.
Testing Communication Between the  
.
.
.
.
.
.
.
Deleting a Virtual Disk From a Replicated  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
205  
Deleting a Primary Virtual Disk .  
.
.
Deleting a Primary Virtual Disk in a  
Replicated Pair From a Storage Array .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
206  
207  
207  
208  
Deleting a Secondary Virtual Disk in a  
Replicated Pair From a Storage Array .  
Removing a Remote Replication From a  
Storage Array.  
Deactivating the Remote Replication  
on the Storage Array . .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Contents  
13  
Disabling the Remote Replication  
Premium Feature .  
.
.
.
14 Configuration: Device Mapper  
Multipath for Linux  
.
.
.
Overview .  
.
.
.
.
.
.
.
.
.
.
Using DM Multipathing Devices.  
Prerequisites  
.
.
.
.
.
.
.
.
.
.
.
212  
213  
Limitations and Known Issues.  
Troubleshooting  
.
.
.
.
.
.
.
15 Management: Installing  
Array Components  
Recommended Tools .  
Front Bezel (Optional)  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Removing the Front Bezel.  
Installing the Front Bezel .  
Hard Drives.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
224  
226  
Installing a Drive Blank .  
Removing a Hard Drive .  
Installing a Hard Drive  
Removing a Hard Drive  
.
From a Hard-Drive Carrier  
.
.
.
.
.
.
.
.
.
.
.
.
227  
Installing a Hard Drive  
Into a Drive Carrier .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
229  
RAID Controller Module .  
229  
14  
Contents  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
231  
233  
Installing a RAID Controller Module Blank.  
Removing a RAID Controller Module .  
Opening the RAID Controller Module  
.
.
.
.
.
.
Removing the RAID Controller  
Installing the RAID Controller  
Module Backup Battery Unit .  
Power Supply/Cooling Fan Module .  
Removing a Power  
Installing a Power  
Control Panel.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
239  
241  
Installing the Control Panel.  
.
.
.
.
.
.
.
.
.
.
.
.
.
242  
Removing the Backplane .  
.
.
.
.
.
.
Downloading RAID Controller  
and NVSRAM Packages .  
.
.
.
.
.
.
.
.
.
.
.
.
.
247  
Downloading Both RAID Controller  
and NVSRAM Firmware .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
248  
250  
Downloading Only NVSRAM Firmware .  
Contents  
15  
Downloading Physical Disk Firmware  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
253  
Downloading MD1200 Series Expansion  
Module EMM Firmware .  
.
.
.
.
.
.
.
.
Self-Monitoring Analysis and Reporting  
Technology (SMART).  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
256  
Media Errors and Unreadable Sectors  
.
.
.
.
Viewing the Firmware Inventory  
.
.
.
.
.
.
.
.
259  
.
.
.
.
.
Microsoft Services.  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
261  
261  
Virtual Disk Service .  
19 Troubleshooting: Your Storage  
Array Software  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
263  
263  
266  
267  
268  
269  
270  
Start–Up Routine .  
.
.
.
.
.
Device Health Conditions .  
Storage Array Support Data .  
Automatically Collect the Support Bundle Data  
Collecting the Physical Disk Data .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Event Log .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Recovery Guru .  
16  
Contents  
Storage Array Profile  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
270  
Viewing the Physical Associations .  
Finding Nodes .  
Using Go To  
.
.
.
.
.
.
.
.
.
.
.
.
Storage Array Condition .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
276  
279  
279  
281  
Locating a Physical Disk.  
.
.
.
.
.
Locating an Expansion Enclosure .  
.
.
SMrepassist Utility  
.
.
.
.
.
.
.
.
.
.
.
.
.
Unidentified Devices  
Recovering From an Unidentified Storage Array .  
Context Agent Software .  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
285  
20 Troubleshooting: Your Array  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
289  
289  
289  
289  
292  
Safety First—For You and Your Array .  
.
.
.
.
.
Troubleshooting Storage Array Startup Failure .  
Troubleshooting Loss of Communication .  
.
.
.
.
.
.
.
.
.
Troubleshooting an SFP Transceiver  
.
.
.
.
Troubleshooting External Connections .  
Contents  
17  
Troubleshooting Power  
Supply/Cooling Fan Module .  
Troubleshooting Expansion  
Enclosure Management Modules .  
Troubleshooting Array and  
.
.
Expansion Enclosure Connections  
.
.
.
.
.
.
.
.
297  
Troubleshooting a Wet Storage Array .  
Troubleshooting a Damaged Array  
Troubleshooting RAID Controller Modules  
.
.
.
.
.
.
.
.
.
.
.
.
300  
Conditions.  
.
.
.
.
.
.
.
300  
21 Getting Help  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
303  
Contacting Dell.  
303  
Index  
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
305  
18  
Contents  
1
Introduction  
NOTE: Throughout the document, Dell PowerVault MD3600f series storage array  
refers to both Dell PowerVault MD3600f and Dell PowerVault MD3620f.  
See the Safety, Environmental, and Regulatory Information document for  
important safety information before following any procedures listed in  
this document.  
About This Document  
This guide familiarizes you with the functions of the MD3600f Series storage  
array. The guide is organized according to the tasks that you must complete  
after receiving your Dell PowerVault MD3600f Series storage array.  
Planning—Information about the storage array and its features.  
Configuration—Tasks that must be completed to ensure that your storage  
array performs optimally.  
Management—Tasks the may be performed to ensure the storage array  
components are up to date and performing properly, including removal and  
installation of storage array components.  
Troubleshooting—Tasks that you must complete to resolve problems that  
may occur with the storage array.  
Additional information on these and other topics can be found in the Dell  
PowerVault MD3600f and MD3620f Storage Array Deployment Guide,  
available at support.dell.com/manuals.  
Introduction  
19  
     
Inside the Box of the Dell PowerVault MD3600f  
Series Storage Array  
Unless specified, MD3600f Series represents Dell PowerVault MD3600f and  
Dell PowerVault MD3620f.  
Your MD3600f Series product package includes:  
Dell PowerVault MD3600f Series storage array  
Small form-factor pluggable (SFP) transceiver (2 per controller)  
Fiber optic cables (optional)  
Power cables (2)  
Bezel (optional)  
Mounting rails (2) (optional)  
Dell PowerVault MD3600f Series resource media  
The Rack Installation Instructions included with your rack solution  
describes how to install your enclosure into a rack  
The Getting Started With Your System provides an overview of enclosure  
features, setting up your enclosure, and technical specifications  
Dell PowerVault MD3600f Series Storage Array  
The Dell PowerVault MD3600f Series is a 2U rack-mounted external  
redundant array of independent disks (RAID) storage array capable of  
accommodating up to 12 3.5" or 24 2.5" 6.0-Gbps Serial-Attached SCSI (SAS)  
disks. The MD3600f Series storage arrays can be daisy-chained with MD1200  
Series expansion enclosures, providing access to a maximum of 96 disks in the  
entire storage array. Connectivity between the storage array and the host  
server is provided by a Fibre Channel (FC) host bus adapter (HBA).  
Dell PowerVault Modular Disk Storage Manager  
Dell PowerVault Modular Disk Storage Manager (MDSM) is a graphical user  
interface (GUI) application, used to configure and manage one or more  
MD3600f Series Storage Arrays. The MDSM software is located on the  
MD3600f Series resource media.  
20  
Introduction  
         
Other Information You May Need  
WARNING: See the safety and regulatory information that shipped with your  
system. Warranty information may be included within this document or as a  
separate document.  
Getting Started Guide provides an overview of setting up and cabling your  
storage array.  
Deployment Guide provides installation and configuration instructions for  
both software and hardware.  
Storage Manager CLI Guide provides information about using the  
command line interface (CLI).  
Resource media contains all system management tools.  
Configuring Fibre Channel with the Dell MD3600f Series Storage Array  
provides information regarding configurations, host bus adapter (HBA)  
installation, and zoning. This document is available at  
support.dell.com/manuals.  
Systems Support Matrix provides information on supported software and  
hardware for MD systems. The document is available at  
support.dell.com/manuals  
.
Dell PowerEdge Cluster Documentation is available at dell.com/ha  
.
Release notes or readme files are included to provide last-minute updates  
to the enclosure or documentation or advanced technical reference  
material intended for experienced users or technicians.  
This document as well as Dell PowerVault MD 1200 series installation  
guide is available at support.dell.com/manuals for users who incorporate  
MD1200 expansion enclosures.  
NOTE: Always check for updates on support.dell.com/manuals and read the  
updates first because they often supersede information in other documents.  
Introduction  
21  
   
22  
Introduction  
2
Planning: About Your Storage Array  
Overview  
The MD3600f Series storage array is designed for high availability, offering  
redundant access to data storage. It’s features support for both single and dual  
RAID controller configuration.  
The Dell PowerVault MD3600f Series storage array provides Fibre Channel  
(FC) connectivity to the host server and enables access for up to eight non-  
redundant servers or four redundant servers.  
The MD3600f Series storage array includes a number of components. These are:  
RAID controller module(s)  
PSU/fan modules  
Disk drives (also called physical disks/hard drives in this document)  
An optional front panel bezel  
A system enclosure, into which, the other components are plugged  
Planning: About Your Storage Array  
23  
     
Hardware Features  
Front-Panel Features and Indicators  
Figure 2-1. Front-Panel Features and Indicators—Dell PowerVault MD3600f  
1
2
3
4
5
6
Figure 2-2. Front-Panel Features and Indicators—Dell PowerVault MD3620f  
1
2
3
4
5
6
24  
Planning: About Your Storage Array  
       
Figure 2-3. Front-Bezel Features and Indicators  
1
2
3
Item Indicator, Button, or  
Icon Description  
Enclosure status LED The enclosure status LED lights when the  
Connector  
1
enclosure power is on.  
Lights blue during normal operation.  
Blinks blue when a host server is identifying the  
enclosure or when the system identification  
button is pressed.  
Lights amber as enclosure boots or is reset.  
Blinks amber when the enclosure is either in a  
fault state or the hosts are not using the preferred  
path to a virtual disk.  
2
Power LED  
The power LED lights green when at least one  
power supply is supplying power to  
the enclosure.  
Planning: About Your Storage Array  
25  
Item Indicator, Button, or  
Connector  
Icon Description  
This LED should be unlit as the split mode  
3
Split mode LED  
function is not supported by the MD3600f Series  
Storage Arrays.  
4
System identification  
button  
The system identification button on the front  
control panel can be used to locate a particular  
enclosure within a rack. When the button is  
pushed, the system status indicators on the  
control panel and the RAID controller module(s)  
blink blue until the button is pushed again.  
5
6
Hard drives  
PowerVault MD3600f—Up to 12 3.5" SAS hot-  
swappable hard drives.  
PowerVault MD3620f—Up to 24 2.5" SAS hot-  
swappable hard drives.  
Enclosure mode  
switch  
The function of this switch is not applicable to  
your storage array. However, if MD1200 Series  
expansion enclosures are daisy chained to the  
storage array, the enclosure mode switches of the  
MD1200 Series expansion enclosures must be set  
to the Unified-Mode position.  
NOTE: This switch must be set prior to turning on  
the MD1200 Series expansion enclosure. Changing  
the switch setting after the expansion enclosure is  
turned on has no effect on enclosure configuration  
until the expansion enclosure goes through a  
complete power cycle.  
26  
Planning: About Your Storage Array  
Back Panel Features and Indicators  
Figure 2-4. Back-Panel Features and Indicators—Dell PowerVault MD3600f Series  
1
1
3
4
2
1
3
600W Power supply/cooling fan  
module  
2
4
RAID Controller Module 0  
RAID Controller Module 1  
600W Power supply/cooling fan  
module  
Planning: About Your Storage Array  
27  
     
Hard–Drive Indicator Patterns  
Figure 2-5. Hard Drive Indicators  
1
2
1
hard-drive activity indicator (green)  
2
hard-drive status indicator (green  
and amber)  
28  
Planning: About Your Storage Array  
     
Drive-Status Indicator Pattern  
Condition  
Off  
The physical disk is:  
• not yet discovered by the host server  
• has been spun down for removal  
• not supported for the RAID controller  
module or is not in the physical disk slot  
NOTE: The drive status indicator remains  
off until all hard drives are initialized after  
system power is turned on. Drives are not  
ready for insertion or removal during this  
time.  
Steady green  
Physical disk is online  
Green flashing (On 250 ms, Off 250 ms) Physical disk is being identified  
Green flashing (On 400 ms, Off 100 ms) Physical disk rebuilding  
Amber flashing (On 150 ms, Off 150 ms) Physical disk failed  
Flashing green, amber, and off (Green On Physical disk failure predicted (SMART)  
500 ms, Amber On 500 ms, Off 1000 ms)  
Flashing green, amber, and off (Green 3 Physical disk rebuild aborted  
seconds, amber 3 seconds, and off 3  
seconds)  
The MD3600f Series storage array includes two integrated, hot-swappable  
power supply/cooling fan modules. Both modules must be installed to ensure  
proper cooling. Each module contains two separate cooling fans. The system  
requires at least one of the cooling fans to function to avoid overheating.  
A power supply/cooling fan module can be replaced without powering down  
the system. For information on removing and replacing the modules, see  
"Power Supply/Cooling Fan Module" on page 237.  
CAUTION: A power supply/cooling fan module can be removed from a powered-  
on enclosure for a maximum period of 5 minutes. Beyond that time, the system may  
automatically shut down to prevent damage.  
Planning: About Your Storage Array  
29  
     
Power Indicator Codes  
Figure 2-6. Power Indicator Codes  
1
2
3
4
5
Item LED Type  
Icon Description  
1
DC power  
The LED lights green when the DC output  
voltage is within the limit.  
If this LED is off, it indicates that the DC output  
voltage are not within the limit.  
2
Power supply/cooling  
fan fault  
The LED lights amber when the DC output  
voltage is not within the limit or a fault with the  
fan is detected.  
If this LED is off, it indicates that no fault  
condition is present.  
3
AC power  
The LED lights green when the AC input voltage  
is within the limit.  
If this LED is off, it indicates either there is no  
power or the AC input voltage is not within  
the limit.  
4
5
Power connector  
Connect the external power supply to  
this connector.  
Power switches (2)  
The power switch controls the power supply  
output to the enclosure.  
30  
Planning: About Your Storage Array  
     
3
Planning: RAID Controller Modules  
RAID Controller Modules  
The RAID controller modules provide high-performance, advanced virtual  
disk configuration, and fault-tolerant disk subsystem management. Each  
RAID controller module contains 2 GB of cache that is mirrored with the  
other controller's cache for high availability and is protected by a battery-  
powered cache offload mechanism.  
RAID controller modules provide the following data path and enclosure  
management functions:  
Monitoring and controlling enclosure environment elements  
(temperature, fans, power supplies, and enclosure LEDs)  
Controlling access to the physical disks  
Communicating enclosure attributes and states to the host server and  
management station  
Each RAID controller module has multiple Fibre Channel (FC) ports  
(labeled FCIN) for host access. The ports provide redundant host  
connections and support a high availability storage environment. Each port  
must have a small form-factor pluggable (SFP) transceiver module installed  
for connectivity. Various configurations can be utilized, in both single  
controller (simplex) and dual controller (duplex) modes, to connect the  
storage enclosure to hosts depending on specific redundancy needs.  
For information on cabling, see the MD3600f and MD3620f Series Storage  
Array’s Deployment Guide, at support.dell.com/manuals.  
Planning: RAID Controller Modules  
31  
       
RAID Controller Module Connectors and Features  
Figure 3-1. MD3600f Series Fibre Channel RAID Controller Module  
3
2
4
7
5
6
1
9
10  
8
12  
21  
20  
19 18 17 16 15 14 13  
11  
Item Component  
Function  
1
SAS OUT port  
Provides SAS connection for cabling to a downchain  
expansion enclosure.  
2
3
4
5
6
FCIN port 0  
FCIN port 1  
FCIN port 2  
FCIN port 3  
Provides host-to-controller FC connection.  
Provides host-to-controller FC connection.  
Provides host-to-controller FC connection.  
Provides host-to-controller FC connection.  
Provides MAC addresses of the management port.  
MAC Address  
label  
7
8
Debug port  
Dell support only.  
Management  
port activity  
LED  
Lights green when Ethernet connection is active.  
Off when Ethernet connection is not active.  
32  
Planning: RAID Controller Modules  
     
Item Component  
Function  
9
Management  
port Ethernet  
connector  
Provides a 100/1000 Mbps Ethernet connection for out-of-  
band management of the enclosure.  
10  
Management  
port speed LED 1000 Mbps.  
Lights green when Ethernet connection is operating at  
Lights amber when Ethernet connection is operating at  
100 Mbps.  
Off when Ethernet connection is operating at 10 Mbps or is  
not active.  
11  
12  
Password reset  
switch  
Activating this switch resets the password.  
Battery fault  
Lights amber when battery backup unit or battery  
has failed.  
Off when battery backup unit is operating normally.  
13  
14  
FCIN port 3  
Displays link rate and port status per the accompanying  
LED 1 / LED 0 host channel LED link/rate table.  
Cache active or Lights green when on-board controller memory  
cache offload  
LED  
contains data.  
If AC power fails, this LED changes to indicate Cache  
Offload status.  
If the password reset function has successfully changed the  
password, this LED flashes on and off briefly.  
15  
16  
FCIN port 2  
Displays link rate and port status per the accompanying  
LED 1 / LED 0 host channel LED link/rate table.  
System  
identification  
LED  
Blinks blue when system identification switch push-button  
on enclosure front panel is pressed.  
17  
Controller fault Lights amber when controller fault detected.  
LED  
Off when controller operating normally.  
18  
19  
FCIN port 1  
Displays link rate and port status per the accompanying  
LED 1 / LED 0 host channel LED link/rate table.  
Controller power Lights green when controller power is on.  
LED  
Off when controller is not powered.  
Planning: RAID Controller Modules  
33  
Item Component  
Function  
20  
FCIN port 0  
Displays link rate and port status per the accompanying  
LED 1 / LED 0 host channel LED link/rate table.  
21  
SAS OUT port  
link/fault LED  
Lights green when all four links are connected.  
Lights amber when one to three links are disconnected.  
Off when all links in the port are disconnected or cable is  
disconnected.  
Host Channel LED Link/Rate Indications  
Below each FC port is a pair of LED indicators. The status of any FC port can  
be determined by applying Table 3-1 to the condition of each LED pair.  
Table 3-1. LED Status Indicators  
LED 1  
Off  
Off  
On  
LED 0  
Off  
On  
CommRate  
Link down  
2 Gbps  
Off  
On  
4 Gbps  
On  
RAID Controller Module—Additional Features  
Battery Backup Unit  
Each RAID controller contains a two-cell lithium ion nanopolymer battery  
backup unit (BBU). It provides power to the RAID controller module in the  
event of a power outage, see "RAID Controller Module Backup Battery Unit"  
on page 235.  
NOTE: For virtual disks, the RAID controller firmware changes the data cache  
setting based on the state of the battery. If the battery is missing or does not have  
sufficient charge, the controller flushes the cache and sets the write cache  
attribute to Write Through for all virtual disks. When the battery is replaced, Write  
Back is re-enabled.  
34  
Planning: RAID Controller Modules  
         
Storage Array Thermal Shutdown  
The system automatically shuts down when system temperature exceeds the  
safe threshold. The battery backup unit protects against data loss by  
providing power to offload cache to non-volatile memory in the event of  
power loss. It is not necessary to shut down any MD1200 Series expansion  
enclosures attached to the storage array when thermal shutdown occurs.  
Temperature threshold values determine the temperature at which shutdown  
occurs. These thresholds cannot be changed.  
Table 3-2. Shutdown Threshold Type  
Threshold Temperature Exceeding Event Description  
Nominal failure threshold  
Maximum failure threshold  
A critical event is set.  
Shutdown of the system power supplies occurs  
within three minutes.  
Shutdown threshold  
Shutdown of the system power supplies occurs  
within five seconds.  
System Password Reset  
The storage array password can be reset if it is forgotten. To reset the password,  
push and hold down the password reset switch for at least 5 seconds. The  
password is deleted. See Figure 3-1 to locate the password reset switch.  
The RAID controller module allows you to change the password. For more  
about setting your password, see "Setting a Password" on page 73.  
NOTE: The reset switch can be accessed by using a small object such as the tip of  
a pen.  
Cache Functions and Features  
Cache Mirroring  
Cache mirroring function copies accepted host-write data from the primary  
controller to the partner controller. This action ensures that host-write data is  
safely mirrored to the partner controller before successful completion status is  
returned to the host. If a controller fails, the surviving controller safely retains  
all mirrored data. Cache mirroring is enabled by default.  
Planning: RAID Controller Modules  
35  
       
Write–Back Cache  
In write-back cache, write operations result in a completion signal being sent  
to the host operating system as soon as the cache receives the data to be  
written. The target physical disk receives the data at a more appropriate time  
in order to increase controller performance. In dual-active controller  
configurations with write-back caching enabled, the write data is always  
mirrored to the cache of the second controller before completion status is  
issued to the host initiator. Write-back cache is enabled by default unless  
cache mirroring is disabled.  
Write–Through Cache  
In write-through cache, data is written to the physical disk before completion  
status is returned to the host operating system. Write-through cache is  
considered more robust than write-back cache, since a power failure is less  
likely to cause loss of data. The RAID controller automatically switches to  
write-through if cache mirroring is disabled or if the battery is missing or has a  
fault condition.  
SFP Transceivers, Fiber Optic, and SAS Cables  
Figure 3-2 shows fiber optic cable and SFP transceivers.  
NOTE: Your SFP transceivers and cables may look different from the ones shown  
below. The differences do not affect the performance of the SFP transceivers.  
FC host connections may operate at 8 Gbps or at a lower data rate. Ports for 8  
Gbps FC host connections require SFP transceivers designed for this data rate.  
SFP transceivers that support other data rates are incompatible.  
WARNING: Do not disassemble or remove any part of a small form-factor  
pluggable (SFP) transceiver due to the possibility of being exposed to laser  
radiation.  
36  
Planning: RAID Controller Modules  
     
Figure 3-2. Fiber Optic Cable Connection  
1
2
1
SFP transceiver  
2
Fiber optic cable  
The RAID expansion ports support SAS expansion connections. Figure 3-3  
shows a SAS cable and an SFF-8088 connector.  
Figure 3-3. SAS Cable  
1
2
1
SF-8088 connector  
2
SAS cable  
Planning: RAID Controller Modules  
37  
   
Interoperability of 2 Gbps, 4 Gbps, and 8 Gbps  
Devices  
The FC standard specifies a procedure for speedy auto-detection. If a 4 Gbps  
port on a switch or device is connected to a 2 Gbps port, it must negotiate  
down for the link to run at 2 Gbps. If there are two 8 Gbps ports on either end  
of a link, the negotiation runs the link at 8 Gbps if the link supports the  
required specifications. If a link connects at 4 Gbps in this scenario, possible  
causes could be excessive cable length, poor cable quality, loose cables or  
SFPs, and so on. Ensure that the cable lengths are not excessive and the  
quality of the cable and connector is good.  
The same rules apply to 8 Gbps devices relative to 4 Gbps and 2 Gbps  
environments. The 8 Gbps devices have the ability to automatically negotiate  
down to either 4 Gbps or 2 Gbps, depending on the attached device and link  
quality. If the link negotiates down to a speed slower than expected, check  
the device.  
38  
Planning: RAID Controller Modules  
 
4
Planning: MD3600f Series Storage  
Array Terms and Concepts  
This chapter explains terms and concepts used for configuration and  
operation of MD3600f Series storage arrays.  
Physical Disks, Virtual Disks, and Disk Groups  
Physical disks in your storage array provide the physical storage capacity for  
your data. Before you can begin writing data to the storage array, you must  
configure the physical storage capacity into logical components, called disk  
groups and virtual disks.  
A disk group is a set of physical disks upon which multiple virtual disks are  
created. The maximum number of physical disks supported in a disk group is  
96 drives for RAID 0, RAID 1, and RAID10, and 30 drives for RAID 5 and  
RAID 6. You can create disk groups from unconfigured capacity on your  
storage array.  
A virtual disk is a partition in a disk group that is made up of contiguous data  
segments of the physical disks in the disk group. A virtual disk consists of data  
segments from all physical disks in the disk group. Virtual disks and disk  
groups are set up according to how you plan to organize your data. For  
example, you might have one virtual disk for inventory, a second virtual disk  
for financial and tax information, and so on.  
All virtual disks in a disk group support the same RAID level. The storage  
array supports up to 255 virtual disks (minimum size of 10 MB each) that can  
be assigned to host servers. Each virtual disk is assigned a Logical Unit  
Number (LUN) that is recognized by the host operating system.  
Planning: MD3600f Series Storage Array Terms and Concepts  
39  
     
Physical Disks  
Only Dell supported 6.0 Gbps SAS physical disks are supported in the storage  
array. If the storage array detects unsupported physical disks, it marks the disk  
as unsupported and the physical disk becomes unavailable for all operations.  
NOTE: The MD3600f Series array must contain at least two physical disks for  
proper operation. This is necessary because the physical disks are used to store  
configuration information.  
Physical Disk States  
The storage array recognizes the physical disk states. Mode and status of  
physical disks are reported in the MDSM application. Possible values for  
mode and status are described in Table 4-1.  
Table 4-1. RAID Controller Physical Disk States  
Status  
Mode  
Description  
Physical Disk  
Status LED  
Optimal Assigned  
The physical disk in the indicated slot Steady Green  
is configured as part of a disk group.  
Optimal Unassigned  
The physical disk in the indicated slot Steady Green  
is unused and available to be  
configured.  
Optimal Hot Spare  
standby  
The physical disk in the indicated slot Steady Green  
is configured as a hot spare.  
Optimal Hot Spare in The physical disk in the indicated slot is Steady Green  
use  
in use as a hot spare within a disk group.  
Failed  
Assigned,  
Unassigned,  
The physical disk in the indicated slot Amber flashing  
has failed because of an unrecoverable (150 ms)  
Hot Spare in error, an incorrect drive type or drive  
use, or Hot size, or by its operational state being  
Spare standby set to failed.  
Replaced Assigned  
The physical disk in the indicated slot Green flashing  
has been replaced and is ready to be, (On 400 ms, Off  
or is actively being, configured into a 100 ms)  
disk group.  
40  
Planning: MD3600f Series Storage Array Terms and Concepts  
         
Table 4-1. RAID Controller Physical Disk States (continued)  
Status  
Mode  
Description  
Physical Disk  
Status LED  
Pending Assigned,  
A Self-Monitoring Analysis and  
Reporting Technology (SMART) error (500 ms), Amber  
Flashing Green  
Failure  
Offline  
Unassigned,  
Hot Spare in has been detected on the physical disk (500 ms), and  
use, or Hot  
Spare standby  
in the indicated slot.  
off (1000ms)  
Not applicable The physical disk has either been spun Flashing Green  
down or had a rebuild aborted by  
user request.  
(3000 ms), Amber  
(3000 ms), and  
off(3000ms)  
Identify Assigned,  
Unassigned,  
The physical disk is being identified. Green flashing  
(250 ms)  
Hot Spare in  
use, or Hot  
Spare standby  
If a disk drive rebuild fails because of a source drive failure or because the  
drive is too small, the MDSM reports a failure of the physical disk even  
though the LED state on the drive indicates the rebuild was aborted (green  
for 3 seconds, amber for 3 seconds, then off for 3 seconds).  
Self-Monitoring Analysis and Reporting Technology  
SMART monitors the internal performance of all physical disk components to  
detect faults indicating the potential for physical disk failure. SMART uses  
this information to report whether failure is imminent so that a physical disk  
can be replaced before failure occurs. The storage array monitors all attached  
drives and notifies you when a predicted failure is reported by a physical disk.  
Virtual Disks and Disk Groups  
1
2
3
4
Organize the physical disks into disk groups.  
Create virtual disks within these disk groups.  
Provide host server access.  
Create mappings to associate the virtual disks with the host servers.  
NOTE: Host server access must be created before mapping virtual disks.  
Planning: MD3600f Series Storage Array Terms and Concepts  
41  
       
Disk groups are always created in the unconfigured capacity of a storage array.  
Unconfigured capacity is the available physical disk space not already  
assigned in the storage array.  
Virtual disks are created within the free capacity of a disk group. Free capacity  
is the space in a disk group that has not been assigned to a virtual disk.  
Virtual Disk States  
The storage array recognizes the following virtual disk states.  
Table 4-2. RAID Controller Virtual Disk States  
State  
Description  
Optimal  
Degraded  
The virtual disk contains physical disks that are all online.  
The virtual disk with a redundant RAID level contains an inaccessible  
physical disk. The system can still work properly, but performance may  
be affected and additional disk failures may result in data loss.  
Offline  
A virtual disk with one or more member disks in an inaccessible (failed,  
missing, or offline) state. Data on the virtual disk is no longer accessible.  
Force online The storage array forces a virtual disk that is in an Offline state to an  
Optimal state. If all the member physical disks are not available, the  
storage array forces the virtual disk to a Degraded state. The storage  
array can force a virtual disk to an Online state only when a sufficient  
number of physical disks are available to support the virtual disk.  
42  
Planning: MD3600f Series Storage Array Terms and Concepts  
   
RAID Levels  
RAID levels determine the way in which data is written to physical disks.  
Different RAID levels provide different levels of accessibility, redundancy,  
and capacity.  
Using multiple physical disks has the following advantages over using a single  
physical disk:  
Placing data on multiple physical disks (striping) allows input/output (I/O)  
operations to occur simultaneously and improve performance.  
Storing redundant data on multiple physical disks using mirroring or parity  
supports reconstruction of lost data if an error occurs, even if that error is  
the failure of a physical disk.  
Each RAID level provides different performance and protection. You should  
select a RAID level based on the type of application, access, fault tolerance,  
and data you are storing.  
The storage array supports RAID levels 0, 1, 5, 6, and 10. The maximum  
number of physical disks that can be used in a disk group depends on the  
RAID level:  
96 for RAID 0, 1 and 10  
30 for RAID 5 and 6.  
RAID Level Usage  
To ensure best performance, you should select an optimal RAID level when  
you create a system physical disk. The optimal RAID level for your disk array  
depends on:  
Number of physical disks in the disk array  
Capacity of the physical disks in the disk array  
Need for redundant access to the data (fault tolerance)  
Disk performance requirements  
Planning: MD3600f Series Storage Array Terms and Concepts  
43  
         
RAID 0  
RAID 0 uses disk striping to provide high data throughput, especially for large  
files in an environment that requires no data redundancy. RAID 0 breaks the  
data down into segments and writes each segment to a separate physical disk.  
I/O performance is greatly improved by spreading the I/O load across many  
physical disks. Although it offers the best performance of any RAID level,  
RAID 0 lacks data redundancy. Choose this option only for non-critical data,  
because failure of one physical disk results in the loss of all data. Examples of  
RAID 0 applications include video editing, image editing, prepress  
applications, or any application requiring high bandwidth.  
RAID 1  
RAID 1 uses disk mirroring so that data written to one physical disk is  
simultaneously written to another physical disk. This RAID level offers fast  
performance and the best data availability, but also the highest disk overhead.  
RAID 1 is recommended for small databases or other applications that do not  
require large capacity for example accounting, payroll, or financial  
applications. RAID 1 provides full data redundancy.  
RAID 5  
RAID 5 uses parity and striping data across all physical disks (distributed  
parity) to provide high data throughput and data redundancy, especially for  
small random access. RAID 5 is a versatile RAID level and is suited for multi-  
user environments where typical I/O size is small and there is a high  
proportion of read activity such as file, application, database, web, e-mail,  
news, and intranet servers.  
RAID 6  
RAID 6 is similar to RAID 5 but provides an additional parity disk for better  
redundancy. RAID 6 is the most versatile RAID level and is suited for multi-  
user environments where typical I/O size is small and there is a high  
proportion of read activity. RAID 6 is recommended when large size physical  
disks are used or large number of physical disks are used in a disk group.  
44  
Planning: MD3600f Series Storage Array Terms and Concepts  
     
RAID 10  
RAID 10, a combination of RAID 1 and RAID 0, uses disk striping across  
mirrored disks. It provides high data throughput and complete data  
redundancy. Utilizing an even number of physical disks (four or more) creates  
a RAID level 10 disk group and/or virtual disk. Because RAID levels 1 and 10  
use disk mirroring, half of the capacity of the physical disks is utilized for  
mirroring. This leaves the remaining half of the physical disk capacity for  
actual storage. RAID 10 is automatically used when a RAID level of 1 is  
chosen with four or more physical disks. RAID 10 works well for medium-  
sized databases or any environment that requires high performance and fault  
tolerance and moderate-to-medium capacity.  
Segment Size  
Disk striping enables data to be written across multiple physical disks. Disk  
striping enhances performance because striped disks are accessed simultaneously.  
The segment size or stripe element size specifies the size of data in a stripe  
written to a single disk. The MD3600f Series array supports stripe element sizes  
of 8, 16, 32, 64, 128, 256, and 512 KB. The default stripe element size is 128 KB.  
Stripe width, or depth, refers to the number of disks involved in an array  
where striping is implemented. For example, a four-disk group with disk  
striping has a stripe width of four.  
NOTE: Although disk striping delivers excellent performance, striping alone does  
not provide data redundancy.  
Virtual Disk Operations  
Virtual Disk Initialization  
Every virtual disk must be initialized. Initialization can be done in the  
foreground or the background. A maximum of four virtual disks can be  
initialized concurrently on each RAID controller module.  
Planning: MD3600f Series Storage Array Terms and Concepts  
45  
             
Background Initialization  
The storage array executes a background initialization when the virtual disk is  
created to establish parity, while allowing full host server access to the virtual  
disks. Background initialization does not run on RAID 0 virtual disks. The  
background initialization rate is controlled by MDSM. To change the rate of  
background initialization, you must stop any existing background  
initialization. The rate change is implemented when the background  
initialization restarts automatically.  
Foreground Initialization  
The storage array supports foreground initialization for virtual disks. All  
access to the virtual disk is blocked during foreground initialization. During  
foreground initialization, zeros (0x00) are written to every sector of the virtual  
disk. The virtual disk is available after foreground initialization is completed.  
Consistency Check  
A consistency check verifies the correctness of data in a redundant array  
(RAID levels 1, 5, 6, and 10). For example, in a system with parity, checking  
consistency involves computing the data on one physical disk and comparing  
the results to the contents of the parity physical disk.  
A consistency check is similar to a background initialization. The difference is  
that background initialization cannot be started or stopped manually, while  
consistency check can.  
NOTE: It is recommended that you run data consistency checks on a redundant  
array at least once a month. This allows detection and automatic replacement of  
unreadable sectors. Finding an unreadable sector during a rebuild of a failed  
physical disk is a serious problem, because the system does not have the  
redundancy to recover the data.  
Media Verification  
Another background task performed by the storage array is media verification  
of all configured physical disks in a disk group. The storage array uses the  
Read operation to perform verification on the space configured in virtual  
disks and the space reserved for the metadata.  
46  
Planning: MD3600f Series Storage Array Terms and Concepts  
               
Cycle Time  
The media verification operation runs only on selected disk groups,  
independent of other disk groups. Cycle time is the time taken to complete  
verification of the metadata region of the disk group and all virtual disks in  
the disk group for which media verification is configured. The next cycle for a  
disk group starts automatically when the current cycle completes. You can set  
the cycle time for a media verification operation between 1 and 30 days. The  
storage controller throttles the media verification I/O accesses to disks based  
on the cycle time.  
The storage array tracks the cycle for each disk group independent of other disk  
groups on the controller and creates a checkpoint. If the media verification  
operation on a disk group is preempted or blocked by another operation on the  
disk group, the storage array resumes after the current cycle. If the media  
verification process on a disk group is stopped due to a RAID controller module  
restart, the storage array resumes the process from the last checkpoint.  
Virtual Disk Operations Limit  
The maximum number of active, concurrent virtual disk processes per RAID  
controller module installed in the storage array is four. This limit is applied to  
the following virtual disk processes:  
Background initialization  
Foreground initialization  
Consistency check  
Rebuild  
Copy back  
If a redundant RAID controller module fails with existing virtual disk  
processes, the processes on the failed controller are transferred to the peer  
controller. A transferred process is placed in a suspended state if there are four  
active processes on the peer controller. The suspended processes are resumed  
on the peer controller when the number of active processes falls below four.  
Planning: MD3600f Series Storage Array Terms and Concepts  
47  
       
Disk Group Operations  
RAID Level Migration  
You can migrate from one RAID level to another depending on your  
requirements. For example, fault-tolerant characteristics can be added to a  
stripe set (RAID 0) by converting it to a RAID 5 set. MDSM provides  
information about RAID attributes to assist you in selecting the appropriate  
RAID level. You can perform a RAID level migration while the system is still  
running and without rebooting, which maintains data availability.  
Segment Size Migration  
Segment size refers to the amount of data (in kilobytes) that the storage array  
writes on a single physical disk in a virtual disk before writing data on the next  
physical disk. Valid values for the segment size are 8, 16, 32, 64, 128, 256, and  
512 KB.  
Dynamic segment size migration enables the segment size of a given virtual disk  
to be changed. A default segment size is set when the virtual disk is created, based  
on such factors as the RAID level and expected usage. You can change the default  
value (128 KB) if segment size usage does not match your needs.  
When considering a segment-size change, two scenarios illustrate different  
approaches to the limitations:  
If I/O activity stretches beyond the segment size, you can increase it to  
reduce the number of disks required for a single I/O. Using a single physical  
disk for a single request frees disks to service other requests, especially  
when you have multiple users accessing a database or storage environment.  
If you use the virtual disk in a single-user, large I/O environment (such as  
for multimedia application storage), performance can be optimized when  
a single I/O request is serviced with a single data stripe (the segment size  
multiplied by the number of physical disks in the disk group used for data  
storage). In this case, multiple disks are used for the same request, but  
each disk is only accessed once.  
48  
Planning: MD3600f Series Storage Array Terms and Concepts  
           
Virtual Disk Capacity Expansion  
When you configure a virtual disk, you select a capacity based on the amount  
of data you expect to store. However, you may need to increase the virtual disk  
capacity for a standard virtual disk by adding free capacity to the disk group.  
This creates more unused space for new virtual disks or to expand existing  
virtual disks.  
Disk Group Expansion  
Because the storage array supports hot pluggable physical disks, you can add  
two physical disks at a time for each disk group while the storage array  
remains online. Data remains accessible on virtual disk groups, virtual disks,  
and physical disks throughout the operation. The data and increased unused  
free space are dynamically redistributed across the disk group. RAID  
characteristics are also reapplied to the disk group as a whole.  
Disk Group Defragmentation  
Defragmenting consolidates the free capacity in the disk group into one  
contiguous area. Defragmentation does not change the way in which the data  
is stored on the virtual disks.  
Disk Group Operations Limit  
The maximum number of active, concurrent disk group processes per  
installed RAID controller module is one. This limit is applied to the following  
disk group processes:  
Virtual disk RAID level migration  
Segment size migration  
Virtual disk capacity expansion  
Disk group expansion  
Disk group defragmentation  
Planning: MD3600f Series Storage Array Terms and Concepts  
49  
               
If a redundant RAID controller module fails with an existing disk group process,  
the process on the failed controller is transferred to the peer controller. A  
transferred process is placed in a suspended state if there is an active disk group  
process on the peer controller. The suspended processes are resumed when the  
active process on the peer controller completes or is stopped.  
NOTE: If you try to start a disk group process on a controller that does not have an  
existing active process, the start attempt fails if the first virtual disk in the disk group  
is owned by the other controller and there is an active process on the other  
controller.  
RAID Background Operations Priority  
The storage array supports a common configurable priority for the following  
RAID operations:  
Background initialization  
Rebuild  
Copy back  
Virtual disk capacity expansion  
Raid level migration  
Segment size migration  
Disk group expansion  
Disk group defragmentation  
The priority of each of these operations can be changed to address  
performance requirements of the environment in which the operations are to  
be executed.  
NOTE: Setting a high priority level impacts storage array performance. It is not  
advisable to set priority levels at the maximum level. Priority should also be  
assessed in terms of impact to host server access and time to complete an  
operation. For example, the longer a rebuild of a degraded virtual disk takes, the  
greater the risk for potential secondary disk failure.  
50  
Planning: MD3600f Series Storage Array Terms and Concepts  
   
Virtual Disk Migration and Disk Roaming  
Virtual disk migration is moving a virtual disk or a hot spare from one array to  
another by detaching the physical disks and re-attaching them to the new  
array. Disk roaming is moving a physical disk from one slot to another on the  
same array.  
Disk Migration  
You can move virtual disks from one array to another without taking the  
target array offline. However, the disk group being migrated must be offline  
prior to performing the disk migration. If the disk group is not offline prior to  
migration, the source array holding the physical and virtual disks within the  
disk group marks them as missing. However, the disk groups themselves  
migrate to the target array.  
An array can import a virtual disk only if it is in an optimal state. You can  
move virtual disks that are part of a disk group only if all members of the disk  
group are being migrated. The virtual disks automatically become available  
after the target array has finished importing all the disks in the disk group.  
When you migrate a physical disk or a disk group from one MD3600f array to  
another, the MD3600f array you migrate to, recognizes any data structures  
and/or metadata you had in place on the migrating MD3600f array. However,  
if you are migrating from any device other than a MD3600f Series Storage  
Array, the MD3600f array does not recognize the migrating metadata and that  
data is lost. In this case, the MD3600f initializes the physical disks and mark  
them as unconfigured capacity.  
NOTE: Only disk groups and associated virtual disks with all member physical disks  
present can be migrated from one storage array to another. it is recommended that  
you only migrate disk groups that have all their associated member virtual disks in  
an optimal state.  
NOTE: The number of physical disks and virtual disks that a storage array supports  
limits the scope of the migration.  
Planning: MD3600f Series Storage Array Terms and Concepts  
51  
     
Use either of the following methods to move disk groups and virtual disks:  
Hot virtual disk migration—Disk migration with the destination storage  
array power turned on.  
Cold virtual disk migration—Disk migration with the destination storage  
array power turned off.  
NOTE: To ensure that the migrating disk groups and virtual disks are correctly  
recognized when the target storage array has an existing physical disk, use hot  
virtual disk migration.  
When attempting virtual disk migration, follow these recommendations:  
Moving physical disks to the destination array for migration—When  
inserting drives into the destination storage array during hot virtual disk  
migration, wait for the inserted physical disk to be displayed in MDSM  
before inserting the next physical disk.  
WARNING: Without the delay between drive insertions, the storage array  
can become unstable and manageability is temporarily lost.  
Migrating virtual disks from multiple storage arrays into a single storage  
array—When migrating virtual disks from multiple or different storage  
arrays into a single destination storage array, move all of the physical disks  
from the same storage array as a set into the new destination storage array.  
Ensure that all of the physical disks from a storage array are migrated to  
the destination storage array before starting migration from the next  
storage array.  
NOTE: If the drive modules are not moved as a set to the destination storage  
array, the newly relocated disk groups might not be accessible.  
Migrating virtual disks to a storage array with no existing physical disks—  
Turn off the destination storage array, when migrating disk groups or a  
complete set of physical disks from a storage array to another storage array  
that has no existing physical disks. After the destination storage array has  
been turned on and has successfully recognized the newly migrated  
physical disks, migration operations can continue.  
NOTE: Disk groups from multiple storage arrays should not be migrated at the  
same time to a storage array that has no existing physical disks.  
52  
Planning: MD3600f Series Storage Array Terms and Concepts  
Enabling premium features before migration—Before migrating disk  
groups and virtual disks, enable the required premium features on the  
destination storage array. If a disk group is migrated from a MD3600f  
storage array that has a premium feature enabled and the destination array  
does not have this feature enabled, an Out of Compliance error message  
can be generated.  
Disk Roaming  
You can move physical disks within an array. The RAID controller module  
automatically recognizes the relocated physical disks and logically places  
them in the proper virtual disks that are part of the disk group. Disk roaming  
is permitted when the RAID controller module is either online or  
powered off.  
NOTE: The disk group must be exported before moving the physical disks.  
Advanced Features  
The RAID enclosure supports several advanced features:  
Virtual Disk Snapshots  
Virtual Disk Copy  
High Performance Tier  
Remote Replication (this premium feature is supported only in storage  
arrays with the Fibre Channel [FC] host ports)  
NOTE: Virtual Disk Snapshot, Virtual Disk Copy, High Performance Tier, and Remote  
Replication are premium features that must be activated separately. If you have  
purchased these features, an activation card is supplied that contains instructions  
for enabling this functionality.  
Host Server–to–Virtual Disk Mapping  
The host server attached to a storage array accesses various virtual disks on the  
storage array through its host ports. Specific virtual disk-to-LUN mappings to  
an individual host server can be defined. In addition, the host server can be part  
of a host group that shares access to one or more virtual disks.  
Planning: MD3600f Series Storage Array Terms and Concepts  
53  
         
You can manually configure a host server-to-virtual disk mapping. When you  
configure host server-to-virtual disk mapping, consider these guidelines:  
You can define one host server-to-virtual disk mapping for each virtual disk  
in the storage array.  
Host server-to-virtual disk mappings are shared between RAID controller  
modules in the storage array.  
A unique LUN must be used by a host group or host server to access a  
virtual disk.  
Not every operating system has the same number of LUNs available for use.  
Host Types  
A host server is a server that accesses a storage array. Host servers are mapped  
to the virtual disks. Host servers have the following attributes:  
Host name—A name that uniquely identifies the host server.  
Host group (used in Cluster solutions only)—Two or more host servers  
associated together to share access to the same virtual disks.  
This host group is a logical entity you can create in MDSM. All host servers  
in a host group must be running the same operating system.  
Host type—The operating system running on the host server.  
Snapshot Virtual Disks  
A snapshot is a point-in-time image of a virtual disk. The snapshot provides  
an image of the virtual disk at the time the snapshot was created. You create a  
snapshot so that an application (for example, a backup application) can  
access the snapshot and read the data while the source virtual disk remains  
online and user-accessible. When the backup is completed, the snapshot  
virtual disk is no longer needed. You can create up to four snapshots per  
virtual disk.  
Snapshots are used to recover previous versions of files that have changed  
since the snapshot was taken. Snapshots are implemented using a copy-on-  
write algorithm, which makes a backup copy of data the instant an error  
occurs. Data on a virtual disk is copied to the snapshot repository before it is  
modified. Snapshots are instantaneous and take up less overhead than a full  
physical copy process.  
54  
Planning: MD3600f Series Storage Array Terms and Concepts  
     
Snapshot Repository Virtual Disk  
When you create a snapshot virtual disk, it automatically creates a snapshot  
repository virtual disk. A snapshot repository is a virtual disk created in the  
storage array as a resource for a snapshot virtual disk. A snapshot repository  
virtual disk contains snapshot virtual disk metadata and copy-on-write data for  
a particular snapshot virtual disk. The repository supports one snapshot only.  
You cannot select a snapshot repository virtual disk as a source virtual disk or  
as a target virtual disk in a virtual disk copy. If you select a Snapshot source  
virtual disk as the target virtual disk of a virtual disk copy, you must disable all  
snapshot virtual disks associated with the source virtual disk.  
CAUTION: Before using the Snapshot Virtual Disks Premium Feature in a  
Windows Clustered configuration, you must map the snapshot virtual disk to the  
cluster node that owns the source virtual disk. This ensures that the cluster nodes  
correctly recognize the snapshot virtual disk.  
Mapping the snapshot virtual disk to the node that does not own the source  
virtual disk before the snapshot enabling process is completed can result in  
the operating system incorrectly identifying the snapshot virtual disk. This  
can result in data loss or an inaccessible snapshot.  
For details on mapping the snapshot virtual disk to the secondary node, see  
the Dell PowerVault MD3600f and MD3620f Storage Arrays With Microsoft  
Windows Server Failover Clusters on support.dell.com/manuals.  
Virtual Disk Copy  
Virtual disk copy is a premium feature you can use to:  
Back up data  
Copy data from disk groups that use smaller-capacity physical disks to disk  
groups using greater capacity physical disks  
Restore snapshot virtual disk data to the source virtual disk  
Virtual disk copy generates a full copy of data from the source virtual disk to  
the target virtual disk in a storage array.  
Planning: MD3600f Series Storage Array Terms and Concepts  
55  
       
Source Virtual Disk  
When you create a virtual disk copy, a copy pair consisting of a source virtual  
disk and a target virtual disk is created on the same storage array. When a  
virtual disk copy is started, data from the source virtual disk is copied  
completely to the target virtual disk.  
Target Virtual Disk  
When you start a virtual disk copy, the target virtual disk maintains a copy of  
the data from the source virtual disk. You can choose whether to use an  
existing virtual disk or create a new virtual disk as the target virtual disk. If  
you choose an existing virtual disk as the target, all data on the target is  
overwritten. A target virtual disk can be a standard virtual disk or the source  
virtual disk of a failed or disabled snapshot virtual disk.  
NOTE: The target virtual disk capacity must be equal to or greater than the source  
virtual disk capacity.  
When you begin the disk copy process, you must define the rate at which the  
copy is completed. Giving the copy process top priority slightly impacts I/O  
performance, while giving it lowest priority makes the copy process longer to  
complete. You can modify the copy priority while the disk copy is in progress.  
For more information, see the online help topics.  
Virtual Disk Recovery  
You can use the Edit host server-to-virtual disk mappings feature to recover  
data from the backup virtual disk. This functionality enables you to unmap  
the original source virtual disk from its host server, then map the backup  
virtual disk to the same host server.  
Ensure that you record the LUN used to provide access to the source virtual  
disk. You need this information when you define a host server-to-virtual disk  
mapping for the target (backup) virtual disk. Also, be sure to stop all I/O activity  
to the source virtual disk before beginning the virtual disk recovery procedure.  
56  
Planning: MD3600f Series Storage Array Terms and Concepts  
       
Using Snapshot and Disk Copy Together  
You can use the Snapshot Virtual Disk and Virtual Disk Copy premium  
features together to back up data on the same storage array, or to restore the  
data on the snapshot virtual disk to its original source virtual disk.  
You can copy data from a virtual disk in one of two ways:  
By taking a point-in-time snapshot of the data  
By copying the data to another virtual disk using a virtual disk copy  
You can select a snapshot virtual disk as the source virtual disk for a virtual  
disk copy. This configuration is one of the best ways you can apply the  
snapshot virtual disk feature, since it enables complete backups without any  
impact to the storage array I/O.  
You cannot use a snapshot repository virtual disk as a source virtual disk or as  
a target virtual disk in a virtual disk copy. If you select the source virtual disk  
as the target virtual disk of a virtual disk copy, you must disable all snapshot  
virtual disks associated with the source virtual disk.  
Multi–Path Software  
Multi-path software (also referred to as the failover driver) is software resident  
on the host server that provides management of the redundant data path  
between the host server and the storage array.  
The multi-path software identifies the existence of multiple paths to a virtual  
disk and establishes a preferred path to that disk. If any component in the  
preferred path fails, the multi-path software automatically re-routes I/O  
requests to the alternate path so that the storage array continues to operate  
without interruption.  
NOTE: Multi-path software is available on the MD3600f Series resource media.  
Preferred and Alternate Controllers and Paths  
A preferred controller is a RAID controller module designated as the owner of  
a virtual disk or disk group. The preferred controller is automatically selected  
by MDSM when a virtual disk is created. You can change the preferred RAID  
controller module owner of a virtual disk after it is created. If a host is only  
connected to only one RAID controller module, the preferred owner must  
manually be assigned to the RAID controller module the host can access.  
Planning: MD3600f Series Storage Array Terms and Concepts  
57  
           
Ownership of a virtual disk is moved from the preferred controller to the  
secondary controller (also called the alternate controller) when the preferred  
controller is:  
Physically removed  
Updating firmware  
Involved in an event that caused failover to the alternate controller  
Paths used by the preferred RAID controller module to access either the disks  
or the host server are called the preferred paths; redundant paths are called  
the alternate paths. If a failure causes the preferred path to become  
inaccessible, the storage array automatically uses the alternate path to access  
data, when this occurs the enclosure status LED blinks amber.  
Virtual Disk Ownership  
MDSM can be used to automatically build and view virtual disks. It uses  
optimal settings to stripe the disk group. Virtual disks are assigned to  
alternating RAID controller modules when they are created. This default  
assignation provides a simple means for load balancing the workload of the  
RAID controller modules.  
Ownership can later be modified to balance workload according to actual  
usage. If virtual disk ownership is not manually balanced, it is possible for one  
controller to have the majority of the work, while the other controller is idle.  
Limit the number of virtual disks in a disk group. If multiple virtual disks are  
in a disk group, consider:  
The impact each virtual disk has on other virtual disks in the same  
disk group.  
The patterns of usage for each virtual disk.  
Different virtual disks have higher usage at different times of day.  
Load Balancing  
A load balance policy is used to determine which path is used to process I/O.  
Multiple options for setting the load balance policies lets you optimize I/O  
performance when mixed host interfaces are configured.  
You can choose one of these load balance policies to optimize  
I/O performance:  
58  
Planning: MD3600f Series Storage Array Terms and Concepts  
       
Round robin with subset—The round robin with subset I/O load balance  
policy routes I/O requests, in rotation, to each available data path to the  
RAID controller module that owns the virtual disks. This policy treats all  
paths to the RAID controller module that owns the virtual disk equally for  
I/O activity. Paths to the secondary RAID controller module are ignored  
until ownership changes. The basic assumption for the round-robin policy  
is that the data paths are equal. With mixed host support, the data paths  
might have different bandwidths or different data transfer speeds.  
Least queue depth with subset—The least queue depth with subset policy  
is also known as the least I/Os or least requests policy. This policy routes  
the next I/O request to a data path that has the least outstanding I/O  
requests queued. For this policy, an I/O request is simply a command in  
the queue. The type of command or the number of blocks that are  
associated with the command are not considered.The least queue depth  
with subset policy treats large block requests and small block requests  
equally. The data path selected is one of the paths in the path group of the  
RAID controller module that owns the virtual disk.  
Least path weight with subset (Windows operating systems only)—The  
least queue depth with subset policy is also known as the least I/Os or least  
requests policy. This policy routes the next I/O request to a data path that  
has the least outstanding I/O requests queued. For this policy, an I/O  
request is simply a command in the queue. The type of command or the  
number of blocks that are associated with the command are not  
considered. The least queue depth with subset policy treats large block  
requests and small block requests equally. The data path selected is one of  
the paths in the path group of the RAID controller module that owns the  
virtual disk.  
Monitoring MD3600f Series System Performance  
You can use the Performance Monitor to select virtual disks and RAID  
controller modules to monitor or to change the polling interval. Keep these  
guidelines in mind when using the Performance Monitor:  
The Performance Monitor does not dynamically update its display if any  
configuration changes occur while the window is open. You must close the  
Performance Monitor window and reopen it for the changes to appear.  
Planning: MD3600f Series Storage Array Terms and Concepts  
59  
   
Using the Performance Monitor to retrieve performance data can affect  
the normal storage array performance depending on the polling interval  
that you set.  
If the storage array you are monitoring begins in or transitions to an  
unresponsive state, an informational dialog appears. The dialog informs  
you that the Performance Monitor cannot poll the storage array for  
performance data.  
To monitor the performance of the arrays:  
1
2
Open MDSM and select the appropriate storage array.  
Open the Array Management Window (AMW) for the selected  
storage array.  
3
4
In the AMW, select Storage Array  
Monitor Performance.  
Click Settings  
.
a
b
Select the components that you want to monitor. You can monitor:  
RAID controller modules  
Virtual disks  
Storage array totals  
NOTE: By default, all items are selected.  
In Polling interval, select how often you want to update the  
performance statistics.  
NOTE: For an accurate elapsed time, do not use the Set RAID Controller  
Module Clocks option while using the Performance Monitor.  
Each time the polling interval elapses, the Performance Monitor queries  
the storage array again and updates the statistics in the table.  
5
Click Start.  
Values appear for the selected storage arrays in the Performance Monitor  
data table. The table is updated at the interval specified in the Polling  
Interval setting.  
6
7
8
Click Update to force an immediate poll of the storage array.  
Click Stop to stop monitoring the storage array.  
Click Save As on the Performance Monitor main dialog to save the  
currently displayed performance statistics.  
60  
Planning: MD3600f Series Storage Array Terms and Concepts  
9
Select an appropriate directory.  
10 Type a file name in the File name text box.  
NOTE: The .perf extension is the default.  
11 Select a file type from the Files of type list.  
Use the Report format (ASCII text) file type to save the data to a  
report form for viewing or printing.  
Use the Comma Delimited Format file type to save the data in a form  
that can be imported into a commercial spreadsheet application for  
further analysis. Most leading commercial spreadsheet applications  
recognize a comma delimiter. These applications use the delimiter to  
import the data into spreadsheet cells.  
12 Click Save  
.
The Performance Monitor data provides information about how your storage  
array is performing. The data is presented in eight columns, which are  
described in this table. Use this data to make performance tuning decisions  
for your storage array.  
Table 4-3. Performance Monitor Table Description  
Column Headings  
Devices  
Description  
Controller, virtual disk or storage array total.  
Cumulative I/O’s per second from last start time.  
Percentage of cumulative I/O’s that are reads.  
Total IOs  
Read Percentage  
Cache Hit Percentage Percentage of cumulative I/O’s that are in-cache.  
Current KB/second  
Snapshot of throughput value per second (1KB =  
1024 bytes).  
Maximum KB/second Maximum recorded throughput value from last start time.  
Current IO/second  
Snapshot of I/O’s per second (IOP = Input/output per  
second or one completed I/O transaction).  
Maximum IO/second Maximum recorded IOP from last start time.  
For more information, see the MD PowerVault Support Matrix at  
support.dell.com/manuals.  
Planning: MD3600f Series Storage Array Terms and Concepts  
61  
62  
Planning: MD3600f Series Storage Array Terms and Concepts  
5
Configuration: Overview  
Dell PowerVault Modular Disk Storage Manager (MDSM) online help  
contains information on the how to use the MDSM application to perform  
the configuration and management tasks described in this document. You can  
access online help by selecting Help located at the top right corner of MDSM  
interface. See the MD3600f and MD3620f Storage Array’s Deployment Guide  
for information on installing the MDSM.  
NOTE: MDSM supports the MD3000 series, MD3200 series, and MD3600 series  
storage arrays and can automatically detect all of these storage arrays.  
User Interface  
The Storage Manager screen is divided into two primary windows:  
Enterprise Management Window (EMW)—The EMW provides high-  
level management of the storage arrays. You can launch the Array  
Management Window from the EMW.  
Array Management Window (AMW)—The AMW provides management  
functions for a single storage array. You can launch more than one AMW at  
the same time to manage different storage arrays.  
The EMW and the AMW consist of the following:  
The title bar at the top of the window shows the name of the application.  
Beneath the title bar, is the menu bar. You can select menu options from  
the menu bar to perform tasks on a storage array.  
Beneath the menu bar, is the toolbar. You can select options in the toolbar  
to perform tasks on a storage array.  
Beneath the toolbar, are the tabs. Tabs are used to group the tasks that you  
can perform on a storage array.  
Beneath the tabs, is the status bar. The status bar shows status messages  
and status icons related to the storage array.  
NOTE: By default, the toolbar and status bar do not appear. To view the toolbar or  
the status bar, select ViewToolbar or ViewStatus Bar, respectively.  
Configuration: Overview  
63  
     
Enterprise Management Window  
The EMW provides high-level management of storage arrays. When you start  
MDSM, the EMW opens. The EMW has these tabs:  
Devices tab—Provides information about the storage arrays.  
Setup tab—Presents the initial setup tasks that guide you through adding  
storage arrays and configuring alerts.  
The Devices tab has a Tree view on the left side of the window that shows  
discovered storage arrays, unidentified storage arrays, and the status  
conditions for the storage arrays. Discovered storage arrays are managed by  
MDSM. Unidentified storage arrays are available to MDSM but not  
configured for management. The right side of the Devices tab has a Table  
view that shows detailed information for each storage array.  
In the EMW, you can:  
Discover hosts and managed storage arrays on the local sub-network.  
Manually add and remove hosts and storage arrays.  
Locate the storage arrays.  
Name or rename discovered storage arrays.  
Add storage array comments to the Table view.  
Sort rows in the Table view according to different criteria.  
Store your EMW view preferences and configuration data in local  
configuration files. The next time you open the EMW, data from the local  
configuration files is used to show customized view and preferences.  
Monitor the status of managed storage arrays and indicate status using  
appropriate icons.  
Add or remove management connections.  
Configure alert notifications for all selected storage arrays through e-mail  
or SNMP traps.  
Report critical events to the configured alert destinations.  
Launch the AMW for a selected storage array.  
Run a script to perform batch management tasks on specific storage arrays.  
Import the operating system theme settings into the MDSM.  
Upgrade firmware on multiple storage arrays concurrently.  
64  
Configuration: Overview  
   
Inheriting the System Settings  
Use the Inherit System Settings option to import the operating system theme  
settings into the MDSM. Importing system theme settings affects such  
features as font type, font size, color, and contrast in the MDSM.  
1
Open the Inherit System Settings window in one of these ways:  
Select ToolsInherit System Settings  
.
Select the Setup tab, and click Inherit System Settings  
.
2
3
Select Inherit system settings for color and font  
.
Click OK  
.
Array Management Window  
You can launch the AMW from the EMW. The AMW provides management  
functions for a single storage array. You can have multiple AMWs open  
simultaneously to manage different storage arrays.  
To launch the AMW:  
1
2
In the EMW, on the Devices tab, double click on the relevant storage array.  
The context menu for the selected storage is displayed.  
In the context menu, select Manage Storage Array.  
The AMW for the selected storage is displayed.  
The AMW has the following tabs:  
Summary tab—You can view the following information about the  
storage array:  
Status  
Hardware components  
Capacity  
Hosts and Mappings  
Storage partitions  
Disk groups and virtual disks  
Logical tab—You can view the organization of the storage array by virtual  
disks, disk groups, free capacity nodes, and any unconfigured capacity for  
the storage array.  
Configuration: Overview  
65  
     
Physical tab—You can view the organization of the storage array by RAID  
controller modules, physical disks, and other hardware components.  
Mappings tab—You can define the hosts, host groups, and host ports. You  
can change the mappings to grant virtual disk access to host groups and  
hosts and create storage partitions.  
Setup tab—You can complete the initial setup tasks to configure the  
storage array.  
SupportYou can complete common support tasks like downloading  
RAID controller module firmware, viewing the online help, and so on.  
In the AMW, you can:  
Provide storage array options, for example, renaming a storage array,  
changing a password, or enabling a background media scan.  
Provide the ability to configure virtual disks from the storage array  
capacity, define hosts and host groups, and grant host or host group access  
to sets of virtual disks called storage partitions.  
Monitor the health of storage array components and report detailed status  
using applicable icons.  
Provide applicable recovery procedures for a failed logical component or a  
failed hardware component.  
Present a view of the Event Log for the storage array.  
Present profile information about hardware components, such as RAID  
controller modules and physical disks.  
Provide RAID controller module management options, such as changing  
ownership of virtual disks or placing a RAID controller module online  
or offline.  
Provide physical disk management options, such as assignment of hot  
spares and locating the physical disk.  
Monitor storage array performance.  
66  
Configuration: Overview  
6
Configuration: About Your Storage  
Array  
Out–of–Band and In–Band Management  
You can manage a storage array in two ways:  
Out-of-band management  
In-band management  
Out-of-Band Management  
Using the out-of-band management method, data is separated from commands  
and events. Data travels through the host-to-controller interface, while  
commands and events travel through the management port Ethernet cables.  
This management method lets you configure the maximum number of  
virtual disks that are supported by your operating system and host adapters. A  
maximum of eight storage management stations can concurrently monitor an  
out-of-band managed storage array. This limit does not apply to systems that  
manage the storage array through the in-band management method.  
When you use out-of-band management, you must set the network  
configuration for each RAID controller module’s management Ethernet port.  
This includes the Internet Protocol (IP) address, subnetwork mask (subnet  
mask), and gateway. If you are using a Dynamic Host Configuration Protocol  
(DHCP) server, you can enable automatic network configuration, but if you are  
not using a DHCP server, you must enter the network configuration manually.  
NOTE: RAID controller module network configurations can be assigned using a  
DHCP server (the default setting). However, if a DHCP server is not available for 150  
seconds, the RAID controller modules assign static IP addresses. The addresses  
assigned are 192.168.128.101 for controller 0 and 192.168.128.102 for controller 1.  
Configuration: About Your Storage Array  
67  
     
In-Band Management  
Using in-band management, commands, events, and data travel through the  
host-to-controller interface. Unlike out-of-band management, commands and  
events are mixed with data.  
NOTE: For detailed information on setting up in-band and out-of-band management  
see the Deployment Guide.  
When you add storage arrays by using this management method, you need to  
specify only the host name or IP address of the host. After you add the  
specific host name or IP address, the host-agent software automatically  
detects any storage arrays that are connected to that host.  
CAUTION: Some operating systems can be used only as storage management  
stations. For more information about the operating system that you are using, see  
the MD PowerVault Support Matrix at support.dell.com/manuals.  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Access Virtual Disk  
Each RAID controller module in an MD3600f Series storage array maintains a  
special virtual disk, called the access virtual disk. The host-agent software uses  
the access virtual disk to communicate management requests and event  
information between the storage management station and the RAID  
controller module in an in-band–managed storage array. The access virtual  
disk is not available for application data storage. The default LUN is 31.  
Storage Arrays  
You must add the storage arrays to MDSM before you can setup the storage  
array for optimal use.  
Adding Storage Arrays  
You can add storage arrays only in the EMW. You can:  
Automatically discover storage arrays  
Manually add storage arrays  
NOTE: Verify that your host or management station network configuration—  
including station IP address, subnet mask, and default gateway—is correct before  
adding a new storage array using the Automatic option.  
68  
Configuration: About Your Storage Array  
         
NOTE: For Linux, set the default gateway so that broadcast packets are sent to  
255.255.255.0. For Red Hat Enterprise Linux, if no gateway exists on the network, set  
the default gateway to the IP address of the NIC.  
NOTE: MDSM uses TCP/UDP port 2463 for communication to the MD Storage Array.  
Automatic Discovery of Storage Arrays  
The Automatic Discovery process sends out a broadcast message across the  
local subnetwork (subnet) and adds any storage array that responds to the  
message. The Automatic Discovery process finds both in-band and out-of-  
band storage arrays.  
NOTE: The Automatic Discovery option and the Re-scan Hosts option in the  
Enterprise Management Window provide automatic methods for discovering  
managed storage arrays.  
Manual Addition of a Storage Array  
Use Manual Addition if the storage array resides outside of the local subnet.  
This process requires specific identification information to manually add a  
storage array.  
To add a storage array that uses out-of-band management, specify the host  
name or management port IP address of each controller in the storage array.  
Before using this option, verify that the applicable network configuration  
tasks have been performed.  
To add an in-band storage array, add the host through which the storage array  
is attached to the network.  
NOTE: It can take several minutes for MDSM to connect to the specified  
storage array.  
To add a storage array manually:  
1
2
Select EditAdd Storage Array  
.
Select the relevant management method:  
Out-of-band management—Enter a host name or an IP address for  
the RAID controller Modules in the storage array.  
In-band management—Enter a name or an IP address for the Host  
through which the storage array is attached to the network.  
Configuration: About Your Storage Array  
69  
   
NOTE: The host agent must be restarted before in-band management  
communication can be established. See "Starting or Restarting the Host  
Context Agent Software" on page 285.  
3
4
Click Add  
.
Use one of these methods to name a storage array:  
In the EMW, select the Setup tab, and select Name/Rename  
Storage Arrays  
In the AMW, select the Setup tab, and select Rename Storage Array  
In the EMW right-click the icon corresponding to the array and  
select Rename  
.
.
.
Setting Up Your Storage Array  
A list of initial setup tasks is displayed on the Setup tab in the AMW. The list  
of initial setup tasks shows you how to set up a storage array. Using the steps  
outlined in the Initial Setup Tasks area, ensures that the basic setup steps are  
completed properly.  
Use the Initial Setup Tasks list the first time that you set up a storage array to  
perform these tasks:  
Locate the storage array—Find the physical location of the storage array on  
your network by turning on the unit identify LEDs. The storage array can  
be identified with a label.  
Give a new name to the storage array—Use a unique name that identifies  
each storage array.  
Set a storage array password—Configure the storage array with a password  
to protect it from unauthorized access. MDSM prompts for the password  
when you attempt to change the storage array configuration. For example,  
when a virtual disk is created or deleted.  
Configure the storage array—Create disk groups, virtual disks, and hot  
spare physical disks by using the Automatic configuration method or the  
Manual configuration method. For more information, see the PowerVault  
Modular Disk Storage Manager online help topics.  
Map Virtual Disks—Map virtual disks to hosts or host groups.  
70  
Configuration: About Your Storage Array  
 
Save Configuration—Save the configuration parameters in a file that you  
can use to restore the configuration, or reuse the configuration on another  
storage array. For more information, see the PowerVault Modular Disk  
Storage Manager online help topics.  
After you complete the basic steps for configuring the storage array, you can  
perform these optional tasks:  
Manually define hosts—Define the hosts and the host port identifiers that  
are connected to the storage array. Use this option only if the host is not  
automatically recognized and shown in the Mappings tab.  
Configure ethernet management ports—Configure the network  
parameters for the Ethernet management ports on the RAID controller  
modules if you are managing the storage array by using the out-of-band  
management connections.  
View and enable premium features—Your MDSM might include premium  
features. View the premium features that are available and the premium  
features that are already started. You can start available premium features  
that are currently stopped.  
Locating Storage Arrays  
You can use the Blink option to physically locate and identify a storage array.  
To locate the storage array:  
1
Select the relevant storage array and:  
In the EMW, right-click the appropriate storage array, and select Blink  
Storage Array  
In the AMW, select the Setup tab, click Blink Storage Array  
In the AMW, select Storage ArrayBlinkStorage Array  
.
.
.
The LEDs blink on the physical disks in the storage array.  
2
3
After locating the storage array, click OK.  
The LEDs stop blinking.  
If the LEDs do not stop blinking, select Storage ArrayBlink  
Stop All Indications  
A confirmation message is displayed.  
.
4
Click OK.  
Configuration: About Your Storage Array  
71  
   
Naming or Renaming Storage Arrays  
Each storage array should be assigned a unique up to 30-character  
alphanumeric name. A name can consist of letters, numbers, and the special  
characters underscore (_), dash (–), and pound sign (#). No other special  
characters are allowed.  
To rename a selected storage array:  
1
Perform one of these actions:  
In the AMW Setup tab, select Rename Storage Array.  
In the EMW Devices tab Tree view, select EditRename.  
In the EMW Devices tab Table view, select EditRename.  
In the EMW Devices tab Tree view, right-click on the desired array  
icon and select Rename.  
The Name/Rename Storage Arrays dialog appears.  
2
Select the relevant storage array from the Select storage array table.  
If you do not know the name or physical location of the storage array, click  
Blink. After locating the storage array, click OK to turn off the LEDs.  
The name of the storage array appears in the Storage array name  
.
3
4
In Storage array name, type the new name of the storage array. If  
applicable, add a comment for the storage array in Additional comment  
.
Click Apply  
.
A message is displayed warning you about the implications of changing the  
storage array name.  
5
6
Click Yes.  
The new storage array name appears in the Select storage array table.  
Repeat step 2 through step 4 to name or rename additional storage arrays.  
NOTE: Avoid arbitrary names or names that might lose meaning in the future.  
72  
Configuration: About Your Storage Array  
   
Setting a Password  
You can configure each storage array with a password to protect it from  
unauthorized access. MDSM prompts for the password when an attempt is  
made to change the storage array configuration, such as, when a virtual disk is  
created or deleted. View operations do not change the storage array  
configuration and do not require a password. You can create a new password  
or change an existing password.  
NOTE: It is recommended that you use a long password with at least 15  
alphanumeric characters to increase security.  
To set a new password or change an existing password:  
1
Select the relevant storage array and navigate to the AMW for that storage  
array. See "Array Management Window" on page 65.  
The AMW for the selected storage array is displayed.  
In the AMW, perform one of these actions:  
2
Select the storage array in the Logical pane, and then select Storage  
ArraySet Password  
.
Select the Setup tab, and then click Set a Storage Array Password.  
Select the Logical tab, right-click and select Set Password.  
The Set Password dialog appears.  
3
If you are resetting the password, type the Current password.  
NOTE: If you are setting the password for the first time, leave the Current  
password blank.  
4
5
Type the New password  
.
Re-type the new password in Confirm new password  
.
NOTE: The password in Confirm new password and New password must be  
exactly the same.  
6
Click OK  
.
NOTE: You are not prompted for a password when you attempt to change the  
storage array configuration in the current management session.  
Configuration: About Your Storage Array  
73  
   
Password Guidelines  
Follow these guidelines when you create a password:  
Use secure passwords for your storage array. A password should be easy for  
you to remember but difficult for others to determine. Consider using  
numbers or special characters in the place of letters, such as a 1 in the  
place of the letter I, or the at sign (@) in the place of the letter a.  
For increased protection, use a long password with at least 15  
alphanumeric characters. The maximum password length is 30 characters.  
Passwords are case sensitive.  
NOTE: You can attempt to enter a password up to ten times before the storage  
array enters a lockout state. Before you can try to enter a password again, you must  
wait 10 minutes for the storage array to reset. To reset the password, press the  
password reset switch on your RAID controller module, see Figure 3-1.  
Viewing Storage Array Connections  
You can use the View Connections option to view the expansion enclosures  
connected to the RAID controller module.  
To view the storage array connections:  
1
From the toolbar in AMW, select Storage ArrayViewConnections.  
The <Storage Array>:Connections dialog is displayed.  
2
3
Click the column name to sort the connections according to your preference.  
Click Close  
.
If you receive an error message for a port, you can use this dialog to identify  
the components on the port that might have caused the error. By isolating  
these components, you prevent accidentally disconnecting components that  
are still in operation, which could cause an interruption in data flow.  
74  
Configuration: About Your Storage Array  
 
Adding/Editing a Comment to an Existing Storage Array  
A descriptive comment, with an applicable storage array name, is a helpful  
identification tool. You can add or edit a comment for a storage array in the  
EMW only.  
To add or edit a comment:  
1
In the EMW, select the Devices tab and select the relevant managed  
storage array.  
2
Select EditComment  
.
The Edit Comment dialog is displayed.  
3
4
Type the 60-character comment.  
Click OK  
.
This option updates the comment in the table view and saves it in your local  
storage management station file system. The comment does not appear to  
administrators who are using other storage management stations.  
Removing Storage Arrays  
You can remove a storage array from the list of managed arrays if you no  
longer want to manage it from a specific storage management station.  
Removing a storage array does not affect the storage array or its data in any  
way. Removing a storage array simply removes it from the list of storage arrays  
that appear in the drop-down list in the Array Selector. If a storage array is  
accidentally removed, it can be added again. See "Adding Storage Arrays" on  
page 68.  
You can remove the storage array only from the EMW.  
To remove the storage array:  
1
In the EMW, select the Devices tab and select the relevant managed  
storage array.  
2
Select EditRemoveStorage Array  
.
A message prompts you for a confirmation for the removal of the selected  
storage array.  
3
To remove the storage array, click Yes.  
Configuration: About Your Storage Array  
75  
   
Enabling Premium Features  
You can enable premium features on the storage array. To enable the premium  
features, you must obtain a feature key file specific to the premium feature  
that you want to enable from your storage supplier.  
To enable premium features:  
1
From the toolbar in AMW, select Storage ArrayPremium Features.  
The Premium Features and Feature Pack Information window is  
displayed.  
2
Select the relevant premium feature, and click Enable  
.
The Select Feature Key File dialog is displayed.  
3
4
Navigate to the relevant folder, select the appropriate key file, and click OK  
.
Click Close  
.
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Failover Alert Display  
You can change the failover alert delay for a storage array. The failover alert  
delay lets you delay the logging of a critical event if the multi-path driver  
transfers virtual disks to the non-preferred controller. If the multi-path driver  
transfers the virtual disks back to the preferred controller within the specified  
delay period, a critical event is not logged. If the transfer exceeds this delay  
period, then a virtual disk-not-on-preferred-path alert is issued as a critical  
event. You can also use this option to minimize multiple alerts when more than  
one virtual disk fails over because of a system error, such as a failed host adapter.  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
76  
Configuration: About Your Storage Array  
   
Changing the Cache Settings on the Storage Array  
1
In the AMW, select Storage ArrayChangeCache Settings.  
The Change Cache Settings window is displayed.  
2
3
4
Select or enter the percentage of unwritten data in the cache to trigger a  
cache flush in Start flushing  
Select or enter the percentage of unwritten data in the cache to stop a  
cache flush in progress in Stop flushing  
Select the appropriate Cache block size  
.
.
.
A smaller cache size is a good choice for file-system use or database-  
application use. A larger cache size is a good choice for applications that  
generate sequential I/O, such as multimedia.  
5
In the Enter Password dialog, type the current password for the storage  
array, and click OK  
.
Changing Expansion Enclosure ID Number  
When an MD1200 series expansion enclosure is connected to an MD3600f  
series storage array for the first time, an enclosure ID number is assigned and  
maintained by the expansion enclosure. This enclosure ID number is also  
displayed in the MDSM.  
To change the enclosure ID numbers:  
1
In the AMW, select the storage array, and select Storage Array  
ChangeEnclosure ID  
.
2
Select a new enclosure ID number from the Change Enclosure ID list.  
The enclosure ID must be between 0 and 99 (inclusive).  
3
To save the changed enclosure ID, click Change.  
Configuration: About Your Storage Array  
77  
   
Changing the Enclosure Order in the Physical Pane  
You can change the order of the RAID controller modules and the expansion  
enclosures in the Physical pane to match the hardware configuration in your  
storage array. The Physical pane that initially appears is a default view that  
may not match your storage array. The enclosure order change remains in  
effect until it is modified again.  
To change the enclosure order in the physical pane:  
1
2
In the AMW, select Storage ArrayChangeEnclosure Order  
.
From the Enclosures list, select the enclosure you want to move and click  
either Up or Down to move the enclosure to the new position.  
3
Click OK.  
If you have set a password for the selected storage array, the Enter  
Password dialog appears.  
4
5
Type the current password for the storage array.  
Click OK  
.
Configuring Alert Notifications  
MDSM can send an alert for any condition on the storage array that requires  
your attention. Alerts can be sent as e-mail messages or as Simple Network  
Management Protocol (SNMP) trap messages.  
You can configure alert notifications either for all the storage arrays or a single  
storage array.  
To configure alert notifications for all storage arrays:  
1
2
Select Configure Alerts  
The Configure Alerts dialog is displayed.  
Select All storage arrays  
Click OK  
.
3
4
.
.
The Configure Alerts dialog is displayed. To configure e-mail alerts, see  
"Configuring E–mail Alerts" on page 79. To configure SNMP alerts, see  
"Configuring SNMP Alerts" on page 81.  
78  
Configuration: About Your Storage Array  
   
To configure alert notifications for a single storage array:  
1
2
In the EMW, select the Devices tab.  
Select the relevant storage array, then select EditConfigureAlerts  
.
The Configure Alerts dialog is displayed. To configure e-mail alerts, see  
"Configuring E–mail Alerts" on page 79. To configure SNMP alerts, see  
"Configuring SNMP Alerts" on page 81.  
Configuring E–mail Alerts  
For more information on configuring alert notifications, see "Configuring  
Alert Notifications" on page 78.  
To configure e-mail alerts:  
1
Open the Configure Alerts dialog by performing one of these actions:  
In the Tree view or the Table view on the Devices tab in the EMW,  
select a node, and then select EditConfigure Alerts. Go to step 3.  
In the Setup tab in the EMW, select Configure Alerts. Go to step 2.  
2
Select one of the following radio buttons to specify an alert level:  
All storage arrays—Select this option to send an alert e-mail about  
events on all storage arrays.  
An individual storage array—Select this option to send an alert e-mail  
about events that occur on only a specified storage array.  
These results occur, depending on your selection:  
If you select all storage arrays, the Configure Alerts dialog appears.  
If you select an individual storage array, the Select Storage Array  
dialog appears. Select the storage array for which you want to receive  
e-mail alerts and click OK. The Configure Alerts dialog appears.  
If you do not know which storage array to select, click Blink to turn on  
the LEDs of the storage array.  
3
4
In the Configure Alerts dialog, select the Mail Server tab.  
In Mail server, type the name of the Simple Mail Transfer Protocol  
(SMTP) mail server.  
The SMTP mail server is the name of the mail server that forwards the  
alert e-mails to the configured e-mail addresses.  
Configuration: About Your Storage Array  
79  
     
5
6
7
In E-mail sender address, type the e-mail sender address. Use a valid e-  
mail address.  
The e-mail address of the sender (the network administrator) appears on  
each e-mail alert sent to the destination.  
To include the contact information of the sender in the e-mail alert, select  
Include contact information with the alerts, and type the contact  
information.  
NOTE: Including the contact information in the e-mail alert is optional.  
Select the e-mail tab to configure the e-mail destinations.  
Adding an e-mail address—In e-mail address, type the e-mail address,  
and click Add  
.
Replacing an e-mail address—In the Configured E-mail addresses  
area, select the e-mail address to be replaced, type the replacement e-  
mail address in E-mail address, and click Replace  
Deleting an e-mail address—In the Configured E-mail addresses area,  
select the e-mail address, and click Delete  
.
.
Validating an e-mail address—Type the e-mail address in E-mail  
address or select the e-mail address in the Configured E-mail  
addresses area, and click Test. A test e-mail is sent to the selected e-  
mail address. A dialog with the results of the test and any error  
appears.  
8
For the selected e-mail address, in Information To Send, select:  
Event Only—The alert e-mail contains only the event information.  
This alert type is the default.  
Event + Profile—The alert e-mail contains the event information and  
the storage array profile.  
Event + Support—The alert e-mail contains the event information  
and a compressed file that contains complete support information for  
the storage array that has generated the alert.  
80  
Configuration: About Your Storage Array  
9
For the selected e-mail address, in Frequency, select:  
Every event—Sends an alert e-mail whenever an event occurs. This is  
the default option.  
Every x hours—Sends an alert e-mail after the specified time interval  
if an event occurred during that time interval. You can select this  
option only if you have selected either Event + Profile or Event +  
Support in the Information To Send drop down list.  
10 Click OK  
.
An alert icon appears next to each node in the Tree view where an alert is set.  
To ensure that the e-mail is sent successfully:  
You must provide a SMTP mail server name and an e-mail sender address  
for the e-mail addresses to work.  
The e-mail addresses that you had previously configured appear in the  
Configured e-mail addresses area.  
You must use fully qualified e-mail addresses, for example,  
name@mycompany.com.  
You can configure multiple e-mail addresses before you click OK  
.
Configuring SNMP Alerts  
To add a management console to the list of addresses configured to receive  
SNMP alerts:  
1
Open the Configure Alerts dialog by performing one of these actions:  
In the Tree view or the Table view on the Devices tab in the EMW,  
select a node, and select Edit Configure Alerts. Go to step 3.  
In the Setup tab in the EMW, select Configure Alerts. Go to step 2.  
2
Select one of the following radio buttons to specify an alert level:  
All storage arrays—Select this option to send an alert notification  
about events on all storage arrays.  
An individual storage array—Select this option to send an alert  
notification about events that occur in only a specified storage array.  
Configuration: About Your Storage Array  
81  
     
These results occur, depending on your selection:  
If you select all storage arrays, the Configure Alerts dialog appears.  
If you select an individual storage array, the Select Storage Array  
dialog appears. Select the storage array for which you want to receive  
alert notifications and click OK. The Configure Alerts dialog appears.  
NOTE: If you do not know which storage array to select, click Blink to turn on  
the LEDs of the storage array.  
3
Select the SNMP tab to configure the SNMP alert destinations.  
Adding an SNMP address—In Community name, type the  
community name. In Trap destination, type the trap destination, and  
click Add  
.
NOTE: The community name is an American Standard Code for Information  
Interchange (ASCII) string that identifies a known set of network management  
stations and is set by the network administrator. The default community name  
is the string “public”.The trap destination is the IP address or the host name of  
a computer running an SNMP management application. An example of a  
SNMP enabled management application is the Dell Management Console.  
Please see dell.com for more information on the Dell Management Console.  
Replacing an SNMP address—Select the SNMP address in the  
Configured SNMP addresses area, type the replacement community  
name in Community name and the trap destination in Trap  
destination, and click Replace  
Deleting an SNMP address—Select the SNMP address in the  
Configured SNMP addresses area, and click Delete  
.
.
Validating an SNMP address—Select the SNMP address in the  
Configured SNMP addresses area, and click Test. A test message is  
sent to the SNMP address. A message box with the results of the  
validation and any error information appears.  
4
Click OK.  
An alert icon appears next to each node in the Tree view for which an alert  
is set.  
82  
Configuration: About Your Storage Array  
 
Follow these guideline for SNMP alerts:  
Any SNMP addresses that you had previously configured appear in the  
Configured SNMP addresses area.  
The SNMP Community Name is determined by the system administrator  
and configured within the a management application, such as the Dell  
Management Console. More information about the Dell Management  
Console is available at dell.com  
.
You can configure multiple SNMP addresses before you click OK  
.
Battery Settings  
A smart battery backup unit (BBU) can perform a learn cycle. The smart BBU  
module includes the battery, a battery gas gauge, and a battery charger. The  
learn cycle calibrates the smart battery gas gauge so that it provides a  
measurement of the charge of the battery module. A learn cycle can only start  
when the battery is fully charged.  
The learn cycle completes the following operations:  
Discharges the battery to a predetermined threshold.  
Charges the battery back to full capacity.  
A learn cycle starts automatically when you install a new battery module.  
Learn cycles for batteries in both RAID controller modules in a duplex system  
occur simultaneously.  
Learn cycles are scheduled to start automatically at regular intervals, at the  
same time and on the same day of the week. The interval between cycles is  
described in weeks.  
Use the following guidelines to adjust the interval:  
You can use the default interval.  
You can run a learn cycle at any time.  
You can set the learn cycle earlier than the currently scheduled time.  
You cannot set the learn cycle to start more than seven days later than the  
currently scheduled time.  
Configuration: About Your Storage Array  
83  
   
To change the battery settings perform these steps:  
1
In the AMW, select Storage ArrayChangeBattery Settings.  
The Battery Settings dialog opens.  
2
3
In Battery location, select a battery.  
Check these details about the battery:  
Battery status  
Battery age  
Days until replacement  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Setting the Storage Array RAID Controller Module Clocks  
You can use the Synchronize RAID Controller Module Clocks option to  
synchronize the storage array RAID controller module clocks with the storage  
management station. This option makes sure that the event timestamps  
written by the RAID controller modules to the Event Log match the event  
timestamps written to host log files. The RAID controller modules remain  
available during synchronization.  
To synchronize the RAID controller module clocks with the storage  
management station:  
1
In the AMW, select Storage Array  
Module Clocks  
If a password is set, in the Enter Password dialog, type the current  
password for the storage array, and click Synchronize  
Synchronize RAID Controller  
.
2
.
The RAID controller module clocks are synchronized with the storage  
management station.  
84  
Configuration: About Your Storage Array  
   
7
Configuration: Event Monitor  
An event monitor is provided with Dell PowerVault Modular Disk Storage  
Manager (MDSM). The event monitor runs continuously in the background  
and monitors activity on the managed storage arrays. If the event monitor  
detects any critical problems, it can notify a host or remote system using e-  
mail, Simple Network Management Protocol (SNMP) trap messages, or both.  
For the most timely and continuous notification of events, enable the event  
monitor on a management station that runs 24 hours a day. Enabling the  
event monitor on multiple systems or having a combination of an event  
monitor and MDSM active can result in duplicate events, but this does not  
indicate multiple failures on the array.  
The Event Monitor is a background task that runs independently of the  
Enterprise Management Window (EMW).  
To use the Event Monitor:  
Set up alert destinations for the managed device that you want to monitor.  
A possible alert destination would be the Dell Management Console. More  
information about the Dell Management Console can be found at  
dell.com  
.
Replicate the alert settings from a particular managed device by copying  
the emwdata.bin file to every storage management station from which you  
want to receive alerts.  
Each managed device shows a check mark that indicates that alerts have been set.  
Configuration: Event Monitor  
85  
   
Enabling or Disabling the Event Monitor  
You can enable or disable the event monitor at any time.  
Disable the event monitor if you do not want the system to send alert  
notifications. If you are running the event monitor on multiple systems,  
disabling the event monitor on all but one system prevents the sending of  
duplicate messages.  
NOTE: It is recommended that you configure the event monitor to start by default  
on a management station that runs 24 hours a day.  
Windows  
To enable or disable the event monitor:  
1
Click StartSettingsControl PanelAdministrative  
Tools  
Services.  
or  
Click StartAdministrative ToolsServices  
From the list of services, select Modular Disk Storage Manager  
Event Monitor  
Select ActionProperties  
To enable the event monitor, in the Service Status area, click Start  
To disable the event monitor, in the Service Status area, click Stop  
.
2
.
3
4
5
.
.
.
Linux  
To enable the event monitor, at the command prompt, type SMmonitor  
startand press <Enter>. When the program startup begins, the system  
displays the following message:  
SMmonitor started.  
To disable the event monitor, start terminal emulation application (console  
ox xterm) and at the command prompt, type SMmonitor stopand press  
<Enter>. When the program shutdown is complete, the system displays the  
following message:  
Stopping Monitor process.  
86  
Configuration: Event Monitor  
           
8
Configuration: About Your Host  
Configuring Host Access  
Dell PowerVault Modular Disk Storage Manager (MDSM) software is  
comprised of multiple modules. One of these modules is the Host Context  
Agent. The host context agent is installed as part of the MDSM installation  
and runs continuously in the background.  
If the host context agent is running on a host, that host and the host ports  
connected from it to the storage array are automatically detected by MDSM  
and appear on the Mappings tab in the Array Management Window (AMW).  
For more information on the Mappings tab, see "Using the Mappings Tab" on  
page 88.  
If the hosts are not detected automatically, you can use the Define Host  
Wizard to define the hosts that access the virtual disks in the storage array.  
Defining a host is one of the steps required to let the storage array know  
which hosts are attached to it and to allow access to the virtual disks. For  
more information on defining the hosts, see "Defining a Host" on page 88.  
To enable the host to write to the storage array, you must map the host to the  
virtual disk. This mapping grants a host or a host group access to a particular  
virtual disk or to a number of virtual disks in a storage array. You can define the  
mappings on the Mappings tab in the Array Management Window (AMW).  
On the Summary tab in the AMW, the Hosts & Mappings area indicates how  
many hosts are configured to access the storage array. You can click  
Configured Hosts in the Hosts & Mappings area to see the names of the hosts.  
A collection of elements, such as default host groups, hosts, and host ports,  
appear as nodes in the Topology pane of the Mappings tab in the AMW.  
Configuration: About Your Host  
87  
     
The host topology is reconfigurable. You can perform the following tasks:  
Create a host and assign an alias or user label.  
Add or associate a new host port identifier to a particular host.  
Change the host port identifier alias or user label.  
Move or associate a host port identifier to a different host.  
Replace a host port identifier with a new host port identifier.  
Manually activate an inactive host port so that the port can gain access to  
host specific or host group specific LUN mappings.  
Set the host port type to another type.  
Move a host from one host group to another host group.  
Remove a host group, a host, or a host port identifier.  
Rename a host group, or a host.  
Using the Mappings Tab  
In the mappings tab you can:  
Define hosts and hosts groups  
Add mappings to the selected host groups  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Defining a Host  
You can use the Define Host Wizard in the AMW to define a host for a  
storage array. Either a known unassociated host port identifier or a new host  
port identifier can be added.  
NOTE: A user label must be specified before the host port identifier may be added  
(the add button is disabled until one is entered). This user label cannot be the same  
as the host name. It is recommended that you use the host name with HBA port  
number. For example: Server1-HBA-0".  
88  
Configuration: About Your Host  
       
To define a host:  
1
2
In the AMW, select the Mappings tab and select the appropriate storage array.  
Perform one of the actions:  
Select MappingsDefineHost  
.
Select the Setup tab, and click Manually Define Hosts  
.
Select the Mappings tab. Right-click the root node (storage array  
name), Default Group node, or Host Group node in the Topology  
pane to which you want to add the host, and select DefineHost  
from the pop-up menu.  
The Specify Host Name window is displayed.  
3
4
In Host name, enter an up to 30-character alphanumeric name.  
Select the relevant option in Do you plan to use the storage partitions in  
the this storage array? and click Next  
.
The Specify Host Port Identifiers window is displayed.  
5
Select the relevant option to add a host port identifier to the host, you can  
select:  
Add by selecting a known unassociated host port identifier—In  
Known unassociated host port identifiers, select the relevant host  
port identifier.  
Add by creating a new host port identifier—In New host port  
identifier, enter a 16 character name and an up to 30 character Alias  
for the host port identifier and click Add  
NOTE: The host port identifier name must contain only the letters A through F.  
Click Next  
.
6
7
8
.
The Specify Host Type window is displayed.  
In Host type, select the relevant operating system for the host.  
The Host Group Question window is displayed.  
In this window, you can select:  
Yes—this host shares access to the same virtual disks with other hosts.  
No—this host does NOT share access to the same virtual disks with  
other hosts.  
Configuration: About Your Host  
89  
9
Click Next.  
If you select Yes, the Specify Host Group window is displayed. If you  
select No, see step 11  
10 Enter the name of the host group or select an existing host group and click  
Next  
The Preview window is displayed.  
.
11 Click Finish  
.
Removing Host Access  
1
In the AMW, select the Mappings tab, select the host node in the  
Topology pane.  
2
Perform one of these actions:  
Select MappingsRemove  
.
Right-click the host node, and select Remove from the pop-up menu.  
The Remove confirmation dialog appears.  
Type yes  
Click OK  
3
4
.
.
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Managing Host Groups  
A host group is a logical entity of two or more hosts that share access to  
specific virtual disks on the storage array. You create host groups with MDSM.  
All hosts in a host group must have the same host type (operating system). In  
addition, all hosts in the host group must have special software, such as  
clustering software, to manage virtual disk sharing and accessibility.  
If a host is part of a cluster, every host in the cluster must be connected to the  
storage array, and every host in the cluster must be added to the host group.  
90  
Configuration: About Your Host  
         
Creating a Host Group  
1
2
3
In the AMW, select the Mappings tab.  
In the Topology pane, select the storage array or the Default Group  
.
Perform one of the following actions:  
Select MappingsDefineHost Group  
Right-click the storage array or the Default Group, and select  
DefineHost Group from the pop-up menu.  
4
5
6
Type the name of the new host group in Enter new host group name.  
Select the appropriate hosts in the Select hosts to add area.  
Click Add  
.
NOTE: To remove hosts, select the hosts in the Hosts in group area, and click  
Remove.  
7
Click OK  
.
Adding a Host to a Host Group  
You can add a host to an existing host group or a new host group using the  
Define Host Wizard. For more information, see "Defining a Host" on  
page 88.  
You can also move a host to a different host group. For more information, see  
"Moving a Host to a Different Host Group" on page 92.  
Removing a Host From a Host Group  
You can remove a host from the Topology pane on the Mappings tab of the  
Array Management Window. For more information, see "Removing a Host  
Group" on page 92.  
Configuration: About Your Host  
91  
       
Moving a Host to a Different Host Group  
1
In the AMW, select the Mappings tab, select the host node in the  
Topology pane.  
2
Perform one of these actions:  
Select MappingsMove  
.
Right-click the host node, and select Move from the pop-up menu.  
The Move Host dialog appears.  
3
4
In the Select host group, select the host group to which you want to move  
the host.  
The Move Host Confirmation dialog appears.  
Click Yes  
.
The host is moved to the selected host group with the following mappings:  
The host retains the specific virtual disk mappings assigned to it.  
The host inherits the virtual disk mappings assigned to the host group  
to which it is moved.  
The host loses the virtual disk mappings assigned to the host group  
from which it was moved.  
Removing a Host Group  
1
In the AMW, select the Mappings tab, select the host node in the  
Topology pane.  
2
Perform one of these actions:  
Select MappingsRemove  
.
Right-click the host node, and select Remove from the pop-up menu.  
The Remove dialog is displayed.  
Click Yes  
The selected host group is removed.  
3
.
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
92  
Configuration: About Your Host  
       
Host Topology  
Host topology is the organization of hosts, host groups, and host interfaces  
configured for a storage array. You can view the host topology in the Mappings  
tab of the AMW. For more information, see "Using the Mappings Tab" on  
page 88.  
The following tasks change the host topology:  
Moving a host or a host connection  
Renaming a host group, a host, or a host connection  
Adding a host connection  
Replacing a host connection  
Changing a host type  
MDSM automatically detects these changes for any host running the host  
agent software.  
Starting or Stopping the Host Context Agent  
The host context agent discovers the host topology. The host context agent  
starts and stops with the host. The topology discovered by the host context  
agent can be viewed by clicking Configure Host Access (Automatic) in the  
Configure tab in the MDSM.  
You must stop and restart the host context agent to see the changes to the  
host topology if:  
A new storage array is attached to the host server.  
A host is added while turning on power to the RAID controller modules.  
Linux  
To start or stop the host context agent, enter the following command at  
the prompt:  
SMagent start  
SMagent stop  
You stop and then restart SMagent after:  
Moving a controller offline or replacing a controller.  
Removing host-to-array connections from or attaching host-to-array  
connections to a Linux host server.  
Configuration: About Your Host  
93  
       
Windows  
To start or stop the host context agent:  
1
Click StartSettingsControl PanelAdministrative Tools  
Services  
.
or  
Click StartAdministrative ToolsServices  
.
2
3
From the list of services, select Modular Disk Storage Manager Agent  
.
If the host context agent is running, click Action  
Stop, then wait  
approximately 5 seconds.  
4
Click Action  
Start.  
I/O Data Path Protection  
You can have multiple host-to-array connections for a host. Ensure that you  
select all the connections to the array when configuring host access to the  
storage array.  
NOTE: See the Deployment Guide for more information on cabling configurations,  
at support.dell.com\manuals.  
NOTE: For maximum redundancy, you must select all host connections to the array  
when manually defining host topology. For example, a host might have two host  
connections listed when manually configuring host access. For this host, you would  
select the two host connections listed in the Available hosts section and add them  
to the Selected hosts section using the Add button.  
NOTE: For more information on configuring hosts see "Configuration: About Your  
Host" on page 87.  
If a component such as a RAID controller module or a cable fails, or an error  
occurs on the data path to the preferred RAID controller module, virtual disk  
ownership is moved to the alternate non-preferred RAID controller module  
for processing. This process is called failover.  
Drivers for multi-path frameworks such as Microsoft Multi-Path IO (MPIO)  
and Linux Device Mapper (DM) are installed on host systems that access the  
storage array and provide I/O path failover.  
94  
Configuration: About Your Host  
   
For more information on Linux DM please see "Configuration: Device  
Mapper Multipath for Linux" on page 211. For more information on MPIO  
please see microsoft.com.  
NOTE: You should have the multi-path driver installed on the hosts at all times,  
even in a configuration where there is only one path to the storage system, such as  
a single port cluster configuration.  
During a failover, the virtual disk transfer is logged as a critical event, and an  
alert notification is sent automatically if you have configured alert  
destinations for the storage array.  
Managing Host Port Identifiers  
You can manage the host port identifiers that are added to the storage array.  
You can:  
Add—Add or associate a new host port identifier to a particular host.  
Edit—Change the host port identifier alias or user label. You can move  
(associate) the host port identifier to a new host.  
Replace—Replace a particular host port identifier with another host port  
identifier.  
Remove—Remove the association between a particular host port identifier  
and the associated host.  
NOTE: User label cannot be the same as the host name. It is recommended that  
you use the host name with HBA port number. For example: Server1-HBA-0".  
Configuration: About Your Host  
95  
   
To manage a host port identifier:  
1 Perform one of these actions:  
Right-click the host in the Topology pane, and select Manage Host  
Port Identifiers in the pop-up menu.  
From the menu bar, select Mappings  
Manage Host  
Port Identifiers  
.
The Manage Host Port Identifiers dialog is displayed. You can choose to  
manage the host port identifiers for a specific host or all of the host port  
identifiers for all of the hosts in Show host port identifiers associated with  
.
2
If you want to manage the host port identifiers for a specific host, select  
the host from the list of hosts that are associated with the storage array. If  
you want to manage the host port identifiers for all hosts, select All hosts  
from the list of hosts that are associated with the storage array.  
3
4
If you are adding a new host port identifier, go to step 4. If you are  
managing an existing host port identifier, go to step 8.  
Click Add  
.
The Add Host Port Identifier dialog appears.  
5
Select the method to add a host port identifier to the host. You can select:  
Add by selecting a know unassociated host port identifier—Select  
the appropriate host port identifier from the existing list of Known  
unassociated host port identifiers  
.
Add by creating a new host port identifier—In New host port  
identifier, enter the name of the new host port identifier.  
6
7
In User label, enter up to 30–character alphanumeric name.  
In Associated with host, select the appropriate host or host group.  
96  
Configuration: About Your Host  
 
8
9
Select the host port identifier that you would like to manage from the list  
of host port identifiers in the Host port identifier information area.  
Perform one of these actions for the selected host port identifier:  
To edit the host port identifier—Select the appropriate host port  
identifier and click Edit, the Edit Host Port Identifier dialog appears,  
update User label and Associated with host and click Save  
.
To replace the host port identifier—Select the appropriate host port  
identifier and click Replace, the Replace Host Port Identifier dialog  
appears, replace the current host port identifier with a known  
unassociated host port identifier or create a new host port identifier,  
update User label and click Replace  
.
To remove the host port identifier—Select the appropriate host port  
identifier and click Edit, the Remove Host Port Identifier dialog  
appears, type yes and click OK  
.
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Configuration: About Your Host  
97  
 
98  
Configuration: About Your Host  
9
Configuration: Disk Groups and  
Virtual Disks  
Creating Disk Groups and Virtual Disks  
Disk groups are created in the unconfigured capacity of a storage array, and  
virtual disks are created in the free capacity of a disk group. The maximum  
number of physical disks supported in a disk group is 30. The hosts attached  
to the storage array read and write data to the virtual disks.  
NOTE: Before you can create virtual disks, you must first organize the physical  
disks into disk groups and configure host access. Then you can create virtual disks  
within a disk group.  
To create a virtual disk, use one of the following methods:  
Create a new disk group from unconfigured capacity. You first define the RAID  
level and free capacity (available storage space) for the disk group, and then  
you define the parameters for the first virtual disk in the new disk group.  
Create a new virtual disk in the free capacity of an existing disk group. You  
only need to specify the parameters for the new virtual disk.  
A disk group has a set amount of free capacity that is configured when the  
disk group is created. You can use that free capacity to subdivide the disk  
group into one or more virtual disks.  
You can create disk groups and virtual disks using:  
Automatic configuration—Provides the fastest method, but with limited  
configuration options.  
Manual configuration—Provides more configuration options.  
When creating a virtual disk, consider the uses for that virtual disk, and select  
an appropriate capacity for those uses. For example, if a disk group has a  
virtual disk that stores multimedia files (which tend to be large) and another  
virtual disk that stores text files (which tend to be small), the multimedia file  
virtual disk requires more capacity than the text file virtual disk.  
Configuration: Disk Groups and Virtual Disks  
99  
     
A disk group should be organized according to its related tasks and subtasks.  
For example, if you create a disk group for the Accounting Department, you  
can create virtual disks that match the different types of accounting  
performed in the department: Accounts Receivable (AR), Accounts Payable  
(AP), internal billing, and so forth. In this scenario, the AR and AP virtual  
disks probably need more capacity than the internal billing virtual disk.  
NOTE: In Linux, the host must be rebooted after deleting virtual disks to reset the  
/dev entries.  
NOTE: Before you can use a virtual disk, you must register the disk with the host  
systems. See "Host-to-Virtual Disk Mapping" on page 124.  
Creating Disk Groups  
You can create disk groups using either Automatic configuration or Manual  
configuration.  
To create disk groups using automatic configuration:  
1
To start the Create Disk Group Wizard, perform one of these actions:  
To create a disk group from unconfigured capacity in the storage  
array—On the Logical tab, select an Unconfigured Capacity node,  
and select Disk GroupCreate. Alternatively, you can right-click the  
Unconfigured Capacity node, and select Create Disk Group from the  
pop-up menu.  
To create a disk group from unassigned physical disks in the storage  
array—On the Physical tab, select one or more unassigned physical  
disks of the same physical disk type, and select Disk GroupCreate  
Alternatively, you can right-click the unassigned physical disks, and  
select Create Disk Group from the pop-up menu.  
.
To create a secure disk group—On the Physical tab, select one or more  
unassigned security capable physical disks of the same physical disk  
type, and select Disk GroupCreate. Alternatively, you can right-  
click the unassigned security capable physical disks, and select Create  
Disk Group from the pop-up menu.  
The Introduction (Create Disk Group) window is displayed.  
Click Next  
2
.
The Disk Group Name and Physical Disk Selection window is displayed.  
100  
Configuration: Disk Groups and Virtual Disks  
   
3
4
Type up to a 30–character name of the disk group in Disk group name.  
Select the appropriate Physical Disk selection choices, you can select:  
Automatic, see step 6  
Manual, see step 9  
5
6
Click Next.  
For Automatic configuration, The RAID Level and Capacity window  
is displayed.  
7
Select the appropriate RAID level in Select RAID level. You can select  
RAID levels 0, 1/10, 6, and 5.  
Depending on your RAID level selection, the physical disks available for  
the selected RAID level is displayed in Select capacity table.  
8
9
In the Select Capacity table, select the relevant disk group capacity, and  
click Finish  
.
For Manual configuration, The Manual Physical Disk Selection window  
is displayed.  
10 Select the appropriate RAID level in Select RAID level. You can select  
RAID levels 0, 1/10, 6, and 5.  
Depending on your RAID level selection, the physical disks available for  
the selected RAID level is displayed in Unselected physical disks table.  
11 In the Unselected physical disks table, select the appropriate physical  
disks and click Add  
.
<Ctrl> or <Shift> and selecting additional physical disks.  
12 To view the capacity of the new disk group, click Calculate Capacity  
.
13 Click Finish  
.
A message prompts you that the disk group is successfully created and that  
you should create at least one virtual disk before you can use the capacity  
of the new disk group. For more information on creating virtual disks, see  
"Creating Virtual Disks" on page 102.  
Configuration: Disk Groups and Virtual Disks  
101  
   
Locating a Disk Group  
You can physically locate and identify all of the physical disks that comprise a  
selected disk group. An LED blinks on each physical disk in the disk group.  
To locate a disk group:  
1
2
In the AMW, select the Logical tab.  
Select the appropriate disk group and from the toolbar select Disk  
GroupBlink  
The LEDs for the selected disk group blink.  
3
4
After locating the disk group, click OK.  
The LEDs stop blinking.  
If the LEDs for the disk group do not stop blinking, from the toolbar in  
AMW, select Storage ArrayBlinkStop All Indications  
If the LEDs successfully stop blinking, a confirmation message appears.  
.
5
Click OK.  
Creating Virtual Disks  
Keep these important guidelines in mind when you create a virtual disk:  
Many hosts can have 256 logical unit numbers (LUNs) mapped per storage  
partition, but the number varies per operating system.  
After you create one or more virtual disks and assign a mapping, you must  
register the virtual disk with the operating system. In addition, you must  
ensure that the host recognizes the mapping between the physical storage  
array name and the virtual disk name. Depending on the operating system,  
run the host-based utilities, hot_add and SMdevices  
.
If the storage array contains physical disks with different media types or  
different interface types, multiple Unconfigured Capacity nodes may  
appear in the Logical pane of the Logical tab. Each physical disk type has  
an associated Unconfigured Capacity node if unassigned physical disks are  
available in the expansion enclosure.  
You cannot create a disk group and subsequent virtual disk from different  
physical disk technology types. Each physical disk that comprises the disk  
group must be of the same physical disk type.  
NOTE: Ensure that you create disk groups before creating virtual disks.  
102  
Configuration: Disk Groups and Virtual Disks  
       
To create virtual disks:  
1 Choose one of these methods to start the Create Virtual Disk Wizard  
:
To create a virtual disk from unconfigured capacity in the storage  
array—On the Logical tab, select an Unconfigured Capacity node,  
and select Virtual DiskCreate. Alternatively, you can right-click the  
Unconfigured Capacity node, and select Create Virtual Disk from  
the pop-up menu.  
To create a virtual disk from free capacity on a disk group—On the  
Logical tab, select a Free Capacity node, and select Virtual Disk  
Create. Alternatively, you can right-click the Free Capacity node, and  
select Create Virtual Disk from the pop-up menu.  
To create a virtual disk from unassigned physical disks in the storage  
array—On the Physical tab, select one or more unassigned physical  
disks of the same physical disk type, and select Virtual DiskCreate  
Alternatively, you can right-click the unassigned physical disks, and  
select Create Virtual Disk from the pop-up menu.  
.
To create a secure virtual disk—On the Physical tab, select one or  
more unassigned security capable physical disks of the same physical  
disk type, and select Virtual DiskCreate. Alternatively, you can  
right-click the unassigned security capable physical disks, and select  
Create Virtual Disk from the pop-up menu.  
If you selected an Unconfigured Capacity node or unassigned physical  
disks to create a virtual disk, the Disk Group Required dialog is displayed.  
Click Yes and create a disk group by using the Create Disk Group Wizard  
.
The Create Virtual Disk Wizard appears after you create the disk group.  
If you selected a Free Capacity node, the The Introduction (Create  
Virtual Disk) window is displayed.  
2
Click Next  
The Specify Capacity /Name window is displayed.  
Select the appropriate unit for memory in Units and enter the capacity of  
the virtual disk in New virtual disk capacity  
Enter an up to 30–character name for the virtual disk in Virtual disk name  
.
3
4
.
.
Configuration: Disk Groups and Virtual Disks  
103  
5
In Advanced virtual disk parameters, you can select:  
Use recommended settings  
.
Customize settings  
.
6
7
Click Next.  
In the Customize Advanced Virtual Disk Parameters window, select the  
appropriate Virtual Disk I/O characteristics type. You can select:  
File system (typical)  
Database  
Multimedia  
Custom  
NOTE: If you select Custom, you must select an appropriate segment size.  
8
9
Select the appropriate Preferred RAID controller module ownership and  
click Next  
.
The Specify Virtual Disk-to-LUN Mapping window is displayed.  
In the Specify Virtual Disk-to-LUN Mapping window, you can select:  
Default Mapping  
The appropriate Host type  
Map later Using Mappings View  
.
10 Click Finish  
.
The virtual disks are created.  
Changing the Virtual Disk Modification Priority  
You can specify the modification priority setting for a single virtual disk or  
multiple virtual disks on a storage array.  
Guidelines to change the modification priority of a virtual disk:  
If more than one virtual disk is selected, the modification priority defaults  
to the lowest priority. The current priority is shown only if a single virtual  
disk is selected.  
Changing the modification priority by using this option modifies the  
priority for the selected virtual disks.  
104  
Configuration: Disk Groups and Virtual Disks  
   
To change the virtual disk modification priority:  
1
2
3
In the AMW, select the Logical tab.  
Select a virtual disk.  
In the toolbar, select Virtual DiskChangeModification Priority  
.
The Change Modification Priority window is displayed.  
4
Select one or more virtual disks.  
To select nonadjacent virtual disks, press <Ctrl> click. To select adjacent  
virtual disks, press <Shift> click. To select all of the available virtual disks,  
click Select All. Move the Select modification priority slider bar to the  
desired priority.  
5
Click OK  
.
A message prompts you to confirm the change in the virtual disk  
modification priority.  
6
7
Click Yes  
.
Click OK  
.
Changing the Virtual Disk Cache Settings  
You can specify the cache memory settings for a single virtual disk or for  
multiple virtual disks on a storage array.  
Guidelines to change cache settings for a virtual disk:  
After opening the Change Cache Settings dialog, the system may prompt  
you that the RAID controller module has temporarily suspended caching  
operations. This action might occur when a new battery is charging, when  
a RAID controller module has been removed, or if a mismatch in cache  
sizes has been detected by the RAID controller module. After the  
condition has cleared, the cache properties selected in the dialog become  
active. If the selected cache properties do not become active, contact your  
Technical Support representative.  
If you select more than one virtual disk, the cache settings default to no  
settings selected. The current cache settings appear only if you select a  
single virtual disk.  
If you change the cache settings by using this option, the priority of all of  
the virtual disks that you selected is modified.  
Configuration: Disk Groups and Virtual Disks  
105  
   
To change the virtual disk cache settings:  
1
2
In the AMW, select the Logical tab and select a virtual disk  
In the toolbar, select Virtual DiskChangeCache Settings  
The Change Cache Settings window is displayed.  
.
3
Select one or more virtual disks.  
To select nonadjacent virtual disks, press <Ctrl> click. To select adjacent  
virtual disks, press <Shift> click. To select all of the available virtual disks,  
click Select All  
.
4
In the Select cache properties area, you can select:  
Enable read caching—to enable read caching.  
Enable dynamic cache read prefetch—to enable dynamic cache  
read prefetch.  
Enable write caching—to enable write caching.  
Enable write caching with mirroring—to mirror cached data  
across two redundant RAID controller modules that have the  
same cache size.  
Enable write caching without batteries—to permit write caching  
to continue even if the RAID controller module batteries are  
discharged completely, not fully charged, or are not present.  
CAUTION: Possible loss of data – Selecting the Enable write caching without  
batteries option lets write caching continue even when the batteries are discharged  
completely or are not fully charged. Typically, write caching is turned off temporarily  
by the RAID controller module until the batteries are charged. If you select this  
option and do not have a universal power supply for protection, you could lose data.  
In addition, you could lose data if you do not have RAID controller module batteries  
and you select the Enable write caching without batteries option.  
NOTE: When the Optional RAID controller module batteries option is enabled,  
the Enable write caching does not appear. The Enable write caching without  
batteries is still available, but it is not checked by default.  
NOTE: Cache is automatically flushed after the Enable write caching check  
box is disabled.  
106  
Configuration: Disk Groups and Virtual Disks  
5
Click OK.  
A message prompts you to confirm the change in the virtual disk  
modification priority.  
6
7
Click Yes  
.
Click OK  
.
Changing the Segment Size of a Virtual Disk  
You can change the segment size on a selected virtual disk. During this  
operation, I/O performance is affected, but your data remains available.  
Guidelines to proceed with changing the segment size:  
You cannot cancel this operation after it starts.  
Do not start this operation unless the disk group is in Optimal status.  
MDSM determines the segment size transitions that are allowed. Segment  
sizes that are inappropriate transitions from the current segment size are  
unavailable on the menu. Allowed transitions usually are double or half of  
current segment size. For example, if the current virtual disk segment size is  
32 KB, a new virtual disk segment size of either 16 KB or 64 KB is allowed.  
NOTE: The operation to change the segment size is slower than other modification  
operations (for example, changing RAID levels or adding free capacity to a disk  
group). This slowness is the result of how the data is reorganized and the temporary  
internal backup procedures that occur during the operation.  
The amount of time that a change segment size operation takes depends on:  
The I/O load from the host  
The modification priority of the virtual disk  
The number of physical disks in the disk group  
The number of physical disk ports  
The processing power of the storage array RAID controller modules  
If you want this operation to complete faster, you can change the  
modification priority, although this might decrease system I/O performance.  
Configuration: Disk Groups and Virtual Disks  
107  
   
To change the segment size of a virtual disk:  
1
2
3
In the AMW, select the Logical tab and select a virtual disk.  
Select Virtual DiskChangeSegment Size  
.
Select the required segment size.  
A message prompts you to confirm the selected segment size.  
Click Yes  
4
.
The segment size modification operation begins. The virtual disk icon in  
the Logical pane shows an Operation in Progress status while the  
operation is taking place.  
NOTE: To view the progress or change the priority of the modification operation,  
select a virtual disk in the disk group, and select Virtual DiskChange  
Modification Priority.  
Changing the I/O Type  
You can specify the virtual disk I/O characteristics for the virtual disks that  
you are defining as part of the storage array configuration. The expected I/O  
characteristics of the virtual disk is used by the system to indicate an  
applicable default virtual disk segment size and dynamic cache read prefetch  
setting.  
NOTE: The dynamic cache read prefetch setting can be changed later by selecting  
Virtual DiskChangeCache Settings. You can change the segment size later by  
selecting Virtual DiskChangeSegment Size.  
The I/O characteristic types shown below are only presented during the create  
virtual disk process.  
When you choose one of the virtual disk I/O characteristics, the  
corresponding dynamic cache prefetch setting and segment size that are  
typically well suited for expected I/O patterns are populated in the Dynamic  
cache read prefetch field and the Segment size field.  
108  
Configuration: Disk Groups and Virtual Disks  
   
To change the I/O type:  
1
Select from these virtual disk I/O characteristic types, based on your  
application needs:  
File system (typical)  
Database  
Multimedia  
Custom  
The corresponding dynamic cache read prefetch setting and segment size  
values that are typically well suited for the selected virtual disk I/O  
characteristic type are populated in the Dynamic cache read prefetch and  
Segment size fields.  
NOTE: If you selected the Custom option, select your preferred dynamic  
cache read prefetch setting (enabled/disabled) and segment size (8 KB to  
512 KB).  
2
Click OK.  
Choosing an Appropriate Physical Disk Type  
You can create disk groups and virtual disks in the storage array. You must  
select the capacity that you want to allocate for the virtual disk from either  
unconfigured capacity or free capacity available in the storage array. Then you  
define basic and optional advanced parameters for the virtual disk.  
With the advent of different physical disk technologies, it is now possible to  
mix physical disks with different media types and different interface types  
within a single storage array. In this release of MDSM, the following media  
types are supported:  
Hard physical disk  
Solid State Disk (SSD)  
Configuration: Disk Groups and Virtual Disks  
109  
   
Physical Disk Security With Self Encrypting Disk  
Self encrypting disk (SED) technology prevents unauthorized access to the  
data on a physical disk that is physically removed from the storage array. The  
storage array has a security key. Self encrypting disks provide access to data  
only through an array that has the correct security key.  
The self encrypting disk or a security capable physical disk encrypts data  
during writes and decrypts data during reads. For more information, see the  
PowerVault Modular Disk Storage Manager online help topics.  
You can create a secure disk group from security capable physical disks. When  
you create a secure disk group from security capable physical disks, the  
physical disks in that disk group become security enabled. When a security  
capable physical disk has been security enabled, the physical disk requires the  
correct security key from a RAID controller module to read or write the data.  
All of the physical disks and RAID controller modules in a storage array share  
the same security key. The shared security key provides read and write access  
to the physical disks, while the physical disk encryption key on each physical  
disk is used to encrypt the data. A security capable physical disk works like any  
other physical disk until it is security enabled.  
Whenever the power is turned off and turned on again, all of the security-enabled  
physical disks change to a security locked state. In this state, the data is  
inaccessible until the correct security key is provided by a RAID  
controller module.  
You can view the self encrypting disk status of any physical disk in the storage  
array from the Physical Disk Properties dialog. The status information  
reports whether the physical disk is:  
Security Capable  
Secure—Security enabled or disabled  
Read/Write Accessible—Security locked or unlocked  
You can view the self encrypting disk status of any disk group in the storage  
array. The status information reports whether the storage array is:  
Security Capable  
Secure  
110  
Configuration: Disk Groups and Virtual Disks  
   
The following table shows how to interpret the security status of a disk group:  
Table 9-1. Interpretation of Security Status of Disk Group  
Secure  
Security Capable–Yes  
Security Capable–No  
Yes  
The disk group is composed of all Not applicable. Only SED  
SED physical disks and is in a  
Secure state.  
physical disks can be in a Secure  
state.  
No  
The disk group is composed of all The disk group is not entirely  
SED physical disks and is in a  
Non-Secure state.  
composed of SED physical disks.  
The Physical Disk Security menu appears in the Storage Array menu. The  
Physical Disk Security menu has these options:  
Create Security Key  
Change Security Key  
Save Security Key File  
Validate Security Key  
Unlock Drives  
NOTE: If you have not created a security key for the storage array, the Create  
Security Key option is active. If you have created a security key for the storage  
array, the Create Security Key option is inactive with a check mark to the left. The  
Change Security Key option, the Save Security Key option, and the Validate  
Security Key option are now active.  
The Secure Physical Disks option appears in the Disk Group menu. The  
Secure Physical Disks option is active if these conditions are true:  
The selected storage array is not security enabled but is comprised entirely  
of security capable physical disks.  
The storage array contains no snapshot base virtual disks or snapshot  
repository virtual disks.  
The disk group is in an Optimal state.  
A security key is set up for the storage array.  
NOTE: The Secure Physical Disks option is inactive if these conditions are not true.  
Configuration: Disk Groups and Virtual Disks  
111  
The Secure Physical Disks option is inactive with a check mark to the left if  
the disk group is already security enabled.  
The Create a secure disk group option appears in the Create Disk Group  
Wizard - Disk Group Name and Physical Disk Selection dialog. The Create  
a secure disk group option is active only when these conditions are met:  
A security key is installed in the storage array.  
At least one security capable physical disk is installed in the storage array.  
All of the physical disks that you selected on the Physical tab are security  
capable physical disks.  
You can erase security-enabled physical disks so that you can reuse the drives  
in another disk group or in another storage array. When you erase security-  
enabled physical disks, ensure that the data cannot be read. When all of the  
physical disks that you have selected in the Physical pane are security  
enabled, and none of the selected physical disks is part of a disk group, the  
Secure Erase option appears in the Physical Disk menu.  
The storage array password protects a storage array from potentially  
destructive operations by unauthorized users. The storage array password is  
independent from self encrypting disk, and should not be confused with the  
pass phrase that is used to protect copies of a security key. However, it is good  
practice to set a storage array password.  
Creating a Security Key  
When you create a security key, it is generated by and securely stored by the  
array. You cannot read or view the security key.  
NOTE: A copy of the security key must be kept on some other storage medium for  
backup in case of system failure or for transfer to another storage array.  
A pass phrase that you provide is used to encrypt and decrypt the security key  
for storage on other media. When you create a security key, you also provide  
information to create a security key identifier. Unlike the security key, you can  
read or view the security key identifier. The security key identifier is also  
stored on a physical disk or transportable media. The security key identifier is  
used to identify which key the storage array is using.  
112  
Configuration: Disk Groups and Virtual Disks  
   
To create a security key:  
1
In the AMW toolbar, select Storage ArrayPhysical Disk Security  
Create Security Key  
.
The Create Security Key window is displayed.  
2
In Security key identifier, enter a string that becomes part of the secure  
key identifier.  
You can enter up to 189 alphanumeric characters without spaces,  
punctuation, or symbols. Additional characters is generated automatically  
and is appended to the end of the string that you enter. The generated  
characters help to ensure that the secure key identifier is unique.  
3
4
Edit the default path by adding a file name to the end of the path or Click  
Browse, navigate to the required folder and enter the name of the file.  
In Pass phrase, enter a string for the pass phrase.  
The pass phrase must:  
be between eight and 32–characters long.  
contain at least one uppercase letter.  
contain at least one lowercase letter.  
contain at least one number.  
contain at least one non-alphanumeric character, for example, < > @ +.  
The pass phrase that you enter is masked.  
NOTE: Create Key is active only if the pass phrase meets the above  
mentioned criterion.  
5
In Confirm pass phrase, re-enter the exact string that you entered in  
Pass phrase  
.
Make a record of the pass phrase that you entered and the security key  
identifier that is associated with the pass phrase. You need this  
information for later secure operations.  
6
7
Click Create Key  
Make a record of the security key identifier and the file name from the  
Create Security Key Complete dialog, and click OK  
.
.
Configuration: Disk Groups and Virtual Disks  
113  
After you have created a security key, you can create secure disk groups from  
security capable physical disks. Creating a secure disk group makes the  
physical disks in the disk group security-enabled. Security-enabled physical  
disks enter Security Locked status whenever power is re-applied. They can be  
unlocked only by a RAID controller module that supplies the correct key  
during physical disk initialization. Otherwise, the physical disks remain  
locked, and the data is inaccessible. The Security Locked status prevents any  
unauthorized person from accessing data on a security-enabled physical disk  
by physically removing the physical disk and installing the physical disk in  
another computer or storage array.  
Changing a Security Key  
When you change a security key, a new security key is generated by the  
system. The new key replaces the previous key. You cannot view or read the  
key. However, a copy of the security key must be kept on some other storage  
medium for backup in case of system failure or for transfer to another storage  
array. A pass phrase that you provide encrypts and decrypts the security key  
for storage on other media.  
When you change a security key, you also provide information to create a  
security key identifier.  
Changing the security key does not destroy any data. You can change the  
security key at any time.  
Before you change the security key, ensure that:  
All virtual disks in the storage array are in Optimal status.  
In storage arrays with two RAID controller modules, both are present and  
working normally.  
To change the security key:  
1
2
In the AMW toolbar, select Storage ArrayPhysical Disk Security  
Change Security Key  
The Confirm Change Security Key window is displayed.  
.
Type yes in the text field, and click OK  
.
The Change Security Key window is displayed.  
114  
Configuration: Disk Groups and Virtual Disks  
   
3
In Secure key identifier, enter a string that becomes part of the secure  
key identifier.  
You may leave the text box blank, or enter up to 189 alphanumeric  
characters without white space, punctuation, or symbols. Additional  
characters are generated automatically.  
4
5
Edit the default path by adding a file name to the end of the path or Click  
Browse, navigate to the required folder and enter the name of the file.  
In Pass phrase, enter a string for the pass phrase.  
The pass phrase must meet the following criteria:  
It must be between eight and 32 characters long.  
It must contain at least one uppercase letter.  
It must contain at least one lowercase letter.  
It must contain at least one number.  
It must contain at least one non-alphanumeric character (for example,  
< > @ +).  
The pass phrase that you enter is masked.  
6
In Confirm pass phrase, re-enter the exact string you entered in Pass phrase.  
Make a record of the pass phrase you entered and the security key identifier it  
is associated with. You need this information for later secure operations.  
7
8
Click Change Key.  
Make a record of the security key identifier and the file name from the  
Change Security Key Complete dialog, and click OK  
.
Saving a Security Key  
You save an externally storable copy of the security key when the security key  
is first created and each time it is changed. You can create additional storable  
copies at any time. To save a new copy of the security key, you must provide a  
pass phrase. The pass phrase you choose does not need to match the pass  
phrase used when the security key was created or last changed. The pass  
phrase is applied to the particular copy of the security key you are saving.  
Configuration: Disk Groups and Virtual Disks  
115  
   
To save the security key for the storage array:  
1
In the AMW toolbar, select Storage ArrayPhysical Disk SecuritySave  
Security Key File  
.
The Save Security Key File - Enter Pass Phrase window is displayed.  
2
3
Edit the default path by adding a file name to the end of the path or Click  
Browse, navigate to the required folder and enter the name of the file.  
In Pass phrase, enter a string for the pass phrase.  
The pass phrase must meet the following criteria:  
It must be between eight and 32 characters long.  
It must contain at least one uppercase letter.  
It must contain at least one lowercase letter.  
It must contain at least one number.  
It must contain at least one non-alphanumeric character (for example,  
< > @ +).  
The pass phrase that you enter is masked.  
4
In Confirm pass phrase, re-enter the exact string you entered in Pass phrase.  
Make a record of the pass phrase you entered. You need it for later  
secure operations.  
5
6
Click Save  
.
Make a record of the security key identifier and the file name from the  
Save Security Key Complete dialog, and click OK  
.
116  
Configuration: Disk Groups and Virtual Disks  
Unlocking Secure Physical Disks  
You can export a security-enabled disk group to move the associated physical  
disks to a different storage array. After you install those physical disks in the  
new storage array, you must unlock the physical disks before data can be read  
from or written to the physical disks. To unlock the physical disks, you must  
supply the security key from the original storage array. The security key on the  
new storage array is different and cannot unlock the physical disks.  
You must supply the security key from a security key file that was saved on the  
original storage array. You must provide the pass phrase that was used to  
encrypt the security key file to extract the security key from this file.  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Erasing Secure Physical Disks  
In the AMW, when you select a security enabled physical disk that is not part  
of a disk group, the Secure Erase menu item is enabled on the Physical Disk  
menu. You can use the secure erase procedure to re-provision a physical disk.  
You can use the Secure Erase option if you want to remove all of the data on  
the physical disk and reset the physical disk security attributes.  
CAUTION: Possible loss of data access – The Secure Erase option removes all of  
the data that is currently on the physical disk. This action cannot be undone.  
Before you complete this option, ensure that the physical disk that you have  
selected is the correct physical disk. You cannot recover any of the data that is  
currently on the physical disk.  
After you complete the secure erase procedure, the physical disk is available  
for use in another disk group or in another storage array. See the PowerVault  
Modular Disk Storage Manager online help topics for more information on the  
secure erase procedure.  
Configuration: Disk Groups and Virtual Disks  
117  
       
Configuring Hot Spare Physical Disks  
Guidelines to configure host spare physical disks:  
You can use only unassigned physical disks with Optimal status as hot  
spare physical disks.  
You can unassign only hot spare physical disks with Optimal, or Standby  
status. You cannot unassign a hot spare physical disk that has the In Use  
status. A hot spare physical disk has the In Use status when it is in the  
process of taking over for a failed physical disk.  
If a hot spare physical disk does not have Optimal status, follow the  
Recovery Guru procedures displayed by the MDSM application to correct  
any problem before trying to unassign the physical disk.  
Hot spare physical disks must be of the same media type and interface type  
as the physical disks that they are protecting.  
If there are secure disk groups and security capable disk groups in the  
storage array, the hot spare physical disk must match the security  
Hot spare physical disks must have capacities equal to or larger than the  
used capacity on the physical disks that they are protecting.  
The availability of enclosure loss protection for a disk group depends on  
the location of the physical disks that comprise the disk group. To ensure  
that enclosure loss protection is not affected, you must replace a failed  
physical disk to initiate the copyback process. See "Enclosure Loss  
Protection" on page 121.  
CAUTION: If a hot spare physical disk does not have Optimal status, follow the  
Recovery Guru procedures to correct the problem before you try to unassign the  
physical disk. You cannot assign a hot spare physical disk if it is in use (taking  
over for a failed physical disk).  
To assign or unassign hot spare physical disks:  
1
2
3
In the AMW, select the Physical tab.  
Select one or more physical disks.  
Perform one of these actions:  
Select Physical diskHot Spare Coverage.  
118  
Configuration: Disk Groups and Virtual Disks  
   
Right-click the physical disk and select Hot Spare Coverage from the  
pop-up menu.  
The Hot Spare Physical Disk Options window is displayed.  
4
Select the appropriate option, you can select:  
View/change current hot spare coverage—to review hot spare coverage  
and to assign or unassign hot spare physical disks, if necessary. See step 5.  
Automatically assign physical disks— to create hot spare physical  
disks automatically for the best hot spare coverage using available  
physical disks.  
Manually assign individual physical disks—to create hot spare  
physical disks out of the selected physical disks on the Physical tab.  
Manually unassign individual physical disks—to unassign the  
selected hot spare physical disks on the Physical tab. See step 12.  
5
To assign hot spares, in the Hot Spare Coverage window, select a disk  
group in the Hot spare coverage area.  
6
7
Review the information about the hot spare coverage in the Details area.  
Click Assign  
.
The Assign Hot Spare window is displayed.  
8
9
Select the relevant Physical disks in the Unassigned physical disks area, as  
hot spares for the selected disk and click OK  
.
To unassign hot spares, in the Hot Spare Coverage window, select physical  
disks in the Hot spare physical disks area.  
10 Review the information about the hot spare coverage in the Details area.  
11 Click Unassign  
A message prompts you to confirm the operation.  
12 Type yes and click OK  
.
.
Configuration: Disk Groups and Virtual Disks  
119  
   
Hot Spares and Rebuild  
A valuable strategy to protect data is to assign available physical disks in the  
storage array as hot spares. A hot spare adds another level of fault tolerance to  
the storage array.  
A hot spare is an idle, powered-on, stand-by physical disk ready for immediate  
use in case of disk failure. If a hot spare is defined in an enclosure in which a  
redundant virtual disk experiences a physical disk failure, a rebuild of the  
degraded virtual disk is automatically initiated by the RAID controller  
modules. If no hot spares are defined, the rebuild process is initiated by the  
RAID controller modules when a replacement physical disk is inserted into  
the storage array.  
Global Hot Spares  
The MD3600f Series supports global hot spares. A global hot spare can  
replace a failed physical disk in any virtual disk with a redundant RAID level  
as long as the capacity of the hot spare is equal to or larger than the size of the  
configured capacity on the physical disk it replaces, including its metadata.  
Hot Spare Operation  
When a physical disk fails, the virtual disk automatically rebuilds using an  
available hot spare. When a replacement physical disk is installed, data from  
the hot spare is copied back to the replacement physical disk. This function is  
called copy back. By default, the RAID controller module automatically  
configures the number and type of hot spares based on the number and  
capacity of physical disks in your system.  
A hot spare may have the following states:  
A standby hot spare is a physical disk that has been assigned as a hot spare  
and is available to take over for any failed physical disk.  
An in-use hot spare is a physical disk that has been assigned as a hot spare  
and is currently replacing a failed physical disk.  
120  
Configuration: Disk Groups and Virtual Disks  
           
Hot Spare Drive Protection  
You can use a hot spare physical disk for additional data protection from  
physical disk failures that occur in a RAID Level 1, or RAID Level 5 disk  
group. If the hot spare physical disk is available when a physical disk fails, the  
RAID controller module uses redundancy data to reconstruct the data from  
the failed physical disk to the hot spare physical disk. When you have  
physically replaced the failed physical disk, a copyback operation occurs from  
the hot spare physical disk to the replaced physical disk.  
If there are secure disk groups and security capable disk groups in the storage  
array, the hot spare physical disk must match the security capability of the  
disk group. For example, a non-security capable physical disk cannot be used  
as a hot spare for a secure disk group.  
NOTE: For a security capable disk group, security capable hot spare physical disks  
are preferred. If security capable physical disks are not available, non-security  
capable physical disks may be used as hot spare physical disks. To ensure that the  
disk group is retained as security capable, the non-security capable hot spare  
physical disk must be replaced with a security capable physical disk.  
If you select a security capable physical disk as hot spare for a non-secure disk  
group, a dialog box appears indicating that a security capable physical disk is  
being used as a hot spare for a non-secure disk group.  
The availability of enclosure loss protection for a disk group depends on the  
location of the physical disks that comprise the disk group. The enclosure loss  
protection might be lost because of a failed physical disk and location of the  
hot spare physical disk. To ensure that enclosure loss protection is not  
affected, you must replace a failed physical disk to initiate the copyback  
process.  
The virtual disk remains online and accessible while you are replacing the  
failed physical disk, because the hot spare physical disk is automatically  
substituted for the failed physical disk.  
Enclosure Loss Protection  
Enclosure loss protection is an attribute of a disk group. Enclosure loss  
protection guarantees accessibility to the data on the virtual disks in a disk  
group if a total loss of communication occurs with a single expansion  
enclosure. An example of total loss of communication might be loss of power  
to the expansion enclosure or failure of both RAID controller modules.  
Configuration: Disk Groups and Virtual Disks  
121  
       
CAUTION: Enclosure loss protection is not guaranteed if a physical disk has  
already failed in the disk group. In this situation, losing access to an expansion  
enclosure and consequently another physical disk in the disk group causes a  
double physical disk failure and loss of data.  
Enclosure loss protection is achieved when you create a disk group where all  
of the physical disks that comprise the disk group are located in different  
expansion enclosures. This distinction depends on the RAID level. If you  
choose to create a disk group by using the Automatic method, the software  
attempts to choose physical disks that provide enclosure loss protection. If  
you choose to create a disk group by using the Manual method, you must use  
the criteria specified in Table 9-2.  
Table 9-2. Criteria for Enclosure Loss Protection  
RAID Level  
Criteria for Enclosure Loss Protection  
RAID level 5 Ensure that all the physical disks in the disk group are located in  
or RAID level different expansion enclosures.  
6
Because a RAID level 5 requires a minimum of three physical disks,  
enclosure loss protections cannot be achieved if your storage array has  
less than three expansion enclosures. Because a RAID level 6requires a  
minimum of five physical disks, enclosure loss protections cannot be  
achieved if your storage array has less than five expansion enclosures.  
122  
Configuration: Disk Groups and Virtual Disks  
 
Table 9-2. Criteria for Enclosure Loss Protection (continued)  
RAID Level Criteria for Enclosure Loss Protection  
RAID level 1 Ensure that each physical disk in a mirrored pair is located in a  
different expansion enclosure. This enables you to have more than  
two physical disks in the disk group within the same expansion  
enclosure.  
For example, if you are creating a six physical disk, disk group (three-  
mirrored pairs), you could achieve enclosure loss protection with only  
two expansion enclosures by specifying that the physical disk in each  
mirrored pair are located in separate expansion enclosures. This  
example shows this concept:  
• Mirror pair 1—Physical disk in enclosure 1 slot 1 and physical disk in  
enclosure 2 slot 1.  
• Mirror pair 2—Physical disk in enclosure 1 slot 2and physical disk in  
enclosure 2 slot 2.  
• Mirror pair 3—Physical disk in enclosure 1 slot 3and physical disk in  
enclosure 2 slot 3.  
Because a RAID level 1 disk group requires a minimum of two physical  
disks, enclosure loss protections cannot be achieved if your storage  
array has less than two expansion enclosures.  
RAID level 0 RAID level 0 does not have consistency, you cannot achieve enclosure  
loss protection.  
Configuration: Disk Groups and Virtual Disks  
123  
Host-to-Virtual Disk Mapping  
After you create virtual disks, you must map them to the host(s) connected to  
the array.  
Guidelines to configure host-to-virtual disk mapping:  
Each virtual disk in the storage array can be mapped to only one host or  
host group.  
Host-to-virtual disk mappings are shared between controllers in the  
storage array.  
A unique LUN must be used by a host group or host to access a virtual disk.  
Each host has its own LUN address space. MDSM permits the same LUN  
to be used by different hosts or host groups to access virtual disks in a  
storage array.  
Not every operating system has the same number of LUNs available.  
You can define the mappings on the Mappings tab in the Array  
Management Window. See "Using the Mappings Tab" on page 88.  
Creating Host-to-Virtual Disk Mappings  
Guidelines to define the mappings:  
An access virtual disk mapping is not required for an out-of-band storage  
array. If your storage array is managed using an out-of-band connection,  
and an access virtual disk mapping is assigned to the Default Group, an  
access virtual disk mapping is assigned to every host created from the  
Default Group. To prevent this action from occurring, remove the access  
Most hosts have 256 LUNs mapped per storage partition. The LUN  
numbering is from 0 through 255. If your operating system restricts LUNs  
to 127, and you try to map a virtual disk to a LUN that is greater than or  
equal to 127, the host cannot access it.  
An initial mapping of the host group or host must be created using the  
Storage Partitioning Wizard before defining additional mappings. See  
"Storage Partitioning" on page 134.  
124  
Configuration: Disk Groups and Virtual Disks  
       
To create host to virtual disk mappings:  
1
2
In the AMW, select the Mappings tab.  
In the Topology pane, select:  
Default Group  
Undefined mappings node  
Individual defined mapping  
Host group  
Host  
3
4
In the toolbar, select MappingsDefineAdditional Mapping.  
The Define Additional Mapping window is displayed.  
In Host group or host, select the appropriate host group or host.  
All defined hosts, host groups, and the default group appear in the list.  
5
6
In Logical unit number, select a LUN. The supported LUNs are 0  
through 255.  
Select the virtual disk to be mapped in the Virtual Disk area.  
The Virtual Disk area lists the names and capacity of the virtual disks that are  
available for mapping based on the selected host group or selected host.  
7
8
9
Click Add.  
NOTE: The Add button is inactive until a host group or host, LUN, and virtual  
disk are selected.  
To define additional mappings, repeat step 4 through step 7.  
NOTE: After a virtual disk has been mapped once, it is no longer available in  
the Virtual Disk area.  
Click Close  
.
The mappings are saved. The Topology pane and the Defined Mappings  
pane in the Mappings tab are updated to reflect the mappings.  
Configuration: Disk Groups and Virtual Disks  
125  
   
Modifying and Removing Host-to-Virtual Disk Mapping  
You can modify or remove a host-to-virtual disk mapping for several reasons,  
such as an incorrect mapping or reconfiguration of the storage array.  
Modifying or removing a host-to-virtual disk mapping applies to both hosts  
and host groups.  
To modify or remove host to virtual disk mapping:  
NOTE: Before you modify or remove a host-to-virtual disk mapping, you must stop  
any data access (I/O) to the virtual disks to prevent data loss.  
1
2
In the AMW, select the Mappings tab.  
In the Defined Mappings pane, perform one of these actions:  
Select a single virtual disk, and select MappingsChange  
Mapping.  
 
Right-click the virtual disk, and select Change Mapping from the  
pop-up menu.  
3
4
5
6
In Host group or host, select the appropriate host group or host.  
By default, the drop-down list shows the current host group or the host  
associated with the selected virtual disk.  
In Logical unit number, select the appropriate LUN.  
The drop down list shows only the currently available LUNs that are  
associated with the selected virtual disk.  
Click OK  
.
Stop any host applications associated with this virtual disk, and unmount  
the virtual disk, if applicable, from your operating system.  
In the Change Mapping dialog, click Yes to confirm the changes.  
The mapping is checked for validity and is saved. The Defined Mappings  
pane is updated to reflect the new mapping. The Topology pane is also  
updated to reflect any movement of host groups or hosts.  
NOTE: If a password is set on the storage array, the Enter Password dialog  
appears. Type the current password for the storage array, and click OK.  
126  
Configuration: Disk Groups and Virtual Disks  
   
7
8
If configuring a Linux host, run the rescan_dm_devs utility on the host,  
and remount the virtual disk if required. This utility is installed on the host  
as part of the MDSM install process.  
Restart the host applications.  
Changing Controller Ownership of the Virtual Disk  
If the host has a single data-path to the MD storage array, the virtual disk  
must be owned by the controller to which the host is connected. You must  
configure this storage array before you start I/O operations and after the  
virtual disk is created.  
You can change the RAID controller module ownership of a standard virtual  
disk or a snapshot repository virtual disk. You cannot directly change the  
RAID controller module ownership of a snapshot virtual disk because the  
snapshot virtual disk inherits the RAID controller module owner of its  
associated source virtual disk. Changing the RAID controller module  
ownership of a virtual disk changes the preferred RAID controller module  
ownership of the virtual disk.  
During a virtual disk copy, the same RAID controller module must own both  
the source virtual disk and the target virtual disk. Sometimes both virtual  
disks do not have the same preferred RAID controller module when the  
virtual disk copy starts. Therefore, the ownership of the target virtual disk is  
automatically transferred to the preferred RAID controller module of the  
source virtual disk. When the virtual disk copy is completed or is stopped,  
ownership of the target virtual disk is restored to its preferred RAID controller  
module. If ownership of the source virtual disk is changed during the virtual  
disk copy, ownership of the target virtual disk is also changed. Under certain  
operating system environments, it might be necessary to reconfigure the  
multi-path driver before an I/O path can be used.  
To change the ownership of the virtual disk to the connected controller:  
1
2
3
In the AMW, select the Logical tab and select a virtual disk.  
Select Virtual DiskChangeOwnership/Preferred Path  
.
Select the appropriate RAID controller module slot and click Yes to  
confirm the selection.  
Configuration: Disk Groups and Virtual Disks  
127  
   
Removing Host-to-Virtual Disk Mapping  
1
2
3
In the AMW, select the Mapping tab.  
Select a virtual disk from the Defined Mappings pane.  
Perform one of these actions:  
Select MappingsRemove.  
Right-click the virtual disk, and select Remove Mapping from the  
pop-up menu.  
4
Click Yes to remove the mapping.  
Changing the RAID Controller Module Ownership of a Disk Group  
You can change the RAID controller module ownership of a a disk group.  
You can change the RAID controller module ownership of a standard virtual  
disk or a snapshot repository virtual disk. You cannot directly change the  
RAID controller module ownership of a snapshot virtual disk because the  
snapshot virtual disk inherits the RAID controller module owner of its  
associated source virtual disk. Changing the RAID controller module  
ownership of a virtual disk changes the preferred RAID controller module  
ownership of the virtual disk.  
During a virtual disk copy, the same RAID controller module must own both  
the source virtual disk and the target virtual disk. Sometimes both virtual  
disks do not have the same preferred RAID controller module when the  
virtual disk copy starts. Therefore, the ownership of the target virtual disk is  
automatically transferred to the preferred RAID controller module of the  
source virtual disk. When the virtual disk copy is completed or is stopped,  
ownership of the target virtual disk is restored to its preferred RAID controller  
module. If ownership of the source virtual disk is changed during the virtual  
disk copy, ownership of the target virtual disk is also changed. Under certain  
operating system environments, it might be necessary to reconfigure the  
multi-path driver before an I/O path can be used.  
128  
Configuration: Disk Groups and Virtual Disks  
       
To change the RAID controller module ownership of a disk group:  
1
2
3
In the AMW, select the Logical tab and select a disk group.  
Select Disk GroupChangeOwnership/Preferred Path  
.
Select the appropriate RAID controller module slot and click Yes to  
confirm the selection.  
CAUTION: Possible loss of data access– Changing ownership at the disk group  
level causes every virtual disk in that disk group to transfer to the other RAID  
controller module and use the new I/O path. If you do not want to set every virtual  
disk to the new path, change ownership at the virtual disk level instead.  
The ownership of the disk group is changed. I/O to the disk group is now  
directed through this I/O path.  
NOTE: The disk group might not use the new I/O path until the multi-path driver  
reconfigures and recognizes the new path. This action usually takes less than  
five minutes.  
Changing the RAID Level of a Disk Group  
Changing the RAID level of a disk group changes the RAID levels of every  
virtual disk that comprises the disk group. Performance might be slightly  
affected during the operation.  
Guidelines to change the RAID level of a disk group:  
You cannot cancel this operation after it begins.  
The disk group must be in Optimal status before you can perform  
this operation.  
Your data remains available during this operation.  
If you do not have enough capacity in the disk group to convert to the new  
RAID level, an error message appears, and the operation does not  
continue. If you have unassigned physical disks, use the Disk Group  
Add Free Capacity (Physical Disks) option to add additional capacity to  
the disk group. Then retry the operation.  
To change the RAID level of a disk group:  
1
2
3
In the AMW, select the Logical tab and select a disk group.  
Select Disk GroupChangeRAID Level  
Select the appropriate RAID level and click Yes to confirm the selection.  
.
The RAID level operation begins.  
Configuration: Disk Groups and Virtual Disks  
129  
   
Restricted Mappings  
Many hosts are able to map up to 256 logical unit numbers (LUNs) (0 to 255) per  
storage partition. However, the maximum number of mappings differs because of  
operating system variables, failover driver issues, and potential data problems.  
The hosts listed in the Table 9-3 have these mapping restrictions.  
If you try to map a virtual disk to a LUN that exceeds the restriction on these  
operating systems, the host is unable to access the virtual disk.  
Table 9-3. Mapping Restrictions  
Operating System  
Highest LUN  
255  
Windows Server 2003 and Windows server 2008  
Linux  
255  
Guidelines when you work with host types with LUN mapping restrictions:  
You cannot change a host adapter port to a restricted host type if there are  
already mappings in the storage partition that would exceed the limit  
imposed by the restricted host type.  
Consider the case of the Default Group that has access to LUNs up to 256  
(0 to 255) and a restricted host type is added to the Default Group. In this  
case, the host that is associated with the restricted host type is able to  
access virtual disks in the Default Group with LUNs within its limits. For  
example, if the Default Group had two virtual disks mapped to LUNs 254  
and 255, the host with the restricted host type would not be able to access  
those two virtual disks.  
If the Default Group has a restricted host type assigned and the storage  
partitions are disabled, you can map only a total of 32 LUNs. Any  
additional virtual disks that are created are put in the Unidentified  
Mappings area. If additional mappings are defined for one of these  
Unidentified Mappings, the Define Additional Mapping dialog shows  
the LUN list, and the Add button is unavailable.  
Do not configure dual mappings on a Windows host.  
130  
Configuration: Disk Groups and Virtual Disks  
     
If there is a host with a restricted host type that is part of a specific storage  
partition, all of the hosts in that storage partition are limited to the  
maximum number of LUNs allowed by the restricted host type.  
You cannot move a host with a restricted host type into a storage partition  
that already has LUNs mapped that are greater than what is allowed by the  
restricted host type. For example, if you have a restricted host type that  
allows only LUNs up to 31, you cannot move that restricted host type into  
a storage partition that has LUNs greater than 31 already mapped.  
The Default Group on the Mappings tab has a default host type. You can  
change this type by selecting Storage ArrayChangeDefault Host Type.  
If you set the default host type to a host type that is restricted, the maximum  
number of LUNs that are allowed in the Default Group for any host are  
restricted to the limit imposed by the restricted host type. If a particular host  
with a non-restricted host type becomes part of a specific storage partition,  
you are able to change the mapping to a higher LUN.  
Changing the RAID Controller Module Ownership of a Virtual Disk or a  
Disk Group  
You can change the RAID controller module ownership of a virtual disk or a  
disk group.  
You can change the RAID controller module ownership of a standard virtual  
disk or a snapshot repository virtual disk. You cannot directly change the  
RAID controller module ownership of a snapshot virtual disk because the  
snapshot virtual disk inherits the RAID controller module owner of its  
associated source virtual disk. Changing the RAID controller module  
ownership of a virtual disk changes the preferred RAID controller module  
ownership of the virtual disk.  
During a virtual disk copy, the same RAID controller module must own both  
the source virtual disk and the target virtual disk. Sometimes both virtual  
disks do not have the same preferred RAID controller module when the  
virtual disk copy starts. Therefore, the ownership of the target virtual disk is  
automatically transferred to the preferred RAID controller module of the  
source virtual disk. When the virtual disk copy is completed or is stopped,  
ownership of the target virtual disk is restored to its preferred RAID controller  
module. If ownership of the source virtual disk is changed during the virtual  
Configuration: Disk Groups and Virtual Disks  
131  
   
disk copy, ownership of the target virtual disk is also changed. Under certain  
operating system environments, it might be necessary to reconfigure the  
multi-path driver before an I/O path can be used.  
To change the RAID controller module ownership of a virtual disk or a disk  
group?  
Change the RAID controller module ownership of a virtual disk  
Change the RAID controller module ownership of a disk group  
To change the RAID controller module ownership of a virtual disk, perform  
these steps:  
1
2
3
Select the Logical tab.  
Select the virtual disk.  
Select Virtual DiskChangeOwnership/Preferred Path. Or, you can  
right-click the virtual disk and select ChangeOwnership/Preferred Path  
from the pop-up menu.  
4
5
Select the RAID controller module.  
CAUTION: Possible loss of data access – If you do not use a multi-path driver,  
shut down any host applications that are currently using the virtual disk. This  
action prevents application errors when the I/O path changes.  
Click Yes  
.
The ownership of the virtual disk is changed. I/O to the virtual disk is now  
directed through this I/O path.  
NOTE: The virtual disk might not use the new I/O path until the multi-path  
driver reconfigures and recognizes the new path. This action usually takes less  
than five minutes.  
To change the RAID controller module ownership of a disk group:  
1
2
3
Select the Logical tab.  
Select the disk group.  
Select Disk GroupOwnership/Preferred Path  
.
or  
4
5
Right-click the disk group and select ChangeOwnership/Preferred Path  
from the pop-up menu.  
Select the RAID controller module.  
132  
Configuration: Disk Groups and Virtual Disks  
CAUTION: Possible loss of data access– Changing ownership at the disk group  
level causes every virtual disk in that disk group to transfer to the other RAID  
controller module and use the new I/O path. If you do not want to set every virtual  
disk to the new path, change ownership at the virtual disk level instead.  
6
Click Yes.  
The ownership of the disk group is changed. I/O to the disk group is now  
directed through this I/O path.  
NOTE: The disk group might not use the new I/O path until the multi-path driver  
reconfigures and recognizes the new path. This action usually takes less than  
five minutes.  
Changing the RAID Level of a Disk Group  
Use the ChangeRAID Level option to change the RAID level on a selected  
disk group. Using this option changes the RAID levels of every virtual disk  
that comprises the disk group. Performance might be slightly affected during  
the operation.  
Keep these guidelines in mind when you change the RAID level of a disk group:  
You cannot cancel this operation after it begins.  
The disk group must be in Optimal status before you can perform  
this operation.  
Your data remains available during this operation.  
If you do not have enough capacity in the disk group to convert to the new  
RAID level, an error message appears, and the operation does not  
continue. If you have unassigned physical disks, use the Disk Group  
Add Free Capacity (Physical Disks) option to add additional capacity to  
the disk group. Then retry the operation.  
To change the RAID level of a disk group:  
1
2
3
Select the Logical tab.  
Select the disk group.  
Select Disk GroupChangeRAID Level  
.
Configuration: Disk Groups and Virtual Disks  
133  
     
4
5
Select the RAID level (RAID Level 0, RAID Level 1, RAID Level 5, or  
RAID Level 6). The currently selected option is designated with a dot.  
Click Yes  
.
The RAID level operation begins.  
Storage Partitioning  
A storage partition is a logical entity consisting of one or more virtual disks  
that can be accessed by a single host or shared among hosts that are part of a  
host group. The first time you map a virtual disk to a specific host or host  
group, a storage partition is created. Subsequent virtual disk mappings to that  
host or host group do not create another storage partition.  
One storage partition is sufficient if:  
Only one attached host accesses all of the virtual disks in the storage array.  
All attached hosts share access to all of the virtual disks in the storage array.  
When you choose this type of configuration, all of the hosts must have the  
same operating system and special software (such as clustering software) to  
manage virtual disk sharing and accessibility.  
More than one storage partition is required if:  
Specific hosts must access specific virtual disks in the storage array.  
Hosts with different operating systems are attached to the same storage  
array. In this case, a storage partition is created for each host type.  
You can use the Storage Partitioning wizard to define a single storage partition.  
The Storage Partitioning wizard guides you through the major steps required to  
specify which host groups, hosts, virtual disks, and associated logical unit  
numbers (LUNs) are to be included in the storage partition.  
Storage partitioning fails when:  
All mappings are defined.  
You create a mapping for a host group that conflicts with an established  
mapping for a host in the host group.  
You create a mapping for a host in a host group that conflicts with an  
established mapping for the host group.  
134  
Configuration: Disk Groups and Virtual Disks  
   
Storage partitioning is unavailable when:  
No valid host groups or hosts exist in the Topology pane on the Mappings  
tab.  
No host ports are defined for the host being included in the storage  
partition.  
All mappings are defined.  
NOTE: You can include a secondary virtual disk in a storage partition.  
However, any hosts that are mapped to the secondary virtual disk has read-only  
access until the virtual disk is promoted to a primary virtual disk, or the mirror  
relationship is removed.  
Storage partitioning topology is the collection of elements, such as Default  
Group, host groups, hosts, and host ports shown as nodes in the Topology  
pane of the Mappings tab in the AMW. For more information, see "Using the  
Mappings Tab" on page 88.  
If a storage partitioning topology is not defined, an informational dialog  
appears each time you select the Mappings tab. You must define the storage  
partitioning topology before you define the actual storage partition.  
Disk Group and Virtual Disk Expansion  
Adding free capacity to a disk group is achieved by adding unconfigured  
capacity on the array to the disk group. Data is accessible on disk groups,  
virtual disks, and physical disks throughout the entire modification operation.  
The additional free capacity can then be used to perform a virtual disk  
expansion on a standard or snapshot repository virtual disk.  
Disk Group Expansion  
1
2
3
In the AMW, select the Logical tab.  
Select a disk group.  
Select Disk GroupAdd Free Capacity (Physical Disks)  
.
The Add Free Capacity window is displayed. Based on the RAID level, and  
the enclosure loss protection of the current disk group, a list of unassigned  
physical disks appears.  
Configuration: Disk Groups and Virtual Disks  
135  
       
NOTE: If the RAID level of the disk group is RAID Level 5, or RAID Level 6, and  
the expansion enclosure has enclosure loss protection, Display only physical disks  
that will ensure enclosure loss protection appears and is selected by default.  
4
in the Available physical disks area, select physical disks up to the allowed  
maximum number of physical disks.  
NOTE: You cannot mix different media types or different interface types  
within a single disk group or virtual disk.  
5
6
Click Add  
A message prompts you to confirm your selection.  
To add the capacity to the disk group, click Yes  
.
.
You can also use the Command Line Interface (CLI) on both Windows and  
Linux hosts to add free capacity to a disk group. See the CLI Guide for more  
information.  
After the capacity expansion is completed, additional free capacity is available  
in the disk group for creation of new virtual disks or expansion of existing  
virtual disks.  
Virtual Disk Expansion  
Virtual disk expansion is a dynamic modification operation that increases the  
capacity of standard virtual disks.  
NOTE: Snapshot repository virtual disks can be expanded from the CLI or from  
MDSM. All other virtual disk types are expandable only from the CLI.  
If you receive a warning that the snapshot repository virtual disk is becoming  
full, you may expand the snapshot repository virtual disk from MDSM. See  
"Snapshot Repository Capacity" on page 154 for step-by-step instructions.  
Using Free Capacity  
You can increase the capacity of a virtual disk using the free capacity on the  
disk group of the standard virtual disk or the snapshot repository virtual disk.  
The Free Capacity node, shown in the Logical pane, is a contiguous region of  
unassigned capacity on a defined disk group. When increasing virtual disk  
capacity, some or all of the free capacity might be used to achieve the required  
final capacity. Data on the selected virtual disk remains accessible while the  
process for increasing virtual disk capacity is in progress.  
136  
Configuration: Disk Groups and Virtual Disks  
       
Using Unconfigured Capacity  
You can increase the capacity of a standard virtual disk or a snapshot  
repository virtual disk using the unconfigured capacity when no free capacity  
exists on a disk group. An increase is achieved by adding unconfigured  
capacity, in the form of unassigned physical disks, to the disk group of the  
standard virtual disk or the snapshot repository virtual disk. See "Disk Group  
Expansion" on page 135.  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Disk Group Migration  
Disk group migration allows to you export a disk group so that you can import  
the disk group to a different storage array. You can also export a disk group so  
that you can store the data offline.  
NOTE: During the export process (before the disk group is imported) you lose  
access to the data on the exported disk group.  
NOTE: You must export a disk group before you move the disk group or import the  
disk group.  
Export Disk Group  
The export disk group operation prepares the physical disks in the disk group  
for removal. You can remove the physical disks for offline storage, or you can  
import the disk group to a different storage array. After you complete the  
export disk group operation, all of the physical disks are offline. Any  
associated virtual disks or free capacity nodes are no longer shown in MDSM.  
Non-Exportable Components  
You must remove or clear any non-exportable settings before you can  
complete the export disk group procedure. Remove or clear the following  
settings:  
Persistent reservations  
Host-to-virtual disk mappings  
Virtual disk copy pairs  
Snapshot virtual disks and snapshot repository virtual disks  
Remote replication pairs  
Replication repositories  
Configuration: Disk Groups and Virtual Disks  
137  
             
Exporting a Disk Group  
On the source storage array:  
1
2
Save the storage array configuration.  
Stop all I/O, and unmount or disconnect the file systems on the virtual  
disks in the disk group.  
3
4
5
6
Back up the data on the virtual disks in the disk group.  
Locate the disk group, and label the physical disks.  
Place the disk group offline.  
Obtain blank physical disk modules or new physical disks.  
On the target storage array:  
1
2
3
4
Verify that the target storage array has available physical disk slots.  
Verify that the target storage array supports the physical disks that you import.  
Verify that the target storage array can support the new virtual disks.  
Verify that the latest version of firmware is installed on the RAID  
controller module.  
Import Disk Group  
The import disk group operation adds the imported disk group to the target  
storage array. After you complete the import disk group operation, all of the  
physical disks have Optimal status. Any associated virtual disks or free  
capacity nodes are now shown in MDSM installed on the target storage array.  
NOTE: You lose access to your data during the export/import process.  
NOTE: You must export a disk group before you move the disk group or import the  
disk group.  
Importing a Disk Group  
NOTE: You must insert all of the physical disks that are part of the disk group into  
the enclosure before the disk group can be imported.  
On the target storage array:  
1
Insert the exported physical disks into the available physical disk slots.  
138  
Configuration: Disk Groups and Virtual Disks  
         
2
Review the Import Report for an overview of the disk group that you are  
importing.  
3
4
Check for non-importable components.  
Confirm that you want to proceed with the import procedure.  
NOTE: Some settings cannot be imported during the import disk group  
procedure.  
The following settings are removed/cleared during the procedure:  
Persistent reservations  
Host-to-virtual disk mappings  
Virtual disk copy pairs  
Snapshot virtual disks and snapshot repository virtual disks  
Remote replication pairs  
Replication repositories  
Non-Importable Components  
Some components cannot be imported during the import disk group  
procedure. These components are removed during the procedure:  
Persistent reservations  
Host-to-virtual disk mappings  
Virtual disk copy pairs  
Snapshot virtual disks and snapshot repository virtual disks  
Remote replication pairs  
Replication repositories  
Storage Array Media Scan  
The media scan is a background operation that examines virtual disks to  
verify that data is accessible. The process finds media errors before normal  
read and write activity is disrupted and reports errors to the event log.  
NOTE: You cannot enable background media scans on a virtual disk comprised of  
Solid State Disks (SSDs).  
Configuration: Disk Groups and Virtual Disks  
139  
   
Errors discovered by the media scan include:  
Unrecovered media error—Data could not be read on the first attempt or  
on any subsequent attempts. For virtual disks with redundancy protection,  
data is reconstructed, rewritten to the physical disk, and verified and the  
error is reported to the event log. For virtual disks without redundancy  
protection (RAID 1, RAID 5, and RAID 6 virtual disks), the error is not  
corrected but is reported to the event log.  
Recovered media error—Data could not be read by the physical disk on the  
first attempt but was successfully read on a subsequent attempt. Data is  
rewritten to the physical disk and verified and the error is reported to the  
event log.  
Redundancy mismatches error—The first 10 redundancy mismatches that  
are found on the virtual disk are reported to the event log.  
Unfixable error—Data could not be read and parity or redundancy  
information could not be used to regenerate the data. For example,  
redundancy information cannot be used to reconstruct the data on a  
degraded virtual disk. The error is reported to the event log.  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Changing Media Scan Settings  
1
2
In the AMW, select the Logical tab and select any virtual disk.  
Select Virtual DiskChangeMedia Scan Settings  
.
The Change Media Scan Settings window is displayed.  
3
4
Deselect Suspend media scan if selected.  
In Scan duration, enter or select the duration (in days) for the media scan.  
The media scan duration specifies the number of days for which the media  
scan runs on the selected virtual disks.  
5
6
To disable media scans on an individual virtual disk, select the virtual disk  
in the Select virtual disks to scan area, and deselect Scan selected virtual  
disks  
.
To enable media scans on an individual virtual disk, select the virtual disk  
in the Select virtual disks to scan area, and select Scan selected virtual  
disks  
.
140  
Configuration: Disk Groups and Virtual Disks  
   
7
8
To enable or disable the consistency check, select either With consistency  
check or Without consistency check  
.
NOTE: A consistency check scans the data blocks in a RAID Level 5 virtual  
disk, or a RAID Level 6 virtual disk and checks the consistency information for  
each block. A consistency check compares data blocks on RAID Level 1  
replicated physical disks. RAID Level 0 virtual disks have no data consistency.  
Click OK  
.
Suspending the Media Scan  
You cannot perform a media scan while performing another long-running  
operation on the disk drive such as reconstruction, copy-back,  
reconfiguration, virtual disk initialization, or immediate availability  
formatting. If you want to perform another long-running operation, you  
should suspend the media scan.  
NOTE: A background media scan is the lowest priority of the long-running  
operations.  
To suspend a media scan:  
1
2
In the AMW, select the Logical tab and select any virtual disk.  
Select Virtual DiskChangeMedia Scan Settings  
The Change Media Scan Settings window is displayed.  
Select Suspend media scan  
NOTE: This applies to all the virtual disks on the disk group.  
Click OK  
.
3
4
.
.
Configuration: Disk Groups and Virtual Disks  
141  
   
142  
Configuration: Disk Groups and Virtual Disks  
10  
Configuration: Premium Feature—  
Snapshot Virtual Disks  
NOTE: If you ordered this feature, you received a Premium Feature Activation card  
shipped in the same box as your Dell PowerVault MD storage array. Follow the  
directions on the card to obtain a key file and to enable the feature.  
NOTE: The snapshot feature allows up to eight snapshots per LUN and 128 per  
array to be present at the same time.  
A snapshot virtual disk is a point-in-time image of a virtual disk in a storage  
array. It is not an actual virtual disk containing a copy of the original data; It is  
a reference to the data that was contained on a virtual disk at a specific time.  
A snapshot virtual disk is the logical equivalent of a complete physical copy.  
However, you can create a snapshot virtual disk much faster than a physical  
copy, using less disk space.  
The virtual disk on which the snapshot is based, called the source virtual disk,  
must be a standard virtual disk in your storage array. Typically, you create a  
snapshot so that an application, such as a backup application, can access the  
snapshot and read the data while the source virtual disk remains online and  
accessible.  
NOTE: No I/O requests are permitted on the source virtual disk while the virtual  
disk snapshot is being created.  
A snapshot repository virtual disk containing metadata and copy-on-write  
data is automatically created when a snapshot virtual disk is created. The only  
data stored in the snapshot repository virtual disk is that which has changed  
since the time of the snapshot.  
Configuration: Premium Feature—Snapshot Virtual Disks  
143  
 
After the snapshot repository virtual disk is created, I/O write requests to the  
source virtual disk resume. Before a data block on the source virtual disk is  
modified, the contents of the block to be modified are copied to the snapshot  
repository virtual disk for safekeeping. Because the snapshot repository virtual  
disk stores copies of the original data in those data blocks, further changes to  
those data blocks write only to the source virtual disk. The snapshot  
repository uses less disk space than a full physical copy, because the only data  
blocks that are stored in the snapshot repository virtual disk are those that  
have changed since the time of the snapshot.  
When you create a snapshot virtual disk, you specify its location, capacity,  
and other parameters. You can disable or delete the snapshot virtual disk  
when it is not required. If you disable a snapshot virtual disk, you can re-  
create and reuse it the next time you perform a backup. For more  
information, see "Re-creating Snapshot Virtual Disks" on page 157. If you  
delete a snapshot virtual disk, you also delete the associated snapshot  
repository virtual disk.  
NOTE: Deleting a snapshot does not affect data on the source virtual disk.  
NOTE: The following host preparation sections also apply when using the snapshot  
feature through the CLI interface.  
Creating a Snapshot Virtual Disk Using the  
Simple Path  
You can choose the simple path to create a snapshot virtual disk if the disk  
group of the source virtual disk has the required amount of free space. A  
snapshot repository virtual disk requires a minimum 8 MB of free capacity.  
The destination of a snapshot repository virtual disk is determined based on  
the free capacity available in the disk group.  
If 8 MB of free capacity is not available in the disk group of the source virtual  
disk, the Create Snapshot Virtual Disks feature defaults to the advanced path.  
For more information, see "Creating a Snapshot Virtual Disk Using the  
Advanced Path" on page 148.  
In the advanced path option, you can choose to place the snapshot repository  
virtual disk in another disk group or you can use unconfigured capacity on the  
storage array to create a new disk group.  
144  
Configuration: Premium Feature—Snapshot Virtual Disks  
   
About the Simple Path  
Using the simple path, you can specify:  
Snapshot Virtual Disk Name—A user-specified name that helps you  
associate the snapshot virtual disk to its corresponding snapshot repository  
virtual disk and source virtual disk.  
Snapshot Repository Virtual Disk Name—A user-specified name that  
helps you associate the snapshot repository virtual disk to its corresponding  
snapshot virtual disk and source virtual disk.  
Snapshot Repository Virtual Disk Capacity—The snapshot repository  
virtual disk capacity is expressed as a percentage of the source virtual disk  
capacity. The maximum percentage allowed is 120 percent.  
Using the simple path, the following defaults are used for the other  
parameters of a snapshot virtual disk:  
Capacity Allocation—The snapshot repository virtual disk is created using  
free capacity on the same disk group where the source virtual disk resides.  
Host-to-Virtual Disk Mapping—The default setting is Map now  
.
Percent Full—When the snapshot repository virtual disk reaches the  
specified repository full percentage level, the event is logged in the Major  
Event Log (MEL). The default snapshot repository full percentage level is  
50 percent of the source virtual disk.  
Snapshot Repository Virtual Disk Full Conditions—When the snapshot  
repository virtual disk becomes full, you are given a choice of failing write  
activity to the source virtual disk or failing the snapshot virtual disk.  
Preparing Host Servers to Create the Snapshot Using the Simple Path  
NOTE: Before using the Snapshot Virtual Disks Premium Feature in a Microsoft  
Windows clustered configuration, you must first map the snapshot virtual disk to  
the cluster node that owns the source virtual disk. This ensures that the cluster  
nodes correctly recognize the snapshot virtual disk.  
NOTE: Mapping the snapshot virtual disk to the node that does not own the source  
virtual disk before the Snapshot enabling process is completed can result in the  
operating system mis-identifying the snapshot virtual disk. This, in turn, can result in  
data loss on the source virtual disk or an inaccessible snapshot.  
Configuration: Premium Feature—Snapshot Virtual Disks  
145  
         
NOTE: For details on mapping the snapshot virtual disk to the secondary node, see  
the Dell PowerVault MD3200 and MD3220 Storage Arrays With Microsoft Windows  
Server Failover Clusters on support.dell.com/manuals.  
NOTE: You can create concurrent snapshots of a source virtual disk on both the  
source disk group and on another disk group.  
Before creating a Snapshot Virtual Disk, note that:  
The following types of virtual disks are not valid source virtual disks:  
Snapshot repository virtual disks  
Snapshot virtual disks  
Replication repository virtual disks  
Target virtual disks that are participating in a virtual disk copy  
NOTE: Virtual Disk Copy is an Advanced (Premium) feature.  
You cannot create a snapshot of a virtual disk that contains unreadable  
sectors.  
You must satisfy the requirements of your host operating system for  
creating snapshot virtual disks. Failure to meet the requirements of your  
host operating system results in an inaccurate snapshot of the source  
virtual disk or the target virtual disk in a virtual disk copy.  
NOTE: Before you create a new snapshot of a source virtual disk, stop any  
data access (I/O) activity or suspend data transfer to the source virtual disk to  
ensure that you capture an accurate snapshot of the source virtual disk. Close  
all applications, including Windows Internet Explorer, to ensure all I/O activity  
has stopped.  
NOTE: Removing the drive letter of the associated virtual disk(s) in Windows  
or unmounting the virtual drive in Linux helps to guarantee a stable copy of the  
drive for the Snapshot.  
Before creating a snapshot virtual disk, the host server has to be in the proper  
state. To ensure that the host server is properly prepared to create a snapshot  
virtual disk, you can either use an application to carry out this task, or you can  
perform the following steps:  
1
2
Stop all I/O activity to the source.  
In the AMW, select the Logical tab and select a valid source virtual disk.  
146  
Configuration: Premium Feature—Snapshot Virtual Disks  
3
Select Virtual DiskSnapshotCreate.  
or  
Right-click the source virtual disk and select Create Snapshot Virtual Disk  
from the pop-up menu.  
The Create Snapshot Virtual Disk Wizard - Introduction dialog appears.  
4
5
Select Simple (Recommended), and click Next.  
The Specify Names window is displayed.  
Enter the Snapshot virtual disk name and the Snapshot repository virtual  
disk name and click Next.  
The Specify Snapshot Repository Capacity window is displayed.  
6
Enter the snapshot repository virtual disks capacity as a percentage of the  
source virtual disks capacity and click Next  
.
The Preview window containing the summary of the snapshot virtual disk  
is displayed.  
7
8
Click Finish  
The Completed window is displayed.  
Click OK  
.
.
After creating one or more snapshot virtual disks, mount the source virtual  
disk, and restart the host application using that source virtual disk.  
9
In the AMW, select the Mappings tab, assign mappings between the  
snapshot virtual disk and the host that accesses the snapshot virtual disk.  
NOTE: In some cases, conflicts might result from mapping the same host to  
both a source virtual disk and its associated snapshot virtual disk. This  
conflict depends on the host operating system and any virtual disk manager  
software in use.  
10 To register the snapshot virtual disk with the host operating system, run  
the host-based hot_add utility.  
11 To associate the mapping between the storage array name and the virtual  
disk name, run the host-based SMdevices utility.  
NOTE: If your operating system requires additional instructions, you can find  
those instructions in your operating system documentation.  
Configuration: Premium Feature—Snapshot Virtual Disks  
147  
Creating a Snapshot Virtual Disk Using the  
Advanced Path  
About the Advanced Path  
Use the advanced path to choose whether to place the snapshot repository  
virtual disk on free capacity or unconfigured capacity and to change the  
snapshot repository virtual disk parameters. You can select the advanced path  
regardless of whether you use free capacity or unconfigured capacity for the  
snapshot virtual disk.  
Using the advanced path, you can specify the following parameters for your  
snapshot virtual disk:  
Snapshot Virtual Disk Name—A user-specified name that helps you  
associate the snapshot virtual disk to its corresponding snapshot repository  
virtual disk and source virtual disk.  
Snapshot Repository Virtual Disk Name—A user-specified name that  
helps you associate the snapshot repository virtual disk to its corresponding  
snapshot virtual disk and source virtual disk.  
Capacity Allocation—This parameter allows you to choose where to create  
the snapshot repository virtual disk. You can allocate capacity by using one  
of the following methods:  
Use free capacity on the same disk group where the source virtual  
disk resides.  
Use free capacity on another disk group.  
Use unconfigured capacity and create a new disk group for the  
snapshot repository virtual disk.  
Dell recommends placing the snapshot repository virtual disk within  
the disk group of the source virtual disk. This ensures that if drives  
associated with the disk group are moved to another storage array, all  
the virtual disks associated with the snapshot virtual disk remain in  
the same group.  
Snapshot Repository Virtual Disk Capacity—The snapshot repository  
virtual disk capacity is expressed as a percentage of the source virtual disk  
capacity. The maximum percentage allowed is 120 percent.  
148  
Configuration: Premium Feature—Snapshot Virtual Disks  
     
Percent Full—When the snapshot repository virtual disk reaches the user-  
specified repository full percentage level, the event is logged in the Major  
Event Log (MEL). The default snapshot repository full percentage level is  
50% of the source virtual disk.  
Snapshot Repository Virtual Disk Full Conditions—Choose whether to  
fail writes to the source virtual disk or fail the snapshot virtual disk when  
the snapshot repository virtual disk becomes full.  
Host-to-Virtual Disk Mapping—Choose whether to map the snapshot  
virtual disk to a host or host group now or to map the snapshot virtual disk  
later. The default setting is Map later.  
Preparing Host Servers to Create the Snapshot Using the Advanced Path  
NOTE: Before using the Snapshot Virtual Disks Premium Feature in a Microsoft  
Windows clustered configuration, you must first map the snapshot virtual disk to  
the cluster node that owns the source virtual disk. This ensures that the cluster  
nodes correctly recognize the snapshot virtual disk.  
NOTE: Mapping the snapshot virtual disk to the node that does not own the source  
virtual disk before the Snapshot enabling process is completed can result in the  
operating system mis-identifying the snapshot virtual disk. This, in turn, can result in  
data loss on the source virtual disk or an inaccessible snapshot.  
NOTE: For details on mapping the snapshot virtual disk to the secondary node, see  
the Dell PowerVault MD3200 and MD3220 Storage Arrays With Microsoft Windows  
Server Failover Clusters on support.dell.com/manuals.  
The destination of a snapshot repository virtual disk is determined based on  
the free capacity available in the disk group. A snapshot repository virtual disk  
requires a minimum 8 MB of free capacity. You can choose your preferred  
creation path—simple or advanced—if the disk group of the source virtual  
disk has the required amount of free space.  
If 8 MB of free capacity is not available in the disk group of the source virtual  
disk, the Create Snapshot Virtual Disks feature defaults to the advanced  
path see "Creating a Snapshot Virtual Disk Using the Advanced Path" on  
page 148. In the advanced path option, you can choose to place the snapshot  
repository virtual disk in another disk group or you can use unconfigured  
capacity on the storage array to create a new disk group.  
NOTE: You can create concurrent snapshots of a source virtual disk on both the  
source disk group and on another disk group.  
Configuration: Premium Feature—Snapshot Virtual Disks  
149  
 
Remember the following guidelines before creating a Snapshot Virtual Disk:  
The following types of virtual disks are not valid source virtual disks:  
Snapshot repository virtual disks.  
Snapshot virtual disks.  
Replication repository virtual disks.  
Target virtual disks that are participating in a virtual disk copy.  
NOTE: Virtual Disk Copy is an Advanced (Premium) feature.  
You cannot create a snapshot of a virtual disk that contains  
unreadable sectors.  
You must satisfy the requirements of your host operating system for  
creating snapshot virtual disks. Failure to meet the requirements of your  
host operating system results in an inaccurate snapshot of the source  
virtual disk or the target virtual disk in a virtual disk copy.  
NOTE: Before you create a new snapshot of a source virtual disk, stop any data  
access (I/O) activity or suspend data transfer to the source virtual disk to ensure that  
you capture an accurate snapshot of the source virtual disk. Close all applications,  
including Windows Internet Explorer, to ensure all I/O activity has stopped.  
NOTE: Removing the drive letter of the associated virtual disk(s) in Windows or  
unmounting the virtual drive in Linux helps to guarantee a stable copy of the drive  
for the Snapshot.  
Before creating a snapshot virtual disk, the host server must be in the proper  
state. To prepare your host server:  
1
2
Stop all I/O activity to the source.  
Using your Windows system, flush the cache to the source. At the host  
prompt, type SMrepassist -f <filename-identifier>and  
press <Enter>. For more information, see "SMrepassist Utility" on  
page 281.  
150  
Configuration: Premium Feature—Snapshot Virtual Disks  
3
4
Remove the drive letter(s) of the source in Windows or unmount the  
virtual drive(s) in Linux to help guarantee a stable copy of the drive for the  
Snapshot. If this is not done, the snapshot operation reports that it has  
completed successfully, but the snapshot data is not updated properly.  
NOTE: Verify that the virtual disk has a status of Optimal or Disabled by  
Follow any additional instructions for your operating system. Failure to  
follow these additional instructions can create unusable snapshot virtual  
disks.  
NOTE: If your operating system requires additional instructions, you can find those  
instructions in your operating system documentation.  
After your host server has been prepared, see "Creating a Snapshot Virtual  
Disk Using the Advanced Path" on page 148 to create the snapshot using the  
advanced path.  
If you want to use a snapshot regularly, such as for backups, use the Disable  
Snapshot and Re-create Snapshot options to reuse the snapshot. Disabling  
and re-creating snapshots preserves the existing virtual disk-to-host mappings  
to the snapshot virtual disk.  
Creating the Snapshot Using the Advanced Path  
NOTE: Removing the drive letter of the associated virtual disk in Windows or  
unmounting the virtual drive in Linux helps to guarantee a stable copy of the drive  
for the Snapshot.  
After first preparing the host server(s) as specified in the preceding procedure,  
complete the following steps to create a virtual disk snapshot using the  
advanced path:  
1
Stop the host application accessing the source virtual disk, and unmount  
the source virtual disk.  
2
3
In the AMW, select the Logical tab, select a valid source virtual disk.  
Select Virtual Disk  
SnapshotCreate.  
or  
4
Right-click the source virtual disk and select Create Snapshot Virtual Disk  
from the pop-up menu.  
The Create Snapshot Virtual Disk Wizard - Introduction dialog appears.  
Configuration: Premium Feature—Snapshot Virtual Disks  
151  
 
5
6
Select Advanced, and click Next.  
The Specify Names window is displayed.  
Enter the Snapshot visual disk name and the Snapshot repository virtual  
disk name and click Next.  
The Allocate Capacity window is displayed.  
7
In the Capacity allocation area, select:  
Free capacity on same disk group as base (recommended)  
Free capacity on different disk group  
Unconfigured capacity (create new disk group)  
8
9
Enter the snapshot repository virtual disks capacity as a percentage of the  
source virtual disks capacity and click Next  
.
The Specify Virtual Disk Parameters window is displayed.  
In the Snapshot virtual disk parameters area, select the relevant mapping  
option, you can select:  
Automatic  
Map later with Storage Partition  
10 In the Snapshot repository virtual disk parameters area, enter the system  
behavior when:  
The snapshot repository virtual disk is full to the selected  
percentage level.  
The snapshot repository virtual disk is full.  
11 Click Next  
.
The Preview window containing the summary of the snapshot virtual disk  
is displayed.  
12 Click Finish  
.
The Completed window is displayed.  
13 Click OK  
.
14 In the Mappings tab, assign mappings between the snapshot virtual disk  
and the host that accesses the snapshot virtual disk.  
152  
Configuration: Premium Feature—Snapshot Virtual Disks  
15 To register the snapshot virtual disk with the host operating system, run  
the host-based hot_add utility.  
16 To associate the mapping between the storage array name and the virtual  
disk name, run the host-based SMdevices utility.  
Specifying Snapshot Virtual Disk Names  
Choose a name that helps you associate the snapshot virtual disk and  
snapshot repository virtual disk with its corresponding source virtual disk. The  
following information is useful when naming virtual disks:  
By default, the snapshot name is shown in the Snapshot virtual disk name  
field as:  
<source-virtual disk-name>—<sequence-number>  
where sequence-numberis the chronological number of the snapshot  
relative to the source virtual disk.  
The default name for the associated snapshot repository virtual disk that is  
shown in the Snapshot repository virtual disk field is:  
<source-virtual disk-name>—R<sequence-number>  
For example, if you are creating the first snapshot virtual disk for a source  
virtual disk called Accounting, the default snapshot virtual disk is  
Accounting-1, and the associated snapshot repository virtual disk default  
name is Accounting-R1. The default name of the next snapshot virtual disk  
you create based on Accounting is Accounting-2, with the corresponding  
snapshot repository virtual disk named as Accounting-R2 by default.  
Whether you use the software-supplied sequence number that (by  
default) populates the Snapshot virtual disk name or the Snapshot  
repository virtual disk name field, the next default name for a snapshot or  
snapshot repository virtual disk still uses the sequence number determined  
by the software. For example, if you give the first snapshot of source virtual  
disk Accounting the name Accounting-8, and do not use the software-  
supplied sequence number of 1, the default name for the next snapshot of  
Accounting is still Accounting-2.  
The next available sequence number is based on the number of existing  
snapshots of a source virtual disk. If you delete a snapshot virtual disk, its  
sequence number becomes available again.  
Configuration: Premium Feature—Snapshot Virtual Disks  
153  
 
You must choose a unique name for the snapshot virtual disk and the  
snapshot repository virtual disks, or an error message is displayed.  
Names are limited to 30 characters. After you reach this limit in either the  
snapshot virtual disk name or the Snapshot repository virtual disk name  
fields, you can no longer type in the field. If the source virtual disk is 30  
characters, the default names for the snapshot and its associated snapshot  
repository virtual disk use the source virtual disk name truncated enough  
to add the sequence string. For example, for Host Software Engineering  
Group GR-1, the default snapshot name is Host Software Engineering GR-  
1, and the default repository name would be Host Software Engineering  
GR-R1.  
Snapshot Repository Capacity  
If you receive a warning that the capacity for the snapshot repository virtual  
disk is approaching its threshold, you can increase the capacity of a snapshot  
repository virtual disk by using one of the following methods:  
Use the free capacity available on the disk group of the snapshot repository  
virtual disk.  
Add unconfigured capacity to the disk group of the snapshot repository  
virtual disk. Use this option when no free capacity exists on the disk group.  
You cannot increase the storage capacity of a snapshot repository virtual disk  
if the snapshot repository virtual disk has any one of the following conditions:  
The virtual disk has one or more hot spare drives in use.  
The virtual disk has a status other than Optimal.  
Any virtual disk in the disk group is in any state of modification.  
The controller that has ownership of this virtual disk is currently adding  
capacity to another virtual disk. Each controller can add capacity to only  
one virtual disk at a time.  
No free capacity exists in the disk group.  
No unconfigured capacity is available to add to the disk group.  
NOTE: You can add a maximum of two physical disks at one time to increase  
snapshot repository virtual disk capacity.  
154  
Configuration: Premium Feature—Snapshot Virtual Disks  
   
To expand the snapshot repository virtual disk:  
1
2
In the AMW, select the Logical tab.  
Select the snapshot repository virtual disk for which you want to increase  
the capacity.  
3
Select Virtual DiskIncrease Capacity  
.
NOTE: If no free capacity or unconfigured capacity is available, the Increase  
Capacity option is disabled.  
The Increase Snapshot Repository Capacity window displays the Virtual  
disk attributes. The snapshot repository virtual disk name, the associated  
snapshot virtual disk name, the associated source virtual disk capacity and  
name, the current capacity, and the amount of free capacity that is  
available for the selected snapshot repository virtual disk appear. If free  
capacity is available, the maximum free space appears in the Increase  
capacity by  
.
If free capacity does not exist on the disk group, the free space that appears  
in the Increase capacity by spinner box is 0. You must add physical disks to  
create free capacity on the disk group.  
4
To increase capacity of the snapshot repository virtual disk, use one of  
these methods:  
Use the free capacity on the disk group of the snapshot repository  
virtual disk – Go to step 5.  
Add unconfigured capacity, or physical disks to the disk group of the  
snapshot repository virtual disk – Go to step 7.  
5
6
In Increase capacity by, enter or select the appropriate capacity.  
Click OK  
.
The Logical tab is updated. The snapshot repository virtual disk having its  
capacity increased shows a status of Operation in Progress. In addition,  
the snapshot repository virtual disk shows its original capacity and the total  
capacity being added. The virtual disk involved shows a reduction in  
capacity. If all of the free capacity is used to increase the size of the virtual  
disk, the Free Capacity node involved is removed from the Logical tab.  
Configuration: Premium Feature—Snapshot Virtual Disks  
155  
 
7
If unassigned physical disks are not available, do you have empty slots in  
the expansion enclosures?  
Yes, there are empty slots—Insert new physical disks by using the  
information in the initial setup guide for your expansion enclosure. Go  
to step 9.  
No, there are no empty slots—Install another expansion enclosure and  
additional physical disks. Use the information in the initial setup  
guides for your RAID controller module and your expansion enclosure.  
Go to step 9.  
NOTE: The physical disks that you add must be of the same media type and  
interface type as the physical disks that already make up the disk group of the  
snapshot repository virtual disk.  
8
9
Click Add Physical Disks.  
NOTE: The physical disks that appear has a capacity that is either the same  
size or larger than the capacity of the physical disks already being used by the  
disk group.  
Select either a single physical disk to add or two physical disks to add.  
10 Click Add  
.
The Add Physical Disks window closes.  
11 Check the Physical Disks to add [enclosure, slot] area to ensure that the  
correct physical disks have been added.  
12 Either accept the final capacity, or enter or select the appropriate capacity  
in Increase capacity by  
.
13 Click OK  
.
The Logical tab is updated. The snapshot repository virtual disk that is  
having its capacity increased shows a status of Operation in Progress. In  
addition, the snapshot repository virtual disk shows its original capacity  
and the total capacity being added. The Free Capacity node involved in  
the increase shows a reduction in capacity. If all of the free capacity is used  
to increase the size of the virtual disk, the Free Capacity node involved is  
removed from the Logical tab.  
156  
Configuration: Premium Feature—Snapshot Virtual Disks  
   
A new Free Capacity node is created and shown in the Logical tab if these  
conditions exist:  
A
Free Capacity node did not exist prior to the addition of capacity.  
Not all of the capacity that is added is used to increase the capacity of  
the snapshot repository virtual disk.  
On the Physical tab, the unassigned physical disks or unconfigured  
capacity that you added to increase the capacity of the snapshot repository  
virtual disk change to assigned physical disks. The new assigned physical  
disks are associated with the disk group of the snapshot repository virtual  
disk.  
Re-creating Snapshot Virtual Disks  
You can re-create a snapshot virtual disk that you have previously disabled.  
CAUTION: Possible loss of data redundancy – If the snapshot virtual disk is in  
Optimal status, it is first disabled prior to being re-created. This action invalidates  
the current snapshot.  
Keep these important guidelines in mind when you re-create a snapshot  
virtual disk:  
To re-create the snapshot virtual disks correctly, follow the instructions for  
your operating system.  
NOTE: Failing to follow these additional instructions could create unusable  
snapshot virtual disks. For more information, see the PowerVault Modular Disk  
Storage Manager online help topics.  
To use this option, the snapshot virtual disk must be either in an Optimal  
status or Disabled status.  
When using this option, the previously configured snapshot name  
parameters and snapshot repository virtual disk are used.  
To recreate the snapshot virtual disk:  
1
2
3
In the AMW, select the Logical tab, select a snapshot virtual disk.  
Select Virtual DiskSnapshotRe-create  
Type yes, and click OK  
.
.
Configuration: Premium Feature—Snapshot Virtual Disks  
157  
   
Disabling a Snapshot Virtual Disk  
Disable a snapshot virtual disk if:  
You do not need the snapshot now.  
You intend to re-create the snapshot at a later time and want to retain the  
associated snapshot repository virtual disk so that you do not need to  
create it again.  
You want to maximize storage array performance by stopping copy-on-  
write activity to the snapshot repository virtual disk.  
NOTE: If you do not intend to re-create the snapshot virtual disk at a later time, in  
the Logical pane, select the snapshot virtual disk, and select Virtual DiskDelete  
to remove it. The associated snapshot repository virtual disk is also removed. See  
the PowerVault Modular Disk Storage Manager online help topics for more  
information on removing a snapshot virtual disk.  
NOTE: The SMdevices utility displays the snapshot virtual disk in its output, even  
after the snapshot virtual disk is disabled.  
To disable a snapshot virtual disk:  
1
In the AMW, select the Logical tab, select the snapshot virtual disk, and  
select Virtual DiskSnapshotDisable  
In the text box, type yes and click OK  
.
2
.
The snapshot virtual disk is disabled. The associated snapshot repository  
virtual disk does not change status. The copy-on-write activity to the  
snapshot repository virtual disk stops until the snapshot virtual disk is re-  
created.  
158  
Configuration: Premium Feature—Snapshot Virtual Disks  
   
Preparing Host Servers to Re-create a Snapshot Virtual Disk  
NOTE: Before you create a new snapshot of a source virtual disk, stop any data  
access (I/O) activity or suspend data transfer to the source virtual disk and  
snapshot virtual disk to ensure that you capture an accurate snapshot of the source  
virtual disk. Close all applications, including Windows Internet Explorer, to ensure  
all I/O activity has stopped.  
NOTE: Removing the drive letter of the associated virtual disk in Windows or  
unmounting the virtual drive in Linux helps to guarantee a stable copy of the drive  
for the Snapshot.  
Before re-creating a snapshot virtual disk, both the host server and the  
associated virtual disk you are re-creating have to be in the proper state.  
To prepare your host server and virtual disk:  
1
2
Stop all I/O activity to the source and snapshot virtual disk (if mounted).  
Using your Windows system, flush the cache to both the source and the  
snapshot virtual disk (if mounted). At the host prompt, type  
SMrepassist -f <filename-identifier>  
and press <Enter>. For more information, see "SMrepassist Utility" on  
page 281.  
3
4
Click the Summary tab, then click Disk Groups & Virtual Disks to ensure  
that the snapshot virtual disk is in Optimal or Disabled status.  
Remove the drive letter(s) of the source and (if mounted) snapshot virtual  
disk in Windows or unmount the virtual drive(s) in Linux to help  
guarantee a stable copy of the drive for the Snapshot. If this is not done,  
the snapshot operation reports that it has completed successfully, but the  
snapshot data is not updated properly.  
5
Follow any additional instructions for your operating system. Failure to follow  
these additional instructions can create unusable snapshot virtual disks.  
NOTE: If your operating system requires additional instructions, you can find those  
instructions in your operating system documentation.  
Configuration: Premium Feature—Snapshot Virtual Disks  
159  
 
Re-creating a Snapshot Virtual Disk  
After first preparing the host server(s), re-create a snapshot virtual disk. For  
more information, see "Preparing Host Servers to Create the Snapshot Using  
the Simple Path" on page 145 or "Preparing Host Servers to Create the  
Snapshot Using the Advanced Path" on page 149.  
To recreate a snapshot virtual disk:  
NOTE: This action invalidates the current snapshot.  
1
2
3
In the AMW, select the Logical tab, select a snapshot virtual disk.  
Select Virtual DiskSnapshotRe-create  
Type yes, and click OK  
.
.
Re-creating a snapshot repository virtual disk uses the previously  
configured snapshot name and parameters.  
160  
Configuration: Premium Feature—Snapshot Virtual Disks  
   
11  
Configuration: Premium Feature—  
Virtual Disk Copy  
NOTE: A virtual disk copy overwrites data on the target virtual disk. Before starting  
a virtual disk copy, ensure that you no longer need the data or back up the data on  
the target virtual disk.  
NOTE: If you ordered this feature, you received a Premium Feature Activation card  
that shipped in the same box as your Dell PowerVault MD storage array. Follow the  
directions on the card to obtain a key file and to enable the feature.  
NOTE: The preferred method for creating a virtual disk copy is to copy from a  
snapshot virtual disk. This allows the original virtual disk used in the snapshot  
operation to remain fully available for read/write activity while the snapshot is used  
as the source for the virtual disk copy operation.  
When you create a virtual disk copy, you create a copy pair that has a source  
virtual disk and a target virtual disk on the same storage array.  
The source virtual disk is the virtual disk that contains the data you want to  
copy. The source virtual disk accepts the host I/O read activity and stores the  
data until it is copied to the target virtual disk. The source virtual disk can be  
a standard virtual disk, a snapshot virtual disk, or the source virtual disk of a  
snapshot virtual disk. When you start a virtual disk copy, all data is copied to  
the target virtual disk, and the source virtual disk permissions are set to read-  
only until the virtual disk copy is complete.  
The target virtual disk is a virtual disk to which you copy data from the source  
virtual disk. The target virtual disk can be a standard virtual disk, or the  
source virtual disk of a failed or disabled snapshot virtual disk.  
After the virtual disk copy is complete, the source virtual disk becomes  
available to host applications for write requests. To prevent error messages, do  
not attempt to access a source virtual disk that is participating in a virtual disk  
copy while the virtual disk copy is in progress.  
Configuration: Premium Feature—Virtual Disk Copy  
161  
 
Reasons to use virtual disk copy include:  
Copying data for improved access—As your storage requirements for a  
virtual disk change, you can use a virtual disk copy to copy data to a virtual  
disk in a disk group that uses drives with larger capacity within the same  
storage array. Copying data for larger access capacity enables you to move  
data to greater capacity physical disks (for example, 61 GB to 146 GB).  
Restoring snapshot virtual disk data to the source virtual disk—The Virtual  
Disk Copy feature enables you first to restore the data from a snapshot  
virtual disk and then to copy the data from the snapshot virtual disk to the  
original source virtual disk.  
Creating a backup copy—The Virtual Disk Copy feature enables you to  
create a backup of a virtual disk by copying data from one virtual disk (the  
source virtual disk) to another virtual disk (the target virtual disk) in the  
same storage array, minimizing the time that the source virtual disk is  
unavailable to host write activity. You can then use the target virtual disk as  
a backup for the source virtual disk, as a resource for system testing, or to  
copy data to another device, such as a tape drive or other media.  
NOTE: Recovering from a backup copy — You can use the Edit Host-to-Virtual Disk  
Mappings feature to recover data from the backup virtual disk you created in the  
previous procedure. The Mappings option enables you to unmap the source virtual  
disk from its host and then to map the backup virtual disk to the same host.  
Creating a Virtual Disk Copy for an MSCS Shared  
Disk  
To create a virtual disk copy for a Microsoft Cluster Server (MSCS) shared  
disk, create a snapshot of the virtual disk, and then use the snapshot virtual  
disk as the source for the virtual disk copy.  
NOTE: An attempt to directly create a virtual disk copy for an MSCS shared disk,  
rather than using a snapshot virtual disk, fails with the following error:  
The operation cannot complete because the selected  
virtual disk is not a source virtual disk  
candidate.  
162  
Configuration: Premium Feature—Virtual Disk Copy  
   
NOTE: When creating a snapshot virtual disk, map the snapshot virtual disk to only  
one node in the cluster. Mapping the snapshot virtual disk to the host group or both  
nodes in the cluster may cause data corruption by allowing both nodes to  
concurrently access data.  
Virtual Disk Read/Write Permissions  
After the virtual disk copy is complete, the target virtual disk automatically  
becomes read-only to the hosts. The target virtual disk rejects read and write  
requests while the virtual disk copy operation has a status of Pending or In  
Progress or if the operation fails before completing the copy. Keep the target  
virtual disk Read-Only enabled if you want to preserve the data on the target  
virtual disk for reasons such as the following:  
If you are using the target virtual disk for backup purposes.  
If you are using the data on the target virtual disk to copy back to the  
source virtual disk of a disabled or failed snapshot virtual disk.  
If you decide not to preserve the data on the target virtual disk after the  
virtual disk copy is complete, change the write protection setting for the  
target virtual disk to Read/Write.  
To set the target virtual disk read/write permissions:  
1
In the AMW, select Virtual DiskCopyCopy Manager  
.
The Copy Manager window is displayed.  
2
3
Select one or more copy pairs in the table.  
Perform one of these actions:  
To enable Read-Only permission, select ChangeTarget Virtual Disk  
PermissionsEnable Read-Only  
.
NOTE: Write requests to the target virtual disk are rejected when the Read-  
Only permission is enabled on the target virtual disk.  
To disable Read-Only permission, select ChangeTarget Virtual  
Disk PermissionsDisable Read-Only  
.
Configuration: Premium Feature—Virtual Disk Copy  
163  
   
Virtual Disk Copy Restrictions  
Before you perform any virtual disk copy tasks, understand and adhere to the  
restrictions listed in this section. The restrictions apply to the source virtual  
disk, the target virtual disk, and the storage array.  
While a virtual disk copy has a status of In Progress, Pending, or Failed,  
the source virtual disk is available for read I/O activity only. After the  
virtual disk copy is complete, read and write I/O activity to the source  
virtual disk are permitted.  
A virtual disk can be selected as a target virtual disk for only one virtual  
disk copy at a time.  
A virtual disk copy for any virtual disk cannot be mounted on the same  
host as the source virtual disk.  
Windows does not allow a drive letter to be assigned to a virtual disk copy.  
A virtual disk with a Failed status cannot be used as a source virtual disk or  
target virtual disk.  
A virtual disk with a Degraded status cannot be used as a target  
virtual disk.  
A virtual disk participating in a modification operation cannot be selected  
as a source virtual disk or target virtual disk. Modification operations  
include the following:  
Capacity expansion  
RAID-level migration  
Segment sizing  
Virtual disk expansion  
Defragmenting a virtual disk  
NOTE: The following host preparation sections also apply when using the virtual  
disk copy feature through the CLI interface.  
164  
Configuration: Premium Feature—Virtual Disk Copy  
   
Creating a Virtual Disk Copy  
Source virtual disks that are participating in a virtual disk copy are available  
for read I/O activity only while a virtual disk copy has a status of In Progress or  
Pending. Write requests are allowed after the virtual disk copy has completed.  
If the source virtual disk has been formatted with a journaling file system, any  
attempt to issue a read request to the source virtual disk might be rejected by  
the storage array, and an error message might appear. The journaling file  
system driver issues a write request before it attempts to issue the read  
request. The storage array rejects the write request, and the read request  
might not be issued due to the rejected write request. This condition might  
result in an error message appearing, which indicates that the source virtual  
disk is write protected.  
To prevent this issue from occurring, do not attempt to access a source virtual  
disk that is participating in a virtual disk copy while the virtual disk copy has a  
status of In Progress. Also, ensure that the Read-Only attribute for the target  
virtual disk is disabled after the virtual disk copy has completed to prevent  
error messages from appearing.  
The Virtual Disk Copy premium feature includes these items:  
The Create Copy Wizard, which assists in creating a virtual disk copy  
The Copy Manager, which monitors virtual disk copies after they have  
been created  
Before You Begin  
A virtual disk copy fails all snapshot virtual disks that are associated with the  
target virtual disk, if any exist. If you select a source virtual disk of a snapshot  
virtual disk, you must disable all of the snapshot virtual disks that are  
associated with the source virtual disk before you can select it as a target  
virtual disk. Otherwise, the source virtual disk cannot be used as a target  
virtual disk.  
A virtual disk copy overwrites data on the target virtual disk and automatically  
makes the target virtual disk read-only to hosts.  
If eight virtual disk copies with a status of In Progress exist, any subsequent  
virtual disk copy has a status of Pending, which stays until one of the eight  
virtual disk copies completes.  
Configuration: Premium Feature—Virtual Disk Copy  
165  
     
Virtual Disk Copy and Modification Operations  
If a modification operation is running on a source virtual disk or a target  
virtual disk, and the virtual disk copy has a status of In Progress, Pending, or  
Failed, the virtual disk copy does not take place. If a modification operation is  
running on a source virtual disk or a target virtual disk after a virtual disk copy  
has been created, the modification operation must complete before the  
virtual disk copy can start. If a virtual disk copy has a status of In Progress, any  
modification operation does not take place.  
Create Copy Wizard  
The Create Copy Wizard guides you through:  
1
2
3
Selecting a source virtual disk from a list of available virtual disks  
Selecting a target virtual disk from a list of available virtual disks  
Setting the copy priority for the virtual disk copy  
When you have completed the wizard dialogs, the virtual disk copy starts, and  
data is read from the source virtual disk and written to the target virtual disk.  
The Operation in Progress icon appears on the source virtual disk and the target  
virtual disk while the virtual disk copy has a status of In Progress or Pending.  
Failed Virtual Disk Copy  
A virtual disk copy can fail due to these conditions:  
A read error from the source virtual disk  
A write error to the target virtual disk  
A failure in the storage array that affects the source virtual disk or the  
target virtual disk  
When the virtual disk copy fails, a critical event is logged in the Event Log, and a  
Needs Attention icon appears in the Array Management window. While a virtual  
disk copy has this status, the host has read-only access to the source virtual disk.  
Read requests from and write requests to the target virtual disk do not take place  
until the failure is corrected by using the Recovery Guru.  
166  
Configuration: Premium Feature—Virtual Disk Copy  
         
Preferred RAID Controller Module Ownership  
During a virtual disk copy, the same RAID controller module must own both  
the source virtual disk and the target virtual disk. If both virtual disks do not  
have the same preferred RAID controller module when the virtual disk copy  
starts, the ownership of the target virtual disk is automatically transferred to  
the preferred RAID controller module of the source virtual disk. When the  
virtual disk copy is completed or is stopped, ownership of the target virtual  
disk is restored to its preferred RAID controller module. If ownership of the  
source virtual disk is changed during the virtual disk copy, ownership of the  
target virtual disk is also changed.  
Failed RAID Controller Module  
You must manually change RAID controller module ownership to the  
alternate RAID controller module to allow the virtual disk copy to complete  
under all of these conditions:  
A virtual disk copy has a status of In Progress  
The preferred RAID controller module of the source virtual disk fails  
The ownership transfer does not occur automatically in the failover  
Copy Manager  
After you create a virtual disk copy by using the Create Copy Wizard, you can  
monitor the virtual disk copy through the Copy Manager. From the Copy  
Manager, a virtual disk copy may be re-copied, stopped, or removed. You can  
also modify the attributes, such as the copy priority and the target virtual disk  
Read-Only attribute. You can view the status of a virtual disk copy in the  
Copy Manager. Also, if you need to determine which virtual disks are  
involved in a virtual disk copy, you can use the Copy Manager or the storage  
array profile.  
Configuration: Premium Feature—Virtual Disk Copy  
167  
           
Copying the Virtual Disk  
You can create a virtual disk copy by using the Create Copy Wizard.  
CAUTION: Possible loss of data access – A virtual disk copy overwrites data on  
the target virtual disk.  
A virtual disk copy automatically makes the target virtual disk read-only to  
hosts. You might want to keep this attribute enabled to preserve the data on  
the target virtual disk.  
CAUTION: If you decide not to preserve the data on the target virtual disk after the  
virtual disk copy has completed, disable the Read-Only attribute for the target  
virtual disk. See "Virtual Disk Read/Write Permissions" on page 163 for more  
information on enabling and disabling the Read-Only attribute for the target virtual  
disk.  
To prevent write-protected error messages from appearing, do not try to  
access a source virtual disk that is participating in a virtual disk copy while the  
virtual disk copy has a status of In Progress. Also, ensure that the Read-Only  
attribute for the target virtual disk is disabled after the virtual disk copy has  
completed to prevent error messages from appearing.  
To copy the virtual disk:  
1
2
3
4
Stop all I/O activity to the source virtual disk and the target virtual disk.  
Unmount any file systems on the source virtual disk and the target virtual disk.  
In the AMW, select the Logical tab and select the source virtual disk.  
Select Virtual DiskCopyCreate  
.
The Create Copy Wizard - Introduction window is displayed.  
5
In the Select source virtual disk area, select the appropriate virtual disk  
and click Next.  
The Select Target Virtual Disk and Copy Priority window is displayed.  
6
7
Click Next  
The Preview window displays the summary of your selections.  
Type yes, and click Finish  
.
.
NOTE: The Operation in Progress icon appears on the source virtual disk and the  
target virtual disk while the virtual disk copy has a status of In Progress or Pending.  
168  
Configuration: Premium Feature—Virtual Disk Copy  
   
Storage Array Performance During Virtual Disk  
Copy  
The following factors contribute to the overall performance of the  
storage array:  
I/O activity  
Virtual disk RAID level  
Virtual disk configuration—Number of drives in the virtual disk groups  
Virtual disk type—Snapshot virtual disks might take more time to copy  
than standard virtual disks  
During a virtual disk copy, resources for the storage array are diverted from  
processing I/O activity to completing a virtual disk copy. This affects the  
overall performance of the storage array. When you create a new virtual disk  
copy, you define the copy priority to determine how much controller  
processing time is diverted from I/O activity to a virtual disk copy operation.  
Setting Copy Priority  
You can use the Copy Manager to select the rate at which a virtual disk copy  
completes for a selected copy pair. You can change the copy priority for a copy  
pair at any of these times:  
Before the virtual disk copy begins  
While the virtual disk copy has a status of In Progress  
When you re-create a virtual disk copy  
To set copy priority:  
1
2
In the AMW, select Virtual DiskCopyCopy Manager  
.
The Copy Manager window is displayed.  
In the table, select one or more copy pairs.  
Configuration: Premium Feature—Virtual Disk Copy  
169  
       
3
4
Select ChangeCopy Priority  
.
The Change Copy Priority window is displayed.  
In the Copy priority area, select the appropriate copy priority, depending  
on your system performance needs.  
NOTE: There are five copy priority rates available: lowest, low, medium, high,  
and highest. If the copy priority is set at the lowest rate, I/O activity is  
prioritized, and the virtual disk copy takes longer.  
Stopping a Virtual Disk Copy  
You can stop a virtual disk copy operation that has an In Progress status, a  
Pending status, or a Failed status. Stopping a virtual disk copy that has a  
Failed status clears the Needs Attention status displayed for the storage array.  
Keep these guidelines in mind when you stop a virtual disk copy:  
To use this option, select only one copy pair in the Copy Manager  
.
When the virtual disk copy is stopped, all of the mapped hosts have write  
access to the source virtual disk. If data is written to the source virtual disk,  
the data on the target virtual disk no longer matches the data on the  
source virtual disk.  
To stop a virtual disk copy, complete the following steps:  
1
In the AMW, select Virtual DiskCopyCopy Manager  
.
The Copy Manager window is displayed.  
2
3
4
Select the copy pair in the table.  
Select CopyStop  
Click Yes  
.
.
Recopying a Virtual Disk  
You can recopy a virtual disk when you have stopped a virtual disk copy and  
you want to start it again or when a virtual disk copy has failed.  
The Recopy option overwrites existing data on the target virtual disk and  
makes the target virtual disk read-only to hosts. This option fails all snapshot  
virtual disks associated with the target virtual disk, if any exist.  
170  
Configuration: Premium Feature—Virtual Disk Copy  
       
Preparing Host Servers to Recopy a Virtual Disk  
NOTE: Before you create a new copy of a source virtual disk, stop any data access  
(I/O) activity or suspend data transfer to the source virtual disk (and, if applicable,  
the target disk) to ensure that you capture an accurate point-in-time image of the  
source virtual disk. Close all applications, including Windows Internet Explorer, to  
ensure all I/O activity has stopped.  
NOTE: Removing the drive letter of the associated virtual disk(s) in Windows or  
unmounting the virtual drive in Linux helps to guarantee a stable copy of the drive  
for the virtual disk copy.  
Before creating a new virtual disk copy for an existing copy pair, both the host  
server and the associated virtual disk you are recopying have to be in the  
proper state. Perform the following steps to prepare your host server and  
virtual disk:  
1
2
Stop all I/O activity to the source and target virtual disk.  
Using your Windows system, flush the cache to both the source and the  
target virtual disk (if mounted). At the host prompt, type  
SMrepassist -f <filename-identifier>  
and press <Enter>. For more information, see "SMrepassist Utility" on  
page 281.  
3
4
Click the Summary tab, then click Disk Groups & Virtual Disks to ensure  
that the virtual disk is in Optimal or Disabled status.  
Remove the drive letter(s) of the source and (if mounted) virtual disk in  
Windows or unmount the virtual drive(s) in Linux to help guarantee a  
stable copy of the drive for the virtual disk. If this is not done, the copy  
operation reports that it has completed successfully, but the copied data is  
not is not updated properly.  
5
Follow any additional instructions for your operating system. Failure to  
follow these additional instructions can create unusable virtual disk copies.  
NOTE: If your operating system requires additional instructions, you can find  
those instructions in your operating system documentation.  
Configuration: Premium Feature—Virtual Disk Copy  
171  
 
Recopying the Virtual Disk  
You can use the Copy Manager to create a new virtual disk copy for a selected  
source virtual disk and a target virtual disk. Use this option when you have  
stopped a virtual disk copy and want to start it again or when a virtual disk  
copy has failed or completed. The virtual disk copy starts over from the  
beginning.  
Possible loss of data—The re-copying operation overwrites existing data on  
the target virtual disk.  
Possible loss of data access—While a virtual disk copy has a status of In  
Progress or Pending, source virtual disks are available for read I/O activity  
only. Write requests are allowed after the virtual disk copy has completed.  
Keep these guidelines in mind when re-copying a virtual disk:  
If hosts are mapped to the source virtual disk, the data that is copied to the  
target virtual disk when you perform the re-copy operation might have  
changed since the previous virtual disk copy was created.  
Select only one virtual disk copy in the Copy Manager dialog.  
To recopy the virtual disk:  
1
2
3
Stop all I/O to the source virtual disk and the target virtual disk.  
Unmount any file systems on the source virtual disk and the target virtual disk.  
In the AMW, select Virtual DiskCopyCopy Manager  
.
The Copy Manager window is displayed.  
4
5
Select the copy pair in the table.  
Select CopyRe-Copy  
.
The Re-Copy window is displayed.  
6
Set the copy priority.  
NOTE: There are five copy priority rates available: lowest, low, medium, high, and  
highest. If the copy priority is set at the lowest rate, I/O activity is prioritized, and the  
virtual disk copy takes longer. If the copy priority is set to the highest priority rate,  
the virtual disk copy is prioritized, but I/O activity for the storage array might be  
affected.  
172  
Configuration: Premium Feature—Virtual Disk Copy  
 
Removing Copy Pairs  
You can remove one or more virtual disk copies by using the Copy Manager.  
Any virtual disk copy-related information for the source virtual disk and the  
target virtual disk is removed from the Virtual Disk Properties dialog and the  
Storage Array Profile dialogs. When you remove a virtual disk copy from the  
storage array, the Read-Only attribute for the target virtual disk is also removed.  
After the virtual disk copy is removed from the Copy Manager, you can either  
select the target virtual disk as a source virtual disk or the target virtual disk for  
a new virtual disk copy. If you remove a virtual disk copy, the source virtual disk  
and the target virtual disk no longer appear in the Copy Manager.  
Keep these guidelines in mind when you remove copy pairs:  
Removing copy pairs does not delete the data on the source virtual disk or  
target virtual disk.  
If the virtual disk copy has a status of In Progress, you must stop the  
virtual disk copy before you can remove the copy pair.  
To remove copy pairs:  
1
In the AMW, select Virtual DiskCopyCopy Manager  
.
The Copy Manager window is displayed.  
2
3
In the table, select one or more copy pairs.  
Select CopyRemove Copy Pairs  
The Remove Copy Pairs dialog appears.  
Click Yes  
.
4
.
Configuration: Premium Feature—Virtual Disk Copy  
173  
   
174  
Configuration: Premium Feature—Virtual Disk Copy  
12  
Configuration: Premium Feature—  
Upgrading to High–Performance  
Tier  
The High Performance Tier premium feature on an MD3600f Series array  
increases the performance of the system beyond that of an MD3600f Series  
array operating at the standard performance level.  
If this feature is ordered, a Premium Feature Activation card is placed in the  
box with the storage array. After reading the information below, follow the  
directions on the card to obtain a key file and to enable the feature.  
CAUTION: Loss of data access – The storage array automatically restarts when  
the High-Performance-Tier feature is enabled or disabled. During the restart data  
is unavailable. Data availability is restored when the array restarts.  
To upgrade from a standard-performance-tier storage array to a high-  
performance-tier storage array, you enable the high-performance-tier  
premium feature, using the Dell PowerVault Modular Disk Storage  
Management (MDSM) software.  
When the high performance tier feature is enabled or disabled the array  
restarts. During this time data access and management access to the  
controller is temporarily lost.  
It is recommended that all I/O to the array be stopped before this feature is  
enabled or disabled.  
While the array is restarting, the state of the array in the MDSM application  
may change from Optimal to Unresponsive. When the restart completes the  
status returns to Optimal.  
When the array status returns to Optimal, I/O to the array can be restarted.  
Configuration: Premium Feature—Upgrading to High–Performance Tier  
175  
 
176  
Configuration: Premium Feature—Upgrading to High–Performance Tier  
13  
Configuration: Premium Feature—  
Remote Replication  
The Remote Replication premium feature enables real-time replication of data  
between two storage arrays in separate locations. Using remote replication, up  
to 16 replicated virtual disk pairs can be created on a storage array.  
A replicated virtual disk pair is created from two standard virtual disks, both  
of which are logical data storage structures created on a storage array. A  
standard virtual disk can only be a member of one replicated pair. A pair  
consists of a primary virtual disk at a local storage array and a secondary  
virtual disk at a remote storage array. Data written by the host to the primary  
virtual disk is replicated directly from the primary storage array to the  
secondary virtual disk. No host server or application resources are used during  
replication. The data can be written either synchronously or asynchronously.  
In the event of a disaster, loss of communications or catastrophic failure in  
the local storage array, the secondary virtual disk in the remote storage array  
can be promoted to the role of primary virtual disk and assume responsibility  
for maintaining business operations.  
For detailed steps on creating virtual disks and implementing remote  
replication, see "Setting Up and Managing Remote Replication" on page 183.  
Replication  
Access to the host server from the local and/or remote storage arrays using  
remote replication must be established through a supported Fibre Channel  
switch configuration. Direct-attached configurations between the host server  
and Dell PowerVault MD storage arrays are not supported.  
For more information on configuring the switch environment, see "Required  
Switch Zoning Configurations" on page 182.  
Configuration: Premium Feature— Remote Replication  
177  
     
Activating Remote Replication  
NOTE: If you ordered the Remote Replication feature, you received a Premium  
Feature Activation card in the Dell PowerVault MD storage array shipping box.  
Follow the directions on the card to install and enable this feature.  
After the Remote Replication feature has been installed following the steps  
on the Premium Feature Activation card, it must be activated using a wizard-  
based process. As part of activation, the following actions occur:  
Two replication repository virtual disks are created using existing capacity  
on either a disk group or storage array specified by the user.  
All hosts currently using the highest-numbered Fibre Channel host port  
(Port 3) on each RAID controller are logged out.  
The highest-numbered Fibre Channel host port (Port 3) on each RAID  
The highest-numbered Fibre Channel host port (Port 3) on each RAID  
controller is not available to direct host access when the Remote  
Replication premium feature is activated. Any attempted host  
communication to this port is rejected.  
Communication between the replication-reserved Fibre Channel ports  
requires established zoning of Fibre Channel switches in the fabric, see  
"Required Switch Zoning Configurations" on page 182.  
Before establishing virtual disk replication pairs, the Remote Replication  
premium feature must be activated on each of the storage arrays  
participating in replication.  
Replication Repository Virtual Disks  
Using the Remote Replication premium feature on the storage array requires  
that two replication repository virtual disks be created on one of the disk  
groups on the storage array. The RAID controller module stores replication  
information on these virtual disks, including information about incomplete  
remote writes that can be used to recover from RAID controller module resets  
and accidental storage array shutdown.  
178  
Configuration: Premium Feature— Remote Replication  
   
Other information about replication repository virtual disks include:  
Specifying that replication repository virtual disks be created from  
unconfigured free capacity on the disk group, or creating a new disk group  
and its member replication repository virtual disks from unconfigured free  
capacity on the storage array.  
NOTE: The disk group containing replication repository virtual disks cannot  
be deleted without disabling the Remote Replication premium feature.  
Default names of the replication repository virtual disks are Replication  
Repository 1 and Replication Repository 2. These names cannot be changed.  
The activation process creates replication repository virtual disk with  
identical capacity. In a dual RAID controller module storage array, the  
default capacity for each replication repository virtual disks is 128 MB per  
controller (requiring a total of 256 MB in the RAID group). This default  
capacity cannot be modified.  
RAID Levels for Replication Repository Virtual Disks  
If replication repository virtual disks are created from unconfigured free capacity  
on the storage array, you can select a RAID level for the disk group of either RAID  
1/10, RAID 5, or RAID 6. However, if the replication repository virtual disks are  
created from an existing disk group, RAID levels are not user configurable.  
Primary and Secondary Virtual Disk Pairs  
After the Remote Replication feature is activated and configuration of the FC  
fabric, individual virtual disk replication pairs are be created. Up to 16  
replication virtual disk pairs can be established. Each pair consists of a  
primary (online) virtual disk and a secondary (offline) virtual disk, both of  
identical capacity. During operation, primary and secondary roles can be  
interchanged to test the replicated data sets without breaking the  
relationship. Independent copies may be created by deleting the relationship,  
in which case both virtual disks are available online to their configured hosts.  
Configuration: Premium Feature— Remote Replication  
179  
   
Before creating a virtual disk pair, verify the following:  
1
2
3
The Remote Replication premium feature has been installed, enabled and  
activated on the primary and secondary storage arrays.  
The storage arrays containing the two virtual disks you want to replicate  
are connected through a Fibre Channel fabric interface.  
The secondary storage array contains a virtual disk that is equal (or larger)  
in capacity than the primary virtual disk in the replicated pair.  
4
5
Both virtual disks are visible to the storage management software.  
All I/O into the secondary virtual disk is stopped and all file systems are  
unmounted.  
CAUTION: Creating a replicated virtual disk pair starts a synchronous process  
between the primary and secondary virtual disks. This process overwrites all  
existing data on the secondary virtual disk and sets the disk to read-only access. If  
you have existing data on your secondary virtual disk, back up the data before  
creating the replicated virtual disk pair.  
Using Remote Replication With Other Features  
Remote replication can be used with other disk utility features such as:  
Storage Partitioning  
Snapshot Virtual Disk Premium Feature  
Virtual Disk Copy Premium Feature  
Virtual Disk Expansion (VDE)  
Snapshot Virtual Disk and Virtual Disk Copy are premium features that must  
be enabled and activated on the primary storage array.  
Storage Partitioning With Remote Replication  
Storage partitioning allows hosts to share access to virtual disks in a storage  
array. A storage partition is created when a collection of hosts (host group) or  
a single host is defined, then a virtual disk-to-logical unit number (LUN)  
mapping is defined. This mapping allows you to define which host or host  
group has access to a particular virtual disk in the storage array.  
Storage partition definitions for the local storage array and remote storage  
array are independent. Establishing similar definitions on the disk serving in a  
secondary role reduces administrative effort required during site recovery.  
180  
Configuration: Premium Feature— Remote Replication  
   
Snapshot Virtual Disk With Remote Replication  
A snapshot virtual disk is a point-in-time image of a virtual disk. In a remote  
replication, do not mount a snapshot virtual disk on the same server as the  
primary virtual disk.  
Virtual Disk Copy With Remote Replication  
The virtual disk copy premium feature copies data from a source virtual disk  
to a target virtual disk within the same storage array.  
A primary virtual disk in a remote replication can be either a source virtual  
disk or a target virtual disk in a virtual disk copy.  
You can create a virtual disk copy on the primary virtual disk in a replicated pair,  
but you cannot create a virtual disk copy of a secondary virtual disk in a replicated  
pair. Instead, you can copy a secondary virtual disk in one of two ways:  
Promote the secondary virtual disk to the primary role.  
CAUTION: If a primary/secondary role reversal is begun while a virtual disk copy  
is in progress, the virtual disk copy fails and be unable to restart.  
Create a snapshot virtual disk of the secondary virtual disk, then perform a  
virtual disk copy of the snapshot virtual disk.  
Virtual Disk Expansion With Remote Replication  
Virtual Disk Expansion (VDE) increases the capacity of an existing virtual  
disk. This increased capacity is achieved by using the free capacity available  
on either:  
The disk group of the standard virtual disk, or  
The snapshot repository virtual disk.  
Performing a VDE operation does not interrupt access to data on disk groups,  
virtual disks or physical disks.  
A VDE operation can be performed on a primary virtual disk or a secondary  
virtual disk of a replicated pair. It cannot be performed on a replication  
repository virtual disk. No other action can be performed on a virtual disk  
while it is undergoing expansion (except normal disk I/O).  
NOTE: To perform a VDE operation, the remote replication must be in Optimal  
status. The properties pane in the logical view displays the status of a virtual disk.  
Configuration: Premium Feature— Remote Replication  
181  
     
Required Switch Zoning Configurations  
NOTE: A limited set of Fibre Channel switches are supported by the storage array.  
For a complete list of supported switches and their required driver, BIOS and  
firmware levels, see the Configuring Fibre Channel with the Dell MD3600f-Series  
Storage Array.  
Due to potential restrictions at the host level, remote replication  
configurations must contain Fibre Channel switches. Fibre Channel switches  
must be zoned so that a single host adapter accesses only one RAID controller  
module in a storage array. Additionally, all zone configurations should specify  
separate zones for ports that are reserved for remote replication functions.  
NOTE: Do not zone the uplink port (E_port) that connects (cascades) switches  
within a fabric.  
Switch zoning configurations are typically set using the switch management  
software provided by the Fibre Channel switch vendor. This software is either  
included with the switch itself, or downloaded from the vendor's website.  
Zoning Guidelines for Remote Replication  
Enabling remote replication on your storage array automatically dedicates one  
port per RAID controller exclusively to replication. These dedicated ports,  
Port 3 on RAID controller 0 and Port 3 on RAID controller 1, must be in  
different zones on the Fibre Channel switch. Generally, switch zoning should  
be established so that:  
All RAID Controller 0 Fibre Channel Port 3s are in a common zone.  
All RAID Controller 1 Fibre Channel Port 3s are in a common zone, but  
separately zoned from RAID Controller 0 ports.  
Host Fibre Channel access ports are in separate zones from those  
established for remote replication.  
Switch Cascading  
When two or more Fibre Channel switches are cascaded, the switch  
management software combines the ports for all of the switches that are  
linked.  
182  
Configuration: Premium Feature— Remote Replication  
         
Journaling File Systems and Remote Replication  
When using a journaling file system, you cannot gain read-only access to a  
remote virtual disk. A journaling file system does not let you mount the  
remote virtual disk in Windows NTFS. However, you can mount a snapshot  
of the remote virtual disk, if available.  
Setting Up and Managing Remote Replication  
Prerequisites  
Before setting up remote replication between two storage arrays, ensure that:  
The Remote Replication premium feature has been activated.  
The local storage array contains two replication repository virtual disks  
(created during activation).  
The local storage array contains the primary virtual disk to be replicated.  
The remote storage array contains the secondary virtual disk to serve as the  
replication target, and:  
Both virtual disks are visible to a single management station;  
replicated disk pairs cannot be managed on separate instances of  
storage management software.  
The RAID level of the secondary virtual disk can be different than that  
of the primary virtual disk.  
The capacity of the secondary virtual disk (after creation and  
primary virtual disk.  
The storage arrays containing the virtual disks you want to replicate are  
connected to each other through a Fibre Channel switch fabric and are  
accessible from a single management station.  
If either the primary or secondary virtual disk does not exist, you must create  
it before proceeding with remote replication setup. See "Activating the  
Remote Replication Premium Feature and Creating Replication Virtual  
Disks" on page 184 and complete the necessary steps.  
If all virtual disks are created, skip to "Creating a Remote Replication" on  
page 186.  
Configuration: Premium Feature— Remote Replication  
183  
       
Activating the Remote Replication Premium Feature and Creating  
Replication Virtual Disks  
To activate the Remote Replication feature and create the two replication  
repository virtual disks required, you can either:  
Activate the feature, then create the two replication virtual disks from the  
total unconfigured capacity on the storage array, or  
Activate the feature, then create the two replication virtual disks from the  
unconfigured capacity in an already-existing disk group  
Activating and Creating Replication Repository Virtual Disks From the  
Storage Array  
To activate the Remote Replication feature and create the two replication  
repository virtual disks required, you can either:  
1
In the Activate Remote Replication wizard, select Unconfigured capacity  
(create a new disk group) and click Next.  
The Activate Remote Replication - Create Disk Group wizard is displayed.  
2
3
In Disk Group Name, type a unique name for the disk group.  
Select one to select a physical disk:  
Automatic—The storage management software generates a list of  
available capacity and physical disk options for each available RAID level.  
Manual—The storage management software generates a list of  
unselected physical disks.  
4
5
Click Next.  
If you selected Automatic, an empty Select Capacity table and a drop-  
down list of available RAID levels is displayed. Go to Step 5.  
If you selected Manual, a populated Unselected Physical Disks table,  
empty Selected Physical Disks table, and a drop-down list of available  
RAID levels is displayed.  
In the Select RAID level dropdown list, select the RAID level for the disk  
group. The Select capacity table displays the available virtual disks for the  
RAID level.  
184  
Configuration: Premium Feature— Remote Replication  
     
6
In the Select capacity table, select the physical disks and capacities for the  
new disk group, then click Next.  
The Preview (Activate Remote Replication) wizard is displayed.  
7
8
Click Finish.  
The Completed (Activate Remote Replication) message is displayed.  
Click OK  
.
The Remote Replication premium feature is now active. The Logical pane  
displays the new disk group and the two replication repository virtual disks.  
Activating and Creating Replication Repository Virtual Disks From an  
Existing Disk Group  
Using this method for creating the replication repository virtual disks, free  
capacity in an existing disk group is used. By default, the replication  
repository virtual disks each have either 128 MB or 256 MB capacity. You  
cannot create the replication repository virtual disks on a disk group with  
insufficient capacity, or change the default capacities of the replication  
repository virtual disks.  
1
In the Introduction (Activate Remote Replication) wizard, select Free  
capacity on existing disk groups.  
2
From the list of available disk groups, select a disk group to contain the  
replication repository virtual disks, then click Next  
The Preview (Activate Remote Replication) wizard is displayed.  
Click Finish  
The Completed (Activate Remote Replication) message is displayed.  
Click OK  
.
3
4
.
.
The Remote Replication premium feature is now active. The Logical pane  
displays the two replication repository virtual disks as part of the disk group.  
Configuration: Premium Feature— Remote Replication  
185  
 
Creating a Remote Replication  
Before creating a remote replication, ensure that all prerequisites are met. See  
"Prerequisites" on page 183 for more information.  
1
2
Open the AMW of both the local and remote storage array.  
Verify that the Remote Replication premium feature has been activated on  
both storage arrays.  
3
4
In the AMW of the local storage array, select the Logical tab.  
In the Logical pane of the local storage array, select the virtual disk you  
created to serve as your primary virtual disk. This is the disk that is  
replicated to the secondary disk.  
5
6
7
In the menu bar in the AMW, select Virtual DiskRemote–  
ReplicationCreate.  
The Introduction (Activate Remote Replication) wizard is displayed.  
Click Next.  
The Select Storage Array (Create Remote Replication) dialog is  
displayed. The Storage Arrays list shows the remote storage arrays.  
Select the storage array where you created the secondary virtual disk, then  
click Next.  
The Select Secondary Virtual Disk (Create Remote Replication) wizard is  
displayed. Go to "Selecting the Secondary Virtual Disk" on page 187.  
186  
Configuration: Premium Feature— Remote Replication  
   
Selecting the Secondary Virtual Disk  
CAUTION: Creating a replicated virtual disk pair starts a process between the  
primary and secondary virtual disks that overwrites all existing data on the  
secondary virtual disk and set the disk to read-only access. If you have existing  
data on your secondary virtual disk stop all I/O to the disk, back up the data, and  
unmount any file systems mounted to the secondary virtual disk before creating  
the replicated virtual disk pair.  
1
2
Select the remote storage array and the primary virtual disk.  
In the Select Secondary Virtual Disk (Create Remote Replication)  
wizard, select the secondary virtual disk.  
NOTE: The secondary virtual disk must have a capacity equal to or greater  
than the capacity of the primary virtual disk.  
3
Click Next.  
The Set Write Mode (Create Remote Replication) wizard is displayed.  
Go to "Setting the Write Mode" on page 187.  
Setting the Write Mode  
The secondary host ports on the storage arrays are reserved for data  
synchronization between the primary virtual disk and the secondary virtual  
disk in a replicated virtual disk pair. You can set remote replication to write  
either synchronously or asynchronously.  
In the synchronous mode, the RAID controller module on the primary  
virtual disk sends an I/O completion message back to the host storage array  
after the data has been successfully copied to the secondary storage array.  
The synchronous mode is the preferred mode of operation, since it offers  
the best chance of full data recovery from the secondary storage array in  
the event of a disaster. However, a synchronous replication mode can  
degrade the I/O performance of the host when long-distance data transfer  
is involved.  
In the asynchronous mode, the RAID controller module on the primary  
storage array sends an I/O completion message to the host storage array  
before the data is successfully copied to the secondary storage array. The  
asynchronous mode offers faster host I/O performance; however, it does  
not guarantee that data was successfully written to the secondary virtual  
disk or that the write requests were completed on the secondary virtual  
disk in the same order they were initiated. In asynchronous mode, you can  
Configuration: Premium Feature— Remote Replication  
187  
   
also specify whether to use a write consistency group option. A write  
consistency group ensures that the secondary virtual disk receives write  
requests in the sequence initiated by the RAID controller module of the  
primary virtual disk. The secondary virtual disk can also be added to a write  
consistency group.  
To set the write mode for the remote replication:  
1
2
3
In the Set Write Mode (Create Remote Replication) wizard, select either  
the Synchronous or Asynchronous mode.  
If you select asynchronous mode, select whether to use the Add to write  
consistency group option.  
Click Next  
.
The Select Synchronization Settings (Create Remote Replication) wizard  
is displayed. Go to "Setting Synchronization Priority and Synchronization  
Method" on page 188.  
Setting Synchronization Priority and Synchronization Method  
You can set the priority for allocating system resources to synchronizing the  
remote replication.  
Higher synchronization priorities allocate more resources to the process  
and might degrade I/O performance.  
Lower synchronization priorities allocate fewer resources to the process  
and have less impact on normal I/O performance.  
The initial synchronization priority and synchronization method you select  
can be changed later. For more information about resynchronizing virtual  
disks in a remote replication, go to "Resynchronizing Virtual Disks" on  
page 196.  
To set Synchronization Priority and Synchronization Method:  
1
In the Select Synchronization Settings (Create Remote Replication)  
wizard, select the synchronization priority on the slide bar.  
2
Select either Manual or Automatic resynchronization.  
Automatic resynchronization—Resynchronization starts immediately  
after communication is restored between unsynchronized replicated  
virtual disks.  
188  
Configuration: Premium Feature— Remote Replication  
 
Manual resynchronization (recommended)—The replicated pair  
must be manually resynchronized each time communication is  
restored between unsynchronized replicated virtual disks.  
3
Click Next.  
The Preview (Create Remote Replication) wizard is displayed. Go to  
"Completing the Remote Replication" on page 189.  
Completing the Remote Replication  
CAUTION: Creating a replicated virtual disk pair starts a process between the  
primary and secondary virtual disks that overwrites ALL existing data on the  
secondary virtual disk and set the disk to read-only access. If you have existing  
data on your secondary virtual disk: (1) stop all I/O to the disk, (2) back up the data  
and (3) unmount any file systems mounted to the secondary virtual disk before  
creating the replicated virtual disk pair.  
After selecting the synchronization settings, perform these steps to complete  
the remote replication:  
1
In the Preview (Create Remote Replication) wizard, type Yes to confirm  
the synchronization. Then, click Finish  
.
The Creation Successful (Create Remote Replication) confirmation  
message is displayed.  
2
If you want to create another replication pair, click Yes and repeat the  
steps. Otherwise, click No to exit.  
disk pairs as members of their disk groups.  
In the local storage array, the Properties pane displays the Replication  
status as Synchronizing and the Synchronization – Progress bar shows  
estimated time to completion.  
To view detailed information about the virtual disks in a remote replication,  
go to either "Viewing Information About a Remote Replication or Replication  
Repository Virtual Disk" on page 190 in the Properties Pane or Viewing  
Information about a Remote Replication or Replication Repository Virtual  
Disk in the Storage Array Profile. Each virtual disk is also represented by a  
status icon in the Devices tab on the Enterprise Management Window. For a  
description of each icon, see "Virtual Disk Status Icons" on page 194.  
Configuration: Premium Feature— Remote Replication  
189  
 
RAID Controller Module Ownership/Preferred Path  
During a remote replication operation, the RAID controller module that owns  
the primary virtual disk must correspond to the RAID controller owning the  
secondary volume on the remote array. If both virtual disks do not have the  
same preferred RAID controller module when a remote replication begins, the  
ownership of the secondary virtual disk is automatically transferred to the  
preferred RAID controller module of the primary virtual disk.  
When the remote replication is completed or is stopped, ownership of the  
secondary virtual disk is restored to its preferred RAID controller module.  
If ownership of the primary virtual disk is changed during the remote  
replication, ownership of the secondary virtual disk is also changed.  
If any of the following conditions apply to your RAID controller module, you  
must manually change module ownership to the alternate RAID controller  
module to allow the remote replication to finish:  
The status of the remote replication is In Progress.  
The preferred RAID controller module of the primary virtual disk fails.  
Transfer of ownership does not occur automatically during failover.  
CAUTION: To avoid possible loss of data, verify that either the virtual disks are  
not in use, or a multi-path driver is installed on the host. If you change the RAID  
controller module ownership/preferred path while an application is using one of  
the virtual disks, I/O activity is disrupted and I/O errors may occur.  
Viewing Information About a Remote Replication or Replication  
Repository Virtual Disk  
In MD Storage Manager, the Storage Array Profile and the Properties pane  
display information about the physical and logical characteristics of the  
remote replication.  
190  
Configuration: Premium Feature— Remote Replication  
   
Viewing the Storage Array Profile  
The Storage Array Profile displays the most detailed information about the  
components of a remote replication and the replication repository virtual  
disks. You can:  
View detailed information about individual and paired virtual disks in a  
remote replication  
View detailed information about the replication repository virtual disks in  
the remote replication and storage array  
Save the Storage Array Profile information as a text file  
To view the Storage Array Profile:  
1
In the Array Management Window of the local or remote storage array,  
select the Summary tab.  
2
In the Status area, click Storage Array Profile  
.
The Storage Array Profile opens.  
3
4
Select the Virtual Disks tab.  
Select either the Replications tab or the Repositories tab. The Profile for  
Storage array page is displayed.  
5
Perform one of the following:  
To return to the Array Management Window without saving the  
information, click Close  
To save the information, click Save As, then go to Step 6.  
.
6
7
In the Section Selection area of the Save Profile window, select each  
section you want to save.  
To save the file, either:  
Save the file in the default directory, or  
Save the file in another directory by choosing it in the Look in drop-  
down list.  
8
In File name, type a name for the file and click Save.  
The file is saved as a .txt file.  
Configuration: Premium Feature— Remote Replication  
191  
Viewing the Properties Pane  
The Properties pane is a view-only display of the physical and logical  
characteristics of a single virtual disk in a replicated pair, or a single  
replication repository virtual disk.  
To view the Properties pane:  
1
2
In the AMW, select the Logical tab.  
Select either the primary or secondary virtual disk of the replicated pair.  
The Properties pane shows properties for the selected virtual disk. Under  
Replication properties, the Replication status displays the synchronization  
status of the replicated pair. When the primary and secondary virtual disks are  
synchronizing, the Replication status shows a synchronizing icon see "Virtual  
Disk Status Icons" on page 194).  
Viewing Logical Elements of the Secondary Virtual Disk  
To view the logical elements of the secondary virtual disk of a remote  
replication:  
1
2
3
In the AMW of the local storage array, select the Logical tab.  
Right-click the secondary virtual disk of the remote replication.  
Select View Associated Logical Elements  
.
The View Associated Logical Elements pop-up is displayed, revealing these  
logical elements:  
Primary and secondary virtual disks and their locations  
Replication repository virtual disks and their locations  
192  
Configuration: Premium Feature— Remote Replication  
   
Viewing the Physical Components or Logical Elements of the Primary  
Virtual Disk  
To view the physical components or logical elements of the primary virtual  
disk in a remote replication:  
1
In the AMW of the storage array that contains t he primary virtual disk,  
select the Logical tab.  
2
Right-click the primary virtual disk and either:  
View the logical elements of the primary virtual disk by selecting  
Select ViewAssociated Logical Elements. The View Associated  
Logical Elements pop-up is displayed, showing visual representations  
of these elements: the primary and secondary virtual disk in the  
remote replication (and their locations) and the replication repository  
virtual disks in the storage array (and their locations). Or,  
View the physical components of the primary virtual disk by clicking  
View Associated Physical Components in the Properties pane. The  
View Associated Physical Components pop-up is displayed, showing a  
visual representation of the primary virtual disk in the remote  
replication.  
Configuration: Premium Feature— Remote Replication  
193  
 
Virtual Disk Status Icons  
In the Devices tab on the EMW, the following icons depict the status of each  
virtual disk:  
Changing Write Mode and Consistency Group Membership  
Table 13-1. Write Mode and Consistency Group Membership  
Icon  
Status  
Replication repository virtual disk  
Primary virtual disk  
Secondary virtual disk  
Primary virtual disk, synchronization in progress  
Primary virtual disk, replication suspended  
Secondary virtual disk, replication suspended  
Primary virtual disk, unsynchronized  
Secondary virtual disk, unsynchronized with communications loss  
194  
Configuration: Premium Feature— Remote Replication  
   
The write mode of a remote replication is selected when it is created.  
However, you can change the write mode in a remote replication at a later  
time, as well as change the secondary virtual disk's membership in a write  
consistency group. For more information about write modes and write  
consistency groups, go to Setting the Write Mode for Creating a Remote  
Replication online help topic.  
To change Write Mode and Consistency Group Membership:  
1
In the AMW the storage array containing the primary virtual disk, select  
the Logical tab.  
2
3
In the Logical pane, right-click the primary virtual disk of the replicated pair.  
Select ChangeWrite Mode  
.
The Change Write Mode dialog is displayed. The Replicated pairs table  
reflects all replicated pairs in both the local and remote storage arrays.  
4
Select one or more replicated pairs. To select all replicated pairs, click  
Select All  
.
5
6
Select either the Synchronous or Asynchronous write mode.  
If you are you adding the secondary virtual disk of the replicated pair to a  
write consistency group, select Yes and then select the Add to consistency  
group check box.  
If you are not adding the secondary virtual disk of the replicated pair to a  
write consistency group, go to Step 7.  
7
8
Click OK  
The Change Write Mode confirmation message is displayed.  
In the Change Write Mode confirmation message, click Yes  
.
.
The Replication properties section (the Properties panes in the Array  
Management Window for the local storage array) displays the  
following information:  
Replication status is Synchronized.  
Write mode is either Synchronous or Asynchronous, depending on your  
selection above.  
The secondary virtual disk is either write consistent/not write consistent.  
Resynchronization method is either Manual or Automatic.  
Configuration: Premium Feature— Remote Replication  
195  
Resynchronizing Virtual Disks  
on page 200.  
Automatic resynchronization—See "Automatically Resynchronizing  
Virtual Disks" on page 200.  
replications, see:  
"Normally Synchronized Virtual Disks" on page 196.  
"Unsynchronized Virtual Disks" on page 199.  
"Setting Synchronization Priority and Synchronization Method" on  
page 188.  
"Changing Synchronization Priority and Method" on page 197.  
"Resynchronizing Virtual Disks" on page 196.  
You might need to periodically test the communication between the primary  
virtual disk and the secondary virtual disk in a remote replication, especially  
after resynchronizing virtual disks. For information about testing the  
communication, go to Testing Communication between the Primary and  
Secondary Virtual Disk.  
Normally Synchronized Virtual Disks  
In a normally synchronized remote replication, RAID controller module  
owners manage the transfer of data from the primary to the secondary virtual  
disk. In a normal remote replication, these events occur:  
1
2
The primary virtual disk receives a write request from a host.  
The RAID controller module owner on the storage array logs information  
about the write operation to a replication repository virtual disk in the  
storage array.  
3
4
The RAID controller module owner writes the data to the primary virtual disk.  
The RAID controller module owner starts a data transfer operation to the  
secondary virtual disk on the secondary storage array.  
196  
Configuration: Premium Feature— Remote Replication  
   
The communication between a primary and secondary virtual disks can be  
suspended or become unsynchronized. If communication between the  
primary virtual disk and the secondary virtual disk is disrupted:  
1
2
3
4
5
The status of the replicated pair changes to Unsynchronized  
.
A
Needs Attention status is displayed for the storage array.  
Data is written to the primary virtual disk.  
Write requests to the primary virtual disk are logged.  
The RAID controller module owner sends an I/O completion message to  
the host sending the write request. Although the host can continue to send  
write requests to the primary virtual disk, no data transfer takes place to  
the secondary virtual disk. Writes to the secondary virtual disk are  
suspended pending restoration of communications between the primary  
virtual disk and the secondary virtual disk.  
When connectivity is restored between the primary virtual disk and the  
secondary virtual disk, the replicated pair is ready to be resynchronized.  
NOTE: When the primary virtual disk and the secondary virtual disk are  
resynchronized, only data that has changed on the primary virtual disk after the  
break in communication is transferred to the secondary virtual disk.  
CAUTION: You should be aware of a possible loss of data if communication is  
broken after resynchronization starts between the primary storage array and the  
secondary storage array. New data might mix with old data on the secondary  
virtual disk and render the data unusable in a disaster recovery situation.  
Changing Synchronization Priority and Method  
Synchronization priority defines how much processing time and resources are  
allocated to synchronizing the primary and secondary virtual disk of a remote  
replication relative to system performance. Increasing the synchronization  
priority of a remote replication might degrade system performance.  
Synchronization priorities can be reset at any time.  
Synchronization priorities can affect these operations:  
Performing a copyback  
Performing a Virtual Disk Expansion (VDE)  
Reconstructing a virtual disk  
Initializing a virtual disk  
Configuration: Premium Feature— Remote Replication  
197  
 
Changing the segment size of a virtual disk  
Defragmenting a disk group  
Adding free capacity to a disk group  
Changing the RAID level of a disk group  
To change the synchronization priority and method for an existing remote  
replication:  
1
In the AMW of the storage array containing the primary virtual disk of the  
replicated pair, right-click the Logical tab.  
2
Select ChangeSynchronization Settings.  
The Change Synchronization Settings dialog is displayed.  
3
4
5
In the Replicated pairs table, select the primary virtual disk and the remote  
virtual disk you want to change. To select all virtual disks, click Select All  
.
On the Select Synchronization Priority slide bar, select the  
synchronization priority for the replicated pair.  
Select either Manual or Automatic resynchronization.  
Automatic —Resynchronization starts immediately after  
communication is restored between unsynchronized replicated  
virtual disks.  
Manual—The replicated pair must be manually resynchronized each  
time communication is restored between unsynchronized replicated  
virtual disks.  
6
7
Click OK  
The Change Synchronization Settings confirmation message is displayed.  
In the Change Synchronization Settings confirmation message, click Yes  
.
.
The Change Synchronization Priority - Progress bar displays the progress of  
the resynchronization.  
8
Click OK.  
198  
Configuration: Premium Feature— Remote Replication  
Unsynchronized Virtual Disks  
Communication between a primary and secondary virtual disks can be either  
suspended, or become unsynchronized. If this communication between both  
virtual disks is disrupted:  
1
2
3
4
5
The status of the replicated pair changes to Unsynchronized  
.
A
Needs Attention status is displayed for the storage array.  
Data is written to the primary virtual disk.  
Write requests to the primary virtual disk are logged.  
The RAID controller module owner sends an I/O completion message to  
the host sending the write request. Although the host can continue to send  
write requests to the primary virtual disk, no data transfer takes place to  
the secondary virtual disk. Writes to the secondary virtual disk are  
suspended pending restoration of communications between the primary  
and secondary virtual disks.  
When connectivity is restored between the primary and secondary virtual  
disks, the replicated pair is ready to be resynchronized.  
NOTE: When the primary virtual disk and the secondary virtual disk are  
resynchronized, only data that has changed on the primary virtual disk after the  
break in communication is transferred to the secondary virtual disk.  
CAUTION: Loss of data can occur if communication is broken after  
resynchronization starts between the primary storage array and the secondary  
storage array. In addition to resulting in an incomplete resynchronization that may  
render the secondary storage array unusable, an interrupted resynchronization  
could cause a loss of data integrity on both arrays.  
Configuration: Premium Feature— Remote Replication  
199  
 
Automatically Resynchronizing Virtual Disks  
When automatic resynchronization is selected, the RAID controller module  
owning the primary virtual disk automatically starts resynchronizing the data  
on the remote replication pair immediately after communication is restored.  
NOTE: When choosing automatic resynchronization, a possible loss of data can  
occur if a resynchronization is interrupted in progress. Since another  
resynchronization begins automatically after communication is restored between  
the primary virtual disk and the remote virtual disk, data integrity could be affected.  
With automatic resynchronization set, you cannot add a secondary virtual  
disk to a write consistency group. Therefore, write consistency during the  
resynchronization process is not preserved. The write order is not consistent  
until the entire group achieves Optimal status.  
Manually Resynchronizing Virtual Disks  
When manual resynchronization is selected, you must manually  
resynchronize and resume the data transfer on a remote replication after  
communication is restored between the primary and remote virtual disks.  
Manual resynchronization is the recommended setting for all remote  
replications for these reasons:  
You determine when resynchronization starts, which allows you to better  
manage and/or mitigate impact on I/O performance.  
In a disaster recovery situation, manual resynchronization offers the best  
chance of retrieving valid data.  
When the secondary virtual disk is in a write consistency group, manual  
resynchronization preserves the write order.  
200  
Configuration: Premium Feature— Remote Replication  
   
Reversing Roles Between the Primary and Secondary Virtual Disks  
Reversing the roles between the virtual disks promotes the secondary virtual  
disk to the role of primary virtual disk and demotes the primary virtual disk to  
the role of secondary virtual. If the primary virtual disk in a remote virtual  
disk replication fails, you can reverse the roles of the primary virtual disk and  
the secondary virtual disk to transfer the data back to the restored virtual disk.  
CAUTION: A potential loss of data access can occur if you reverse roles between  
the secondary virtual disk and the primary virtual disk while a virtual disk copy is  
in progress. The role reversal may succeed, but the virtual disk copy fails and  
cannot be restarted.  
You cannot perform a virtual disk copy on a secondary virtual disk in a  
virtual disk remote replication. To create a virtual disk copy of a secondary  
virtual disk, you must reverse the roles of the secondary virtual disk and the  
primary virtual disk, then perform the virtual disk copy on the new primary  
virtual disk.  
While a virtual disk remote replication is synchronizing, you cannot  
perform a virtual disk copy on either the primary or secondary virtual disk.  
If you reverse roles between a secondary virtual disk with less capacity than  
the primary virtual disk, the role reversal succeeds but the usable capacity  
of the new primary virtual disk equals the total capacity of the secondary  
virtual disk.  
Configuration: Premium Feature— Remote Replication  
201  
 
Promoting the Secondary Virtual Disk or Demoting the Primary Virtual  
Disk  
To promote the secondary virtual disk to the role of primary virtual disk, or  
demote the primary virtual disk to the role of secondary virtual disk:  
1
In the AMW of the storage array that contains the virtual disk you are  
changing, click the Logical tab.  
2
Right-click the virtual disk you are changing.  
If you are promoting the secondary virtual disk to the role of primary  
virtual disk, select ChangeRole to Primary. The Change to Primary  
message is displayed. Clicking Yes reverses the roles of the primary and  
secondary virtual disks.  
If you are demoting the primary virtual disk to the role of secondary  
virtual disk, select ChangeRole to Secondary. The Change to  
Secondary message is displayed. Clicking Yes reverses the roles of the  
primary and secondary virtual disks.  
Suspending a Remote Replication  
1
In the AMW of the storage array with the primary virtual disk, select the  
Logical tab.  
2
In the Logical tab, right-click the primary virtual disk of a replicated pair,  
then select Suspend Replication  
.
The Suspend Replicated Pair dialog is displayed. The Replicated pairs  
table shows all replicated pairs in the local and remote storage arrays.  
3
4
Select one or more of the replicated pairs to suspend. To select all  
replicated pairs, click Select All  
.
Click Suspend.  
The Suspend Replication Relationship - Confirmation message is displayed.  
In the Confirmation text box, type Yes, then click OK  
5
.
The Suspend Replicated Pair - Progress bar indicates the progress of  
the suspension. The Properties pane in the AMW that contains the suspended  
primary virtual disk shows the Replication status as Suspended. The suspended  
icon is displayed next to both the primary virtual disk icon and the secondary  
virtual disk icon in the Logical pane in the AMW.  
202  
Configuration: Premium Feature— Remote Replication  
   
Resuming a Remote Replication  
1
In the AMW of the storage array with the primary virtual disk, select the  
Logical tab.  
2
In the Logical pane, right-click the primary virtual disk of the replicated  
pair, then select Resume Replication  
.
The Resume Replicated Pair dialog is displayed. The Replicated pairs table  
shows all suspended replicated pairs in the local and remote storage arrays.  
3
4
Select one or more replicated pairs. To select all replicated pairs, click  
Select All  
.
Click Resume  
.
The Resume Replication Pair - Confirmationmessage is  
displayed.  
5
In the Confirmation text box, type Yes  
.
The remote virtual disk replication resumes.  
The Properties panes in the AMW for the local storage array and the remote  
storage array show the Replication status as Synchronized for both the  
primary and secondary virtual disks.  
About Resumed Remote Replications  
When a remote virtual disk replication is suspended, data continues to write  
to the primary virtual disk, but is not replicated to the secondary virtual disk.  
Writes to the primary virtual disk are persistently logged to the replication  
repository virtual disks.  
After communications are restored, data transfer between the primary virtual  
disk and the secondary virtual disk must be resynchronized.  
Depending on which resynchronization method you specify, the following occurs:  
Automatic resynchronization—Data transfer automatically starts  
immediately after the virtual disks are resynchronized.  
Manual resynchronizationYou must manually resume the remote virtual  
disk replication to restart the data transfer. A suspended remote virtual disk  
replication stays in a Suspended status until it is manually resumed.  
Configuration: Premium Feature— Remote Replication  
203  
   
After the remote virtual disk replication resumes, data is automatically  
written to the secondary virtual disk. Only the regions of the primary virtual  
disk that changed since the replicated pair was suspended are written to the  
secondary virtual disk.  
CAUTION: Possible loss of data access: When you resume a remote virtual disk  
replication involving a primary or secondary virtual disk that is a member of a  
write consistency group, any other suspended remote replications for replicated  
pairs in the write consistency group also resumes.  
NOTE: If the write mode is synchronous, you do not need to resynchronize the  
primary and secondary virtual disk after resuming a remote virtual disk replication.  
Disks  
you might want to test the communication between the primary virtual disk  
and secondary virtual disk in a remote replication. For more information  
replications, go to these topics:  
"Unsynchronized Virtual Disks" on page 199.  
"Setting Synchronization Priority and Synchronization Method" on  
page 188.  
"Changing Synchronization Priority and Method" on page 197.  
"Resynchronizing Virtual Disks" on page 196.  
"Automatically Resynchronizing Virtual Disks" on page 200.  
To test the communication between virtual disks in a remote replication:  
1
In the AMW of either the primary or secondary virtual disk, select the  
Logical tab.  
2
3
In the Logical pane, right-click the virtual disk.  
Select Test Replication Communication  
.
The Replication Communication Test Progress message is displayed.  
204  
Configuration: Premium Feature— Remote Replication  
 
Deleting a Virtual Disk From a Replicated Pair in a Storage Array  
Follow these steps to delete either a primary virtual disk, a secondary virtual  
disk, or both virtual disks from a replicated pair in a storage array.  
NOTE: Do not remove a replication relationship to back up a replicated virtual disk. To  
perform backups of either the primary virtual disk or the secondary virtual disk, suspend  
the remote virtual disk replication so that the replication relationship is not broken.  
Deleting a Primary Virtual Disk  
CAUTION: Depending on which premium features are enabled on the storage  
array, deleting a primary virtual disk might delete all associated virtual disks,  
resulting in a possible loss of data on those virtual disks.  
NOTE: You cannot delete a primary virtual disk while it is synchronizing.  
When a primary virtual disk is deleted from a remote virtual disk replication,  
the following occurs:  
The primary virtual disk is deleted from the storage array.  
NOTE: The virtual disk is permanently deleted from the storage array and all  
data on the primary virtual disk is permanently lost.  
The replication relationship breaks.  
The capacity of the deleted virtual disk becomes unconfigured free capacity in  
the storage array and is available for creation of new virtual disks.  
The secondary virtual disk becomes a regular, standard virtual disk and is  
can accept both reads and writes.  
For instructions on how to delete a primary virtual disk, go to Deleting a  
Primary Virtual Disk in a Replicated Pair from a Storage Array online  
help topic.  
Configuration: Premium Feature— Remote Replication  
205  
   
Deleting a Secondary Virtual Disk  
NOTE: Depending on which premium features are enabled on the storage array,  
deleting a secondary virtual disk might delete all associated virtual disks, resulting  
in a possible loss of data on those virtual disks.  
NOTE: You cannot delete a secondary virtual disk while it is synchronizing.  
When a secondary virtual disk is deleted, the replication relationship is  
removed and the remote virtual disk replication is destroyed.  
NOTE: Deleting a secondary virtual disk results in the permanent loss of the data  
on the secondary virtual disk.  
For steps describing how to delete a secondary virtual disk, see "Deleting a  
Virtual Disk From a Replicated Pair in a Storage Array" on page 205.  
Deleting a Primary Virtual Disk in a Replicated Pair From a Storage  
Array  
1
2
3
Stop all I/O activity to the primary virtual disk and unmount any file  
systems on the primary virtual disk.  
In the AMW of the storage array containing the primary virtual disk, select  
the Logical tab.  
In the Logical pane, right-click the primary virtual disk, then select Delete  
.
The Delete Virtual Disks dialog is displayed.  
4
5
6
Select one or more virtual disks to delete, then click Delete.  
The Confirm Delete Virtual Disk(s) message is displayed.  
In the confirmation text box, type Yes. Then, click OK  
.
The Delete Virtual Disks - Progress bar is displayed.  
When the deletion is complete, click OK  
.
The primary virtual disk is deleted from the storage array. The secondary  
virtual disk in the replicated pair now becomes a regular standard virtual disk  
in the storage array.  
NOTE: The primary virtual disk is deleted from the storage array and all data on the  
virtual disk is permanently lost.  
206  
Configuration: Premium Feature— Remote Replication  
   
Deleting a Secondary Virtual Disk in a Replicated Pair From a Storage Array  
1
2
3
Stop all I/O activity on the secondary virtual disk and unmount any file  
systems on the secondary virtual disk.  
In the AMW of the storage array that contains the secondary virtual disk,  
select the Logical tab.  
In the Logical pane, right-click the secondary virtual disk, then select Delete  
.
The Delete Virtual Disks dialog is displayed.  
4
5
6
Select one or more virtual disks to delete, then click Delete.  
The Confirm Delete Virtual Disk(s) message is displayed.  
In the confirmation text box, type Yes. Then, click OK  
.
The Delete Virtual Disks - Progress bar is displayed.  
When the deletion is complete, click OK  
.
The replication relationship is removed and the remote virtual disk  
replication is destroyed.  
NOTE: The secondary virtual disk is deleted from the storage array and all data on  
the virtual disk is permanently lost.  
Removing a Remote Replication From a Storage Array  
Removing a remote virtual disk replication from a storage array returns both  
the primary virtual disk and the secondary virtual disk to regular standard  
virtual disks. After the removal is complete, normal I/O operations continue  
on the former primary virtual disk; the former secondary virtual disk also  
becomes available for normal I/O operations. Both virtual disks become read-  
write enabled. Unless one of disks has been deleted, a replication relationship  
between the two virtual disks can be re-created.  
NOTE: No data on either virtual disk is deleted.  
CAUTION: Do not remove a replication relationship to back up a replicated  
virtual disk. To back up either the primary virtual disk or the secondary virtual  
disk, suspend the remote virtual disk replication so that the replication  
relationship is not broken.  
Configuration: Premium Feature— Remote Replication  
207  
   
To remove remote replication from a storage array:  
1
In the AMW of the storage array containing the primary virtual disk, select  
the Logical tab.  
2
In the Logical pane, right-click the primary virtual disk of a replicated pair,  
then select Remove Replication Relationship  
.
The Remove Replication Relationship dialog is displayed. The Replicated  
pairs table shows all replicated pairs in both the local and remote storage array.  
3
4
To remove a replication relationship, select one or more replicated pairs.  
To select all replicated pairs, click Select All  
Click Remove  
The Remove Replication Relationship - Confirmation message is displayed.  
Click Yes  
.
.
5
.
The Remove Replicated Pair - Progress bar shows the progress of the  
removal process.  
Deactivating the Remote Replication on the Storage Array  
Before you can permanently disable the Remote Replication premium feature  
you must deactivate it on the storage array, see "Disabling the Remote  
Replication Premium Feature" on page 209. Deactivating the Remote  
Replication premium feature on this storage array does not affect remote  
virtual disk replications or the Remote Replication premium features on other  
storage arrays. However, another storage array cannot use this storage array as  
a remote storage array for creating a separate remote virtual disk replication.  
To deactivate remote replication in a storage array:  
1
In the Array Management Window, select Storage ArrayRemote  
ReplicationDeactivate  
The Deactivate Remote Replication confirmation message is displayed.  
.
2
Click Yes  
.
The Remote Replication premium feature is deactivated and the two  
replication repository virtual disks are deleted from the storage array.  
208  
Configuration: Premium Feature— Remote Replication  
 
Disabling the Remote Replication Premium Feature  
Before disabling the Remote Replication premium feature, all remote virtual  
disk replications must have been removed on the storage array. See "Disabling  
the Remote Replication Premium Feature" on page 209.  
NOTE: To re-enable the Remote Replication premium feature on a storage array  
after it has been disabled, you must either retrieve the premium feature key or  
obtain a new one from your Dell support representative.  
To disable the remote replication premium feature:  
1
In the Array Management Window, select Storage Array >  
Premium Features  
.
The Premium Features and Feature Pack Information screen is displayed.  
2
3
Select Remote Replication in the Premium Features window, then  
click Disable  
.
A warning message is displayed.  
If you are sure you want to disable the Remote Replication feature, click  
Yes to confirm.  
The Remote Replication premium feature is disabled.  
Configuration: Premium Feature— Remote Replication  
209  
 
210  
Configuration: Premium Feature— Remote Replication  
14  
Configuration: Device Mapper  
Multipath for Linux  
Overview  
The MD3600f Series storage array uses a Linux operating system software  
framework, known as Device Mapper (DM), to enable multipath capabilities  
on Linux Host Servers. The DM multipath functionality is provided by a  
combination of drivers and utilities. This chapter describes how to use those  
utilities to complete the process of enabling an MD3600f Series storage array  
on a Linux system.  
The Device Mapper technology replaces an earlier, proprietary technology,  
known as MPP. MPP was used to enable multipathing for a previous MD  
generation, the MD3000 Series of storage arrays.  
NOTE: The required Device Mapper software components are installed on a Linux  
host server by running the MD3600f Series Resource media installation program on  
the server, and selecting either the Full or Host install option. For detailed  
installation procedures see the Dell PowerVault MD3600f and MD3620f storage  
arrays Deployment Guide.  
Benefits of using DM Multipath include:  
Detects path failure and re-routes I/O to other available paths  
Revalidates failed paths after path restoration  
Utilizes multiple available paths to maximize performance  
Reconfigures path usage based on path states and error conditions  
Unifies multiple device nodes into a single logical multipath device node  
Indentifies a new multipathed LU and automatically configures a new  
multipath node  
Provides device name persistency for DM devices under /dev/mapper/  
Configuration: Device Mapper Multipath for Linux  
211  
     
Using DM Multipathing Devices  
NOTE: Using or modifying any nodes other than the multipathing device nodes can  
result in array or file system problems, including loss of communication with the  
array and corruption of the file system. Avoid accessing any device other than the  
multipathing device.  
NOTE: After creating a partition on a multipathing device, all IO operations,  
including filesystem creation, raw I/O and file system I/O, must be done through the  
partition node, not the multipathing device nodes.  
Prerequisites  
The following tasks must be completed before proceeding. For more  
information about step 1and step 2, see the MD3600f and MD3620f storage  
arrays Deployment Guide. For more information about step 3, see "Creating  
Virtual Disks" on page 102.  
1
Install the host software from MD3600f Series resource media—Insert  
the resource media to start the installation of Modular Disk Storage  
Manager (MDSM).  
NOTE: Red Hat install of 5.x needs a remount of the DVD media to make  
contents executable.  
2
3
Reboot when prompted by the install program—The installation program  
prompts for and needs a reboot at completion of install.  
Create and map virtual disks using MDSM—After configuring the arrays  
run the MDSM to create and map virtual disks.  
Using the MDSM software:  
1
2
3
Map the host server to the MD3600f Series storage array.  
Create the Virtual Disks.  
Map newly created arrays to your host server.  
NOTE: Any array configured with MDCU automatically adds to the list of  
Devices in the PowerVault Modular Disk Storage Manager Enterprise  
Management Window (EMW).  
212  
Configuration: Device Mapper Multipath for Linux  
         
Device Mapper Configuration Steps  
available to the Linux host server:  
1
2
3
4
5
6
Scan for virtual disks. See "Scan for Newly Added Virtual Disks" on  
page 213.  
Topology Using the Multipath Command" on page 214.  
Create a partition on a multipath device node. See "Create a New fdisk  
Partition on a Multipath Device Node" on page 215.  
Add a partition to DM. See "Add a New Partition to Device Mapper" on  
page 215.  
Create a file system on a DM partition. See "Create a File System on a  
Device Mapper Partition" on page 216.  
Mount a DM partition. See "Mount a Device Mapper Partition" on  
page 216.  
NOTE: In the following command descriptions <x> is used to indicate where a  
substitution should be made. On RHEL systems <x> is the number assigned to the  
device. On SLES systems <x> is the letter(s) assigned to the device.  
Scan for Newly Added Virtual Disks  
The rescan_dm_devs command scans the host server system looking for  
existing and newly added virtual disks mapped to the host server.  
# rescan_dm_devs  
If an array virtual disk (VD) is mapped to the host server at a later time the  
rescan_dm_devicescommand must be run again to make the VD a  
visible LUN to the operating system.  
Configuration: Device Mapper Multipath for Linux  
213  
   
Display the Multipath Device Topology Using the Multipath Command  
The multipath command adds newly scanned and mapped virtual disks to the  
Device Mapper tables and creates entries for them in the /dev/mapper  
directory on the host server. These devices are the same as any other block  
devices in the host.  
To list all the multipath devices run the following command.  
# multipath –ll  
The output should be similar to this example, which shows the output for one  
mapped virtual disk.  
mpath1 (3600a0b80005ab177000017544a8d6b92) dm-0 DELL,  
MD36xxf  
[size=5.0G][features=3 queue_if_no_path  
pg_init_retries 50][hwhandler=1 rdac][rw]  
\_ round-robin 0 [prio=6][active]  
\_ 5:0:0:0 sdc 8:32 [active][ready]  
\_ round-robin 0 [prio=1][enabled]  
\_ 4:0:0:0 sdb 8:16 [active][ghost]  
where:  
mpath1is the name of the virtual device created by device mapper. It is  
located in the /dev/mapper directory.  
DELLis the vendor of the device  
MD36xxfis the model of the device  
Sdcis the physical path to the owning controller for the device  
Sdbis the physical path to the non-owning controller for the device  
The following is an example of SLES output:  
mpathb(360080e500017b2f80000c6ca4a1d4ab8) dm-21  
DELL,MD36xxf  
[size=1.0G][features=3 queue_if_no_path  
pg_init_retries 50][hwhandler=1 rdac][rw]  
\_ round-robin 0 [prio=6][active]  
\_ 4:0:0:22 sdx 65:112 [active][ready]  
214  
Configuration: Device Mapper Multipath for Linux  
 
\_ round-robin 0 [prio=1][enabled]  
\_ 6:0:0:22 sdcl 69:144 [active][ghost  
where:  
mpathbis the name of the virtual device created by device mapper. It is  
located in the /dev/mapper directory.  
DELLis the vendor of the device  
MD36xxfis the model of the device  
Sdxis the physical path to the owning controller for the device  
Sdclis the physical path to the non-owning controller for the device  
Create a New fdisk Partition on a Multipath Device Node  
The fdisk command allows you to create partition space for a file system on  
the newly scanned and mapped virtual disks that have been presented to  
Device Mapper.  
To create a partition with the multipathing device nodes  
/dev/mapper/mpath<x>, for example, use the following command, where  
mpath<x> is the multipathing device node on which you want to create  
the partition:  
# fdisk /dev/mapper/mpath<x>  
NOTE: The <x> value is an alphanumeric operating system dependent format. The  
corresponding value for mapped virtual disks can be seen using the previously run  
multipathcommand. Please refer to your operating system documentation for  
additional information on fdisk usage.  
Add a New Partition to Device Mapper  
The kpartxcommand adds the newly fdisk created partition to the Device  
Mapper list of usable partitions. See examples below, where mpath<x> is the  
device node on which the partition was created.  
# kpartx –a /dev/mapper/mpath<x>  
If successful, the command does not show an output. To verify success and  
view exact partition naming, the user can use these commands to see the full  
partition names assigned.  
# cd /dev/mapper  
# ls  
Configuration: Device Mapper Multipath for Linux  
215  
   
The following examples provide information about the general mapping formats:  
On RHEL hosts, a partition node has the format  
/dev/mapper/mpath<x>p<y>, where <x> is the alphabetic number for the  
multipathing device, <y> is the partition number for this device.  
On SLES 11.x hosts, a partition node has the format  
/dev/mapper/mpath<x>-part<y>, where <x> is letter(s) assigned to the  
multipathing device and <y> is the partition number.  
On SLES 10.3 hosts, a partition node has the format  
/dev/mapper/mpath<x>_part<y>, where <x> is the letter(s) assigned to  
the multipathing device and <y> is the partition number.  
NOTE: After creating a partition on a multipathing device, all I/O operations,  
including file system creation, raw I/O and file system I/O, must be done through the  
partition node, and not through the multipathing device nodes.  
Create a File System on a Device Mapper Partition  
Use the standard mkfscommand to create the file system on the newly  
created Device Mapper partition  
For example:  
# mkfs –t <filesystem type> /dev/mapper/<partition  
node>  
where <partition node> is the partition on which the file system is created.  
Mount a Device Mapper Partition  
Use the standard mountcommand to mount the Device Mapper partition, as  
shown below.  
# mount /dev/mapper/<partition_node> <mounting_point>  
Ready for Use  
The newly created virtual disks created on the MD3600f Series array are now  
setup and ready to be used. Future reboots automatically finds multipathing  
devices along with their partitions.  
216  
Configuration: Device Mapper Multipath for Linux  
   
Important Information About Special Partitions  
When using Device Mapper with the MD3600f Series array, all physical disks  
are assigned a disk device node. This includes a special device type used for  
in-band management of the MD3600f Series array, known as the Access Disk  
or Universal Xport device.  
CAUTION: Certain commands, such as lsscsi, displays one or more instances of  
Universal Xport devices. These device nodes must never be accessed, mounted, or  
used in any way. Doing so could cause loss of communication to the storage array  
and possibly cause serious damage to the storage array, potential making data  
stored on the array inaccessible.  
Only multipathing device nodes and partition nodes created using the  
directions provided above should be mounted or in any way accessed by the  
host system or its users.  
Table 14-1. Device Mapper Commands  
Command  
Descriptions  
multipath -h  
multipath -ll  
Prints usage information.  
Displays the current multipath topology using all  
available information (sysfs, the device mapper,  
path checkers, and so on).  
multipath  
Re-aggregates multipathing device with  
simplified output.  
multipath -f  
<multipath_dev_node>  
Flushes out Device Mapper for the specified  
multipathing device. Used if the underlying  
physical devices are deleted/unmapped.  
multipath -F  
Flushes out all unused multipathing device maps.  
rescan_dm_devs  
Dell provided script. Forces a rescan of the host's  
SCSI bus and aggregates multipathing devices as  
needed. For use when:  
• LUNs are dynamically mapped to the hosts.  
• New targets are added to the host.  
• Failback of the storage array is required.  
Configuration: Device Mapper Multipath for Linux  
217  
Limitations and Known Issues  
In certain error conditions with the no_path_retry or the  
queue_if_no_path feature is set, applications may hang. To overcome  
these conditions the you must enter the following command for each  
affected multipath device:  
dmsetup message [device] 0 "fail_if_no_path"  
where [device]is the multipath device name (for example. mpath2; do  
not specify the path)  
I/O may hang when a Device Mapper device is deleted before the volume  
is unmounted.  
If the scsi_dh_rdac module is not included in initrd, slower device  
discovery may be seen and the syslog may become populated with buffer  
I/O error messages.  
I/O may hang if the host server or storage array is rebooted while I/O is  
active. All I/O to the storage array should be stopped before shutting down  
or rebooting the host server or storage array.  
With the MD3600f Series array, after a failed path is restored, failback does not  
automatically occur because the driver cannot auto-detect devices without a  
forced rescan. Run the command rescan_dm_devsto force a rescan of the  
host server. This restores the failed paths enabling failback to occur.  
Failback can be slow when the host system is experiencing heavy I/O. The  
problem is exacerbated if the host server is also experiencing very high  
CPU utilization.  
The Device Mapper Multipath service can be slow when the host system is  
experiencing heavy I/O. The problem is exacerbated if the host server also  
experiences very high CPU utilization.  
If the root disk is not blacklisted in the multipath.conf file, a multipathing  
node may be created for the root disk. The command multipath –ll  
lists vendor/product ID which can help identify this issue.  
218  
Configuration: Device Mapper Multipath for Linux  
 
Troubleshooting  
Question  
Answer  
How can I check if multipathd Run the following command  
is running?  
/etc/init.d/multipathd status  
Why does the multipath  
–ll command output not  
show any devices?  
First verify if the devices are discovered or not. The  
command #cat /proc/scsi/scsidisplays all  
the devices that are already discovered.  
Then verify the multipath.conf to ensure that it is  
been updated with proper settings.  
After this, run multipath. Then run multipath  
–ll, the new devices should be displayed.  
Why is a newly-mapped LUN Run “rescan_dm_devs” in any directory. This should  
not assigned a multipathing  
device node?  
bring up the devices.  
I have no LUNs mapped before. Run rescan_dm_devsinstead of rescan-  
Then I map a few LUNs. After scsi-busfor LUN 0 reconfiguration.  
running rescan-scsi-  
bus.sh, LUN 0 does not  
show up.  
I removed a LUN, but the  
multipathing mapping is  
still present.  
The multipathing device is still there after you  
remove the LUNs. Run multipath –f <device  
node for the deleted LUN> to remove the  
multipathing mapping. For example, if device  
devices related with /dev/dm-1 is deleted, you must  
run multipath –f /dev/dm-1to remove  
/dev/dm-1 from DM mapping table. If multipathing  
daemon is stopped/restarted, run multipath –F  
to flush out all stale mappings.  
Failback does not happen as  
expected with the array.  
Sometimes the low level driver cannot auto-detect  
devices coming back with the array. Run  
rescan_dm_devsto do a rescan of the host server  
SCSI bus and re-aggregate devices at  
multipathing layer.  
Configuration: Device Mapper Multipath for Linux  
219  
 
220  
Configuration: Device Mapper Multipath for Linux  
15  
Management: Installing Array  
Components  
Recommended Tools  
You may need the following items to perform the procedures in this section:  
Key to the system keylock  
#2 Phillips screwdriver  
Wrist grounding strap  
Management: Installing Array Components  
221  
       
Front Bezel (Optional)  
Removing the Front Bezel  
1
2
3
4
Using the system key, unlock the front bezel (if locked).  
Lift up the release latch next to the keylock.  
Rotate the left end of the bezel away from the front panel.  
Unhook the right end of the bezel and pull the bezel away from  
the system.  
Figure 15-1. Removing and Installing the Front Bezel  
1
2
3
4
1
3
bezel  
2
4
keylock  
release latch  
hinge tab  
Installing the Front Bezel  
1
2
3
Hook the right end of the bezel onto the chassis.  
Fit the free end of the bezel onto the system.  
Secure the bezel with the keylock. See Figure 15-1.  
222  
Management: Installing Array Components  
               
Hard Drives  
SAFETY: Models AMT, E03J, and E04J  
Models AMT, E03J, and E04J are intended for installation only in restricted  
access locations as defined in cl 1.2.7.3 of IEC 60950-1:2005.  
Depending on your configuration, your array either supports up to 24 2.5-inch  
SAS hard drives or up to 12 3.5" SAS hard drives in internal drive bays. Hard  
drives are connected to a backplane through hard-drive carriers and can be  
configured as hot-swappable.  
Removing a Drive Blank  
CAUTION: To maintain proper system cooling, all empty hard-drive bays must  
have drive blanks installed.  
1
2
If installed, remove the front bezel. See "Removing the Front Bezel" on  
page 222.  
Press the release tab and slide the drive blank out until it is free of the  
drive bay. See Figure 15-2 for PowerVault MD3600f and Figure 15-3 for  
PowerVault MD3620f.  
Figure 15-2. Removing and Installing a 3.5-Inch Hard-Drive Blank  
1
2
1
drive blank  
2
release tab  
Management: Installing Array Components  
223  
       
Figure 15-3. Removing and Installing a 2.5-Inch Hard-Drive Blank  
1
2
1
drive blank  
2
release tab  
Installing a Drive Blank  
1
If installed, remove the front bezel. See "Removing the Front Bezel" on  
page 222.  
2
3
4
Insert the drive blank into the drive bay until the blank is fully seated.  
Close the handle to lock the blank in place.  
If applicable, replace the front bezel. See "Installing the Front Bezel" on  
page 222.  
Removing a Hard Drive  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came with  
the product.  
1
2
If installed, remove the front bezel. See "Removing the Front Bezel" on  
page 222.  
From the Modular Disk Storage Manager (MDSM) software, prepare the  
drive for removal. Wait until the hard-drive indicators on the drive  
carrier signal that the drive can be removed safely. See your controller  
documentation for information about hot-swap drive removal.  
224  
Management: Installing Array Components  
           
If the drive has been online, the green activity/fault indicator flashes as  
the drive is powered down. When the drive indicators are off, the drive is  
ready for removal.  
3
4
Press the release button to open the drive carrier release handle. See  
Figure 15-4.  
Slide the hard drive out until it is free of the drive bay.  
CAUTION: To maintain proper system cooling, all empty hard-drive bays must  
have drive blanks installed.  
5
6
Insert a drive blank in the empty drive bay. See "Installing a Drive Blank"  
on page 224.  
If applicable, replace the front bezel. See "Installing the Front Bezel" on  
page 222.  
Figure 15-4. Removing and Installing a Hard Drive  
1
2
1
release button  
2
hard-drive carrier handle  
Management: Installing Array Components  
225  
 
Installing a Hard Drive  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is  
not covered by your warranty. Read and follow the safety instructions that came  
with the product.  
CAUTION: Use only hard drives that have been tested and approved for use with  
the MD3600f series.  
CAUTION: When installing a hard drive, ensure that the adjacent drives are fully  
installed. Inserting a hard-drive carrier and attempting to lock its handle next to a  
partially installed carrier can damage the partially installed carrier's shield spring  
and make it unusable.  
1
2
If applicable, remove the front bezel. See "Removing the Front Bezel" on  
page 222.  
If applicable, remove the drive blank from the bay. See "Removing a Drive  
Blank" on page 223.  
3
4
Press the release button to open the drive carrier release handle.  
Insert the hard-drive carrier into the drive bay until the carrier contacts  
the backplane.  
5
Close the handle to lock the drive in place.  
226  
Management: Installing Array Components  
     
Removing a Hard Drive From a Hard-Drive Carrier  
Remove the screws from the slide rails on the hard-drive carrier and separate  
the hard drive from the carrier. See Figure 15-5 for PowerVault MD3600f and  
Figure 15-6 for PowerVault MD3620f.  
Figure 15-5. Removing and Installing a Hard Drive Into a 3.5-Inch Drive Carrier  
4
1
3
2
1
3
screws (4)  
2
4
drive carrier  
hard drive  
SAS screw hole  
Management: Installing Array Components  
227  
     
Figure 15-6. Removing and Installing a Hard Drive Into a 2.5-Inch Drive Carrier  
4
1
3
2
1
3
screws (4)  
2
4
drive carrier  
hard drive  
SAS screw hole  
228  
Management: Installing Array Components  
 
Installing a Hard Drive Into a Drive Carrier  
1
Insert the hard drive into the hard-drive carrier with the connector end  
of the drive at the back. See Figure 15-5.  
2
Align the screw holes on the hard drive with the back set of holes on the  
hard-drive carrier.  
When aligned correctly, the back of the hard drive is flush with the back of  
the hard-drive carrier.  
3
Attach the four screws to secure the hard drive to the hard-drive carrier.  
RAID Controller Module  
An MD3600f series storage array supports single as well as dual RAID  
controller configurations. If only one RAID controller module is installed in  
your array, it must be installed in slot 0. You must install the RAID controller  
module blank in slot 1.  
CAUTION: RAID controller modules can be removed and installed without turning  
off the array. It is recommended that you do not remove the RAID controller module  
while data is being transferred. Replacing or installing a RAID controller module  
that is connected to a host server causes it to loose communication with the array  
and may require a reboot of the host server.  
Removing a RAID Controller Module Blank  
module blank in the empty slot.  
1
2
3
Turn off the array and host server.  
Disconnect all the power cables connected to the array.  
To remove the RAID controller module blank, press down on the release  
latch and pull the blank away from the array. See Figure 15-7.  
4
Install RAID controller modules in slot 0 and 1. See "Installing a RAID  
Controller Module" on page 232.  
Management: Installing Array Components  
229  
         
5
6
Connect all the power cables to the array.  
Turn on the array and the host server.  
Figure 15-7. Removing and Installing a RAID controller module Blank  
1
2
1
release latch  
2
RAID controller module blank  
Installing a RAID Controller Module Blank  
To install a RAID controller module blank, align the blank with the RAID  
controller module bay and insert the blank into the chassis until it clicks  
into place.  
230  
Management: Installing Array Components  
     
Removing a RAID Controller Module  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
CAUTION: If your configuration uses fiber optic cables, remember that they are  
fragile. Bending, twisting, folding, or pinching fiber optic cables can cause  
damage to the cables, degraded performance, or loss of data. To prevent damage,  
do not twist, fold, pinch, or step on the cables. Do not bend the cables in less than  
a 5-cm (2") radius.  
1
If small form-factor pluggable (SFP) transceivers are present, record the  
ports in which they are installed, and remove them.  
2
3
Take the desired RAID controller module offline.  
Disconnect the cables connected to the RAID controller module.  
If fiber optic cables are present, it might be necessary to partially remove the  
RAID controller module using the two release levers. Opening the release  
levers makes it easier to press down the fiber optic cable release tab. If the  
system is still running, do not disturb the second RAID controller module.  
4
5
Push down on the release tab and pull the release lever away from the  
chassis. See Figure 15-8.  
Grasp the release lever and pull the module away from the chassis.  
NOTE: To avoid damage to the sensitive EMI contacts on the RAID controller  
module, do not stack RAID controller modules.  
Management: Installing Array Components  
231  
   
Figure 15-8. Removing and Installing a RAID controller module  
3
2
1
1
3
RAID controller module  
release lever  
2
release tab  
Installing a RAID Controller Module  
WARNING: Fiber optic cables are fragile. Bending, twisting, folding, or pinching  
fiber optic cables can cause damage to the cables, degraded performance, or loss  
of data. To prevent damage, do not twist, fold, pinch, or step on the cables. Do not  
bend the cables in less than a 5 cm (2") radius.  
CAUTION: Many repairs may only be done by a certified service technician. You  
should only perform troubleshooting and simple repairs as authorized in your product  
documentation, or as directed by the online or telephone service and support team.  
Damage due to servicing that is not authorized by Dell is not covered by your warranty.  
Read and follow the safety instructions that came with the product.  
1
2
Insert the RAID controller module into the RAID controller module bay  
until it seats into place.  
Push the release lever toward the chassis until it clicks into place.  
232  
Management: Installing Array Components  
       
3
4
Connect all the cables to the RAID controller module. If SFP transceivers  
and fiber optic cables were present, reinstall them in the correct ports.  
If applicable, update the firmware for the RAID controller module. For  
information about the latest firmware, see the Dell Support website at  
support.dell.com  
.
Opening the RAID Controller Module  
CAUTION: Many repairs may only be done by a certified service technician. You  
documentation, or as directed by the online or telephone service and support team.  
Read and follow the safety instructions that came with the product.  
1
2
3
Turn off the array and host server.  
Disconnect all the power cables connected to the array.  
Remove the RAID controller module. See "Removing a RAID Controller  
Module Blank" on page 229.  
4
5
Remove the screws from the sides of the RAID controller module. See  
Figure 15-9.  
While pressing the indent, slide the cover in the direction of the arrow and  
lift it away from the RAID controller module. See Figure 15-9.  
Management: Installing Array Components  
233  
 
Figure 15-9. Opening and Closing the RAID Controller Module  
3
2
1
1
3
screws (2)  
indent  
2
RAID controller module  
Closing the RAID Controller Module  
CAUTION: Many repairs may only be done by a certified service technician. You  
should only perform troubleshooting and simple repairs as authorized in your product  
documentation, or as directed by the online or telephone service and support team.  
Damage due to servicing that is not authorized by Dell is not covered by your warranty.  
Read and follow the safety instructions that came with the product.  
1
Place the cover onto the RAID controller module and offset it slightly  
towards the back, so that the hooks on the cover fit over the corresponding  
slots on the RAID controller module.  
2
3
4
Slide the cover toward the front till it snaps into place. See Figure 15-9.  
Replace the screws on the RAID controller module. See Figure 15-9.  
Connect all the cables to the array. If SFP transceivers and fiber optic  
cables were previously in place, reinstall them in the correct ports.  
5
Turn on the array and the host server.  
234  
Management: Installing Array Components  
   
RAID Controller Module Backup Battery Unit  
Removing the RAID Controller Module Backup Battery Unit  
CAUTION: Many repairs may only be done by a certified service technician. You  
should only perform troubleshooting and simple repairs as authorized in your product  
documentation, or as directed by the online or telephone service and support team.  
Read and follow the safety instructions that came with the product.  
1
2
3
Turn off the array and host server.  
Disconnect all the cables connected to the array.  
Remove the RAID controller module. See "Removing a RAID Controller  
Module" on page 231.  
4
5
6
Open the RAID controller module. See "Opening the RAID Controller  
Module" on page 233.  
Loosen the screw that secures the backup battery unit to the RAID  
controller module. See Figure 15-10.  
Slide the backup battery unit in the direction of the arrow and lift it out of  
the RAID controller module. See Figure 15-10.  
Figure 15-10. Removing and Installing the RAID Controller Module Backup Battery Unit  
1
2
1
backup battery unit  
2
screw  
Management: Installing Array Components  
235  
       
Installing the RAID Controller Module Backup Battery Unit  
CAUTION: Many repairs may only be done by a certified service technician. You  
should only perform troubleshooting and simple repairs as authorized in your product  
documentation, or as directed by the online or telephone service and support team.  
Damage due to servicing that is not authorized by Dell is not covered by your warranty.  
1
2
Align the backup battery unit with the slots on the RAID controller module.  
controller module.  
3
4
5
Tighten the screw that secures the backup battery unit to the RAID  
controller module.  
Close the RAID controller module. See "Closing the RAID Controller  
Module" on page 234.  
Replace the RAID controller module. See "Installing a RAID Controller  
Module" on page 232.  
6
7
Connect all the cables to the array.  
Turn on the array and the host server.  
236  
Management: Installing Array Components  
 
Power Supply/Cooling Fan Module  
The array supports two hot-swappable power supply/cooling fan modules.  
While the array can operate temporarily with one module, both the modules  
must be present for proper system cooling.  
CAUTION: A single power supply/cooling fan module can be removed from a  
powered-on array for a maximum period of five minutes. Beyond that time, the  
array may automatically shut down to prevent damage.  
Removing a Power Supply/Cooling Fan Module  
CAUTION: Many repairs may only be done by a certified service technician. You  
should only perform troubleshooting and simple repairs as authorized in your product  
documentation, or as directed by the online or telephone service and support team.  
Damage due to servicing that is not authorized by Dell is not covered by your warranty.  
Read and follow the safety instructions that came with the product.  
NOTE: If you remove a fully functioning power supply/cooling fan module, the fan  
speed in the remaining module increases significantly to provide adequate cooling.  
The fan speed decreases gradually when a new power supply/cooling fan module  
is installed.  
1
2
3
Turn off the power supply/cooling fan module.  
Disconnect the power cable from the power source.  
Remove the velcro straps that secure the power cable and then disconnect  
the power cable from the power supply/cooling fan module.  
WARNING: The power supply/cooling fan modules are heavy. Use both hands  
while removing the module.  
4
Press the release tab and pull the power supply out of the chassis.  
Management: Installing Array Components  
237  
       
Figure 15-11. Removing and Installing a Power Supply/Cooling Fan Module  
1
2
3
1
3
release tab  
2
power supply  
power supply handle  
Installing a Power Supply/Cooling Fan Module  
CAUTION: Many repairs may only be done by a certified service technician. You  
should only perform troubleshooting and simple repairs as authorized in your product  
documentation, or as directed by the online or telephone service and support team.  
Damage due to servicing that is not authorized by Dell is not covered by your warranty.  
Read and follow the safety instructions that came with the product.  
1
2
3
Slide the power supply/cooling fan module into the chassis until it is fully  
seated and the release tab clicks into place. See Figure 15-11.  
Connect the power cable to the power supply/cooling fan module and plug  
the cable into a power outlet.  
Secure the power cable using the velcro strap. See Figure 15-12.  
238  
Management: Installing Array Components  
     
Figure 15-12. Securing the Power Cable  
1
1
velcro strap  
CAUTION: When connecting the power cable, secure the cable with the velcro strap.  
NOTE: If the array is powered on, all the power supply LEDs remain off until the AC  
power cable is connected to the power supply/cooling fan module and the power  
switch is turned on.  
4
Turn on the power supply/cooling fan module.  
Control Panel  
Removing the Control Panel  
1
2
3
Turn off the array and host server.  
Disconnect all the power cables connected to the array.  
Remove the hard drives from:  
slots 0 to 2 in PowerVault MD3600f  
slots 0 to 5 in PowerVault MD3620f  
See "Removing a Hard Drive" on page 224.  
NOTE: Mark each hard drive with it’s slot position as you remove it.  
Management: Installing Array Components  
239  
         
4
Slide the control panel out of the chassis after:  
Pushing the release tab toward the front of the array in PowerVault  
MD3600f. See Figure 15-13.  
Pulling the release pin toward the front of the array in PowerVault  
MD3620f. See Figure 15-14.  
Figure 15-13. Removing and Installing the Control Panel—PowerVault MD3600f  
1
2
1
control panel  
2
release tab  
240  
Management: Installing Array Components  
 
Figure 15-14. Removing and Installing the Control Panel—PowerVault MD3620f  
2
1
1
control panel  
2
release pin  
Installing the Control Panel  
1
2
Align the control panel with the slot on the array.  
Slide the control panel into the array until:  
The release tab clicks into place in PowerVault MD3600f. See  
Figure 15-13.  
The release pin clicks into place in PowerVault MD3620f. See  
Figure 15-14.  
3
Replace the hard drives in their respective slots. See "Installing a Hard  
Drive" on page 226.  
4
5
Connect all the power cables to the array.  
Turn on the array and the host server.  
Management: Installing Array Components  
241  
       
Backplane  
WARNING: Whenever you need to lift the array, get others to assist you. To avoid  
injury, do not attempt to lift the array by yourself.  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
Removing the Backplane  
1
2
3
4
Turn off the array and disconnect it from the electrical outlet.  
Disconnect all the cables connected to the array.  
Remove the hard drives. See "Removing a Hard Drive" on page 224.  
Remove the RAID controller modules. See "Removing a RAID Controller  
Module" on page 231.  
5
Supply/Cooling Fan Module" on page 237.  
6
7
Remove the control panel. See "Removing the Control Panel" on page 239.  
Remove the screws that secure the RAID controller module /power supply  
cage to the chassis.  
8
9
Grasp the cage removal ring at the bottom center of the array and pull the  
RAID controller module /power supply cage toward the back of the chassis.  
See Figure 15-15.  
Lift the RAID controller module /power supply cage away from the chassis.  
See Figure 15-15.  
10 Loosen the captive screw that secures the backplane to the chassis. See  
Figure 15-16 for PowerVault MD3600f or Figure 15-17 for  
PowerVault MD3620f.  
11 Remove the screws that secure the backplane and pull the backplane out  
of the array. See Figure 15-16 for PowerVault MD3600f or Figure 15-17 for  
PowerVault MD3620f.  
242  
Management: Installing Array Components  
       
Figure 15-15. Removing and Installing the RAID controller module /Power Supply Cage  
2
1
1
RAID controller module /power  
supply cage  
2
screws (6)  
Management: Installing Array Components  
243  
 
Figure 15-16. Removing and Installing the Backplane—PowerVault MD3600f  
1
2
3
1
3
screws (5)  
2
backplane  
captive screw  
Figure 15-17. Removing and Installing the Backplane—PowerVault MD3620f  
1
2
3
1
3
screws (4)  
2
backplane  
captive screw  
244  
Management: Installing Array Components  
   
Installing the Backplane  
1
2
Align the holes on the backplane with the holes on the array.  
Tighten the captive screw to secure the backplane to the chassis. See  
Figure 15-16 for PowerVault MD3600f or Figure 15-17 for  
PowerVault MD3620f.  
3
Replace the screws that secure the backplane to the chassis. See  
Figure 15-16 for PowerVault MD3600f or Figure 15-17 for  
PowerVault MD3620f.  
4
5
6
Align the slots on the RAID controller module /power supply cage with the  
tabs on the chassis. See Figure 15-15.  
Push the RAID controller module /power supply cage toward the front of  
the array.  
Replace the screws that secure the RAID controller module /power supply  
cage to the chassis.  
7
8
Replace the control panel. See "Installing the Control Panel" on page 241.  
Replace the power supply/cooling fan modules. See "Installing a Power  
Supply/Cooling Fan Module" on page 238.  
9
Replace the hard drives. See "Installing a Hard Drive" on page 226.  
10 Connect all the cables to the array.  
11 Turn on the array and the host server.  
Management: Installing Array Components  
245  
     
246  
Management: Installing Array Components  
16  
Management: Firmware Downloads  
Downloading RAID Controller and NVSRAM  
Packages  
A version number exists for each firmware file. The version number indicates  
whether the firmware is a major version or a minor version. You can use the  
Enterprise Management Window (EMW) to download and activate both the  
major firmware versions and the minor firmware versions. You can use the  
Array Management Window (AMW) to download and activate only the  
minor firmware versions.  
NOTE: Firmware versions are of the format aa.bb.cc.dd. Where aa is the major  
firmware version. bb.cc.dd is the minor firmware version. Depending on which one  
changes, firmware can be updated from EMW and AMW or only EMW.  
You can activate the files immediately or wait until a more convenient time.  
You might want to activate the firmware or NVSRAM files at a later time  
because of these reasons:  
Time of day—Activating the firmware and the NVSRAM can take a long  
time, so you can wait until I/O loads are lighter. The RAID controller  
modules goes offline briefly to load the new firmware.  
Type of packageYou might want to test the new firmware on one storage  
array before loading the files onto other storage arrays.  
The ability to download both files and activate them later depends on the  
type of RAID controller module in the storage array.  
NOTE: You can use the command line interface to download and activate the firmware  
to several storage arrays by using a script. See the PowerVault Modular Disk Storage  
Manager online help topics for more information on the command line interface.  
Management: Firmware Downloads  
247  
         
Downloading Both RAID Controller and NVSRAM  
Firmware  
NOTE: I/O to the array can continue while you are upgrading RAID controller and  
NVSRAM firmware.  
NOTE: Dell recommends that the firmware and NVSRAM be upgraded during a  
maintenance period when the array is not being used for I/O.  
NOTE: The RAID enclosure must contain at least two disk drives in order to update  
the firmware on the controller.  
To download RAID controller and NVSRAM firmware in a single operation:  
1
2
If you are using the EMW, go to step 9. If you are using the AMW, go to step 2.  
Perform one of these actions:  
Select AdvancedMaintenanceDownloadRAID Controller  
Module Firmware  
.
Select the Support tab, and click Download Firmware. In Select  
download task, select the Download RAID controller module  
firmware and click OK  
.
NOTE: The RAID Controller Module Firmware area and the NVSRAM area list  
the current firmware and the current NVSRAM versions respectively.  
3
4
To locate the directory in which the file to download resides, click Select File.  
In the File Selection area, select the file to download.  
By default, only the downloadable files that are compatible with the  
current storage array configuration appear.  
When you select a file in the File Selection area of the dialog, applicable  
attributes (if any) of the file appear in the File Information area. The  
attributes indicate the version of the file.  
5
6
If you want to download an NVSRAM file with the firmware, select  
Transfer NVSRAM file with RAID controller module firmware, and click  
Select File next to Selected NVSRAM file  
.
To transfer the files to the RAID controller module without activating  
them, click Transfer files but don't activate them (activate later)  
.
248  
Management: Firmware Downloads  
     
7
Click Transfer  
.
Keep these guidelines in mind:  
If the Transfer button is inactive, ensure that you either select an  
NVSRAM file or deselect the Transfer NVSRAM file with RAID  
controller module firmware option.  
If the file selected is not valid or is not compatible with the current storage  
array configuration, the File Selection Error dialog appears. Click OK to  
close it, and choose a compatible firmware or NVSRAM file.  
8
9
In the Confirm Download dialog, click Yes.  
The download starts.  
Perform one of these actions:  
Select Tools  
Select the Setup tab, and click Upgrade RAID Controller  
Module Firmware  
Upgrade RAID Controller Module Firmware.  
.
10 In the Storage array pane, select the storage array for which you want to  
upgrade the RAID controller module firmware or the NVSRAM.  
You can select more than one storage array.  
NOTE: The Details pane shows the details of only one storage array at a time.  
If you select more than one storage array in the Storage Array pane, the  
details of the storage arrays are not shown in the Details pane.  
11 Click Firmware in the Download area.  
If you select a storage array that cannot be upgraded, the Firmware button  
is disabled.  
The Download Firmware dialog appears. The current firmware version and  
the NVSRAM version of the selected storage arrays appear.  
NOTE: If you select the storage arrays with different RAID controller module  
types that cannot be updated with the same firmware or NVSRAM file and  
click Firmware, the Incompatible RAID Controller Modules dialog appears.  
Click OK to close the dialog and select the storage arrays with similar RAID  
controller module types.  
Management: Firmware Downloads  
249  
   
12 To locate the directory in which the file to download resides, click Browse  
in the Select files area.  
The Select File dialog appears.  
13 Select the file to download.  
14 Click OK  
.
15 If you want to download the NVSRAM file with the RAID controller  
module firmware, select Download NVSRAM file with firmware in the  
Select files area.  
Any attributes of the firmware file appear in the Firmware file information  
area. The attributes indicate the version of the firmware file.  
Any attributes of the NVSRAM file appear in the NVSRAM file  
information area. The attributes indicate the version of the NVSRAM file.  
16 If you want to download the file and activate the firmware and NVSRAM later,  
select the Transfer files but don’t activate them (activate later) check box.  
NOTE: If any of the selected storage arrays do not support downloading the  
files and activating the firmware or NVSRAM later, the Transfer files but don’t  
activate them (activate later) check box is disabled.  
17 Click OK  
.
The Confirm Download dialog appears.  
18 Click Yes  
.
The download starts and a progress indicator appears in the Status column  
of the Upgrade RAID Controller Module Firmware window.  
You also can use the command line interface (CLI) to download and activate  
NVSRAM to several storage arrays. For more information, see the PowerVault  
Modular Disk Storage Manager online help topics.  
To download only NVSRAM firmware:  
1
To download the NVSRAM firmware from:  
EMW—Go to step 9.  
AMW—Go to step 10  
250  
Management: Firmware Downloads  
   
2
Perform one of these actions:  
Select Advanced  
Module NVSRAM  
Maintenance  
Download  
RAID Controller  
Select the Support tab, and click Download Firmware. In Select  
download task, select Download RAID controller module NVSRAM  
and click OK  
.
An error message appears.  
3
4
5
Click OK to close it, and select a compatible file.  
To locate the directory in which the file to download resides, click Select File  
.
Select the file to download in the File selection area, and click OK  
.
By default, only downloadable files that are compatible with the current  
storage array configuration appear.  
When you select a file in the File selection area, applicable attributes (if  
any) of the file appear in the NVSRAM File information area. The  
attributes indicate the version of the NVSRAM file.  
6
Click Transfer  
.
NOTE: If the file selected is not valid or is not compatible with the current  
storage array configuration, the File Selection Error dialog appears. Click OK  
to close it, and choose a compatible NVSRAM file.  
7
8
Click Yes in the Confirm Download dialog.  
The download starts.  
Perform one of these actions:  
Select Tools  
Upgrade RAID Controller Module Firmware.  
Select the Setup tab, and click Upgrade RAID Controller  
Module Firmware.  
The Upgrade RAID Controller Module Firmware window appears.  
The Storage array pane lists the storage arrays. The Details pane shows the  
details of the storage array that is selected in the Storage array pane.  
9
In the Storage array pane, select the storage array for which you want to  
download the NVSRAM firmware.  
You can select more than one storage array.  
Management: Firmware Downloads  
251  
 
NOTE: The Details pane shows the details of only one storage array at a time.  
If you select more than one storage array in the Storage array pane, the  
details of the storage arrays are not shown in the Details pane.  
10 Click NVSRAM in the Download area.  
NOTE: If you select a storage array that cannot be upgraded, the NVSRAM  
button is disabled.  
The Download NVSRAM dialog is displayed. The current firmware  
version and the NVSRAM version of the selected storage arrays appear.  
NOTE: If you select the storage arrays with different RAID controller module  
types that cannot be updated with the same NVSRAM file and click NVSRAM, the  
Incompatible RAID Controller Modules dialog appears. Click OK to close the  
dialog and select the storage arrays with similar RAID controller module types.  
11 To locate the directory in which the NVSRAM file to download resides,  
click Browse in the Select file area.  
The Select File dialog appears.  
12 Select the file to download.  
13 Click OK  
.
Any attributes of the NVSRAM file appear in the NVSRAM file  
information area. The attributes indicate the version of the NVSRAM file.  
14 Click OK  
.
The Confirm Download dialog appears.  
15 Click Yes  
.
The download starts and a progress indicator appears in the Status column of  
the Upgrade RAID Controller Module Firmware window.  
252  
Management: Firmware Downloads  
Downloading Physical Disk Firmware  
CAUTION: When updating physical disk firmware, you should stop all I/O activity  
to the array to prevent data loss.  
The physical disk firmware controls various features of the physical disk. The  
disk array controller (DAC) uses this type of firmware. Physical disk firmware  
stores information about the system configuration on an area of the physical  
disk called DACstore. DACstore and the physical disk firmware enable easier  
reconfiguration and migration of the physical disks. The physical disk  
firmware performs these functions:  
The physical disk firmware records the location of the physical disk in an  
expansion enclosure. If you take a physical disk out of an expansion  
enclosure, you must insert it back into the same physical disk slot, or the  
physical disk firmware cannot communicate with the RAID controller  
module or other storage array components.  
RAID configuration information is stored in the physical disk firmware and  
is used to communicate with other RAID components.  
CAUTION: Risk of application errors – Downloading the firmware could cause  
application errors.  
Keep these important guidelines in mind when you download firmware to  
avoid the risk of application errors:  
Downloading firmware incorrectly could result in damage to the physical  
disks or loss of data. Perform downloads only under the guidance of your  
Technical Support representative.  
Stop all I/O to the storage array before the download.  
Ensure that the firmware that you download to the physical disks are  
compatible with the physical disks that you select.  
Do not make any configuration changes to the storage array while  
downloading the firmware.  
NOTE: Downloads can take several minutes to complete. During a download, the  
Download Physical Disk - Progress dialog appears. Do not attempt another  
operation when the Download Physical Disk - Progress dialog is shown.  
Management: Firmware Downloads  
253  
   
To download Physical Disk Firmware:  
1
From the AMW, select Advanced  
Download Physical Disk.  
The Download Physical Disk - Introduction window appears.  
Maintenance  
2
3
Click Next.  
The Download Physical Disk Firmware - Add Packages window is displayed.  
In the Selected Packages area, click Add. Navigate to the location of the  
packages and click OK  
The selected package is added to the Packages to be transferred area.  
Click Next  
.
4
5
6
7
.
The Download Physical Disk Firmware - Select Physical Disks window  
is displayed.  
In the Compatible Physical Disks tab, select the appropriate physical  
disks or Select all the physical disks.  
The Confirm Download dialog is displayed.  
Type yes and click OK  
.
The Download Physical Disk Firmware - Progress window is displays the  
progress of physical disk firmware download.  
After the firmware download is complete click Close  
.
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
254  
Management: Firmware Downloads  
Downloading MD1200 Series Expansion Module  
EMM Firmware  
NOTE: Do not make any configuration changes to the storage array while you are  
downloading the expansion enclosure EMM firmware. Doing so could cause the  
firmware download to fail, damage the storage array, or cause loss of data accessibility.  
NOTE: Due to a limitation with Linux, expansion enclosure EMM firmware updates  
must be performed using out-of-band management only. Failure to do so may result  
in the host server becoming unresponsive, and it may require a reboot.  
You can transfer a downloadable firmware file to the expansion enclosure  
EMM in the expansion enclosures attached to the storage array.  
CAUTION: Risk of possible loss of data or risk of damage to the storage array –  
Downloading the expansion enclosure EMM firmware incorrectly could result in  
loss of data or damage to the storage array. Perform downloads only under the  
guidance of your Technical Support representative.  
CAUTION: Risk of making expansion enclosure EMM unusable – Do not make  
any configuration changes to the storage array while downloading expansion  
enclosure EMM firmware. Doing so could cause the firmware download to fail  
and make the selected expansion enclosure unusable.  
To download the EMM firmware:  
1
Perform one of these actions:  
In the AMW, select Advanced  
EMM Firmware  
Select the Support tab, and click Download Firmware. In the dialog  
that appears, select the EMM firmware, and click OK  
Maintenance  
Download  
.
.
The Download Environmental (EMM) Firmware dialog appears.  
2
In the Select enclosures area, either select each expansion enclosure to  
which you want to download firmware, or select the Select All to select all  
of the expansion enclosures in the storage array.  
Each selected expansion enclosure should have the same product ID.  
3
4
Click Select File  
The Select Environmental (EMM) Card Firmware File dialog appears.  
Select the file to download, and click OK  
.
.
Management: Firmware Downloads  
255  
   
5
6
Click Start.  
Click Yes to continue with the firmware download.  
NOTE: If you click Stop while a firmware download is in progress, the  
download-in-progress finishes before the operation stops. The status for the  
remaining expansion enclosures changes to Canceled.  
Monitor the progress and completion status of the download to the  
expansion enclosures. The progress and status of each expansion enclosure  
that is participating in the download appears in the Status column of the  
Select enclosures table.  
NOTE: Each firmware download can take several minutes to complete.  
7
Perform one of these actions depending on whether the download succeeded:  
The download succeeded – The statuses of all the expansion enclosures  
show Complete. You can close the Download environmental (EMM)  
Card Firmware dialog by clicking Close. The expansion enclosure EMM  
cards are now operating with the new firmware.  
The download failed – The status of one expansion enclosure shows  
Failed, and the remainder of the expansion enclosures show Canceled.  
Ensure that the new firmware file is compatible before attempting  
another firmware download.  
Self-Monitoring Analysis and Reporting  
Technology (SMART)  
Self-Monitoring Analysis and Reporting Technology (SMART) monitors the  
internal performance of all physical disk components to detect faults indicating  
the potential for physical disk failure. SMART uses this information to report  
whether failure is imminent so that a physical disk can be replaced before failure  
occurs. The RAID controller monitors all attached drives and notifies users when  
a predicted failure is reported by a physical disk.  
256  
Management: Firmware Downloads  
 
Media Errors and Unreadable Sectors  
If the RAID controller detects a media error while accessing data from a  
physical disk that is a member of a disk group with a redundant RAID level  
(RAID 1, RAID 5 or RAID 10), the controller tries to recover the data from  
peer disks in the disk group and uses recovered data to correct the error. If the  
controller encounters an error while accessing a peer disk, it is unable to  
recover the data and affected sectors are added to the unreadable sector log  
maintained by the controller. Other conditions under which sectors are added  
to the unreadable sector log include:  
A media error is encountered when trying to access a physical disk that is a  
member of a nonredundant disk group (RAID 0 or degraded RAID 1,  
RAID 5 or RAID 10).  
An error is encountered on source disks during rebuild.  
NOTE: Data on an unreadable sector is no longer accessible.  
Management: Firmware Downloads  
257  
   
258  
Management: Firmware Downloads  
17  
Management: Firmware Inventory  
A storage array is made up of many components, which might include RAID  
controller modules, physical disks, and enclosure management modules  
(EMMs). Each of these components contains firmware. Some versions of  
firmware are dependent on other versions of firmware. To capture  
information about all of the firmware versions in the storage array, view the  
firmware inventory.  
If the firmware inventory does not contain information for a particular storage  
array, the firmware inventory service is not available on that storage array.  
You can also save the firmware inventory to a text file. You can then send the  
file to your Technical Support representative for analysis. Your Technical  
Support representative is able to detect any firmware mismatches.  
Viewing the Firmware Inventory  
1
Perform one of these actions based on whether you want to view the  
firmware information for one storage array or all storage arrays:  
One storage array—From the Array Management Window, select  
AdvancedMaintenance Firmware Inventory  
All storage arrays—From the Enterprise Management Window, select  
Tools Firmware Inventory  
To save the firmware inventory to a text file, Click Save As  
.
.
2
3
.
In Specify filename, enter a name for the file to be saved. You might also  
specify another physical disk and directory if you want to save the file in a  
location other than the default.  
NOTE: The suffix *.txt is added to the file name automatically if you do not  
specify a suffix for the file name.  
4
Click Save.  
An ASCII text file that contains the firmware inventory is saved to the  
designated directory.  
Management: Firmware Inventory  
259  
       
260  
Management: Firmware Inventory  
18  
Management: System Interfaces  
Microsoft Services  
Virtual Disk Service  
The Microsoft Virtual Disk Service (VDS) is a component of the Windows  
operating system. The VDS component utilizes third-part vendor specific  
software modules, known as providers, to access and configure third-party  
storage resources, such as MD3600f storage arrays. The VDS component  
exposes a set of application programming interfaces (APIs) that provides a  
single interface for managing disks and other storage hardware. The MD3600f  
Series VDS Provider enables Windows tools, including the Disk Manager, to  
access and configure storage array virtual disks.  
The VDS Provider for the MD3600f Series arrays is available on the MD3600f  
Series resource media. For more information on VDS, see microsoft.com.  
Volume Shadow-Copy Service  
The Microsoft Volume Shadow-copy Service (VSS) is a component of the  
Microsoft Windows operating system. The VSS component utilizes third-part  
vendor specific software modules, known as providers, to access and utilize  
snapshot and disk copy functionality provided by third-party storage  
resources, such as MD3600f Series storage arrays. The combination of the  
VSS component and the VSS Provider, included on the MD3600f Resource  
media, enable the MD3600f Series arrays to be utilized by third-party and  
Windows backup and snapshot applications.  
NOTE: Virtual disks that is used as source virtual disks for VSS snapshots should  
have names no longer than 16 characters.  
The VSS hardware provider uses the source virtual disk name as a prefix for the  
snapshot and repository virtual disk names. The resulting snapshot and repository  
names are too long if the source virtual disk name exceeds 16 characters.  
Management: System Interfaces  
261  
           
VSS attaches to the service and uses it to coordinate the creation of snapshot  
virtual disks on the storage array. VSS-initiated snapshot virtual disks can be  
triggered through backup tools, known as requestors. The VSS Provider  
Configuration Tool makes available the following configuration options:  
Snapshot Repository Virtual Disk Properties—This section contains a  
drop-down list for the RAID level and a field for entering source virtual  
disk capacity percentage for snapshot repositories.  
Snapshot Repository Virtual Disk Location—This section contains a list of  
preferences for the location of the snapshot repository virtual disk. These  
preferences are honored whenever conditions permit.  
The Microsoft VSS installer service for storage provisioning is available on the  
MD3600f resource media in the \windows\VDS_VSS directory.  
NOTE: When registering VSS during your Windows setup, the registration  
graphical user interface (GUI) prompts you to provide the name of your array  
because settings in the GUI are array-specific, not host-specific.  
Storage Management VSS Hardware Provider Tips:  
The number of snapshot virtual disks that can be created using a single  
snapshot set varies with the I/O load on the RAID controller modules.  
Under little or no I/O load, the number of virtual disks in a snapshot set  
should be limited to eight. Under high I/O loads, the limit should be three.  
The snapshot virtual disks created in the storage management software are  
differential snapshots. Plex snapshots are not supported.  
Virtual disks to be used as source virtual disks for VSS snapshots should  
have names no longer than 16 characters. The VSS hardware provider uses  
the base virtual disk name as a prefix for the snapshot and repository  
virtual disk names. The resulting snapshot and repository names are too  
long if the source virtual disk name exceeds 16 characters.  
NOTE: A volume is another term for virtual disk.  
For more information on VDS and VSS, see microsoft.com.  
262  
Management: System Interfaces  
19  
Array Software  
Start–Up Routine  
Look and listen during the array’s start-up routine for the indications  
described in Table 19-1. For a description of the front- and back-panel  
indicators, see "Planning: About Your Storage Array" on page 23.  
Table 19-1. Start-Up Routine Indications  
Look/listen for  
Action  
Alert messages.  
See your storage  
management documentation.  
An unfamiliar constant scraping or grinding See "Getting Help" on page 303.  
sound when you access a physical disk.  
NOTE: At least two physical disks must be installed in the array.  
Device Health Conditions  
When you open the Enterprise Management Window (EMW), the Dell  
PowerVault Modular Disk Storage Management software (MDSM)  
establishes communication with each managed storage array and determines  
the current storage array status. The current status is represented by icons  
next to the managed storage array.  
The status icons shown in the Tree view in the EMW represent a summary  
status for each storage array. If a storage array has a status of Needs Attention  
or a status of Fixing, you must determine the condition that is causing this  
status before attempting any management actions. You can determine the  
condition causing the Needs Attention status or the Fixing status by selecting  
the storage array and launching its Array Management Window (AMW).  
Troubleshooting: Your Storage Array Software  
263  
           
To launch the AMW, perform one of these actions:  
On the Devices tab, in either the Tree view or the Table view, double-click  
a storage array. Alternatively, you can right-click a storage array and select  
Manage Storage Array from the pop-up menu.  
On the Setup tab, select Manage a Storage Array  
.
After the AMW opens, select the Physical tab to see the components in the  
storage array. A component that has a problem is indicated by a status icon.  
The status icons indicate the status of the components that comprise the  
storage array. Also, the Recovery Guru option provides a detailed explanation  
of the conditions and the applicable steps to remedy any Needs Attention  
status. For more information, see "Recovery Guru" on page 270.  
For the status of a storage array, the icons shown in the following table are  
used in the Tree view, the Table view, and both the EMW Status Bar and the  
AMW Status Bar.  
Table 19-2. Status Icon  
Status  
Icon  
Description  
Optimal  
Each component in the managed storage array is in  
the desired working condition.  
Needs Attention  
Unresponsive  
There is a problem with the managed storage array  
that requires your intervention to correct it.  
The storage management station cannot  
communicate with the storage array or one RAID  
controller module or both RAID controller modules  
in the storage array.  
Fixing Status  
Unsupported  
A Needs Attention status has been corrected and the  
managed storage array is currently transitioning to an  
Optimal state.  
The node is currently not supported by this version of  
MDSM.  
Software  
Unsupported  
The storage array is running a level of software that is  
no longer supported by MDSM.  
264  
Troubleshooting: Your Storage Array Software  
In the Table view, every managed storage array is listed once, regardless of the  
number of attachments it has in the Tree view. After the storage array has  
been contacted by MDSM, an icon representing its hardware status appears.  
Hardware status can be Optimal, Needs Attention, or Fixing. If, however, all  
of the network management connections from the storage management  
station to the storage array shown in the Tree view are Unresponsive, the  
storage array status is represented as Unresponsive.  
In the EMW Status Bar and the AMW Status Bar, the icons also have these  
behaviors:  
Hold the mouse over the icon in the EMW Status Bar and the AMW  
Status Bar to show a tooltip with a brief description of the status.  
The icons for the Needs Attention status and Unresponsive status appear  
in EMW Status Bar and the AMW Status Bar if there are discovered  
storage arrays with either condition.  
The EMW Tree view has additional status icons that are shown in the  
following table.  
Table 19-3. Additional Status Icons  
Status  
Icon  
Description  
Unsupported Alerts  
with a Needs  
Upgrade Status  
Setting an alert on a s storage array with a Needs  
Upgrade status is not supported. In this case, the  
storage array shows both a Needs Upgrade status  
and an Unsupported Alerts icon in the Tree view.  
The Unsupported Alerts icon indicates that the  
storage array cannot be monitored.  
Alert Set  
If you installed the Event Monitor with MDSM,  
and if you have set alerts, the Alert Set icon  
appears next to the storage array status in the  
Tree view for which the alerts have been set.  
Setting an Alert at  
the Parent Node  
Level  
You can set alerts at any of the nodes in the Tree  
view. Setting an alert at a parent node level, such  
as at a host level, sets alert fir any child nodes. If  
you set an alert at a parent node level and any of  
the in-band storage array child nodes have a  
Needs Upgrade status, the Alert Disables status  
icon appears next to the parent node in the tree  
view.  
Troubleshooting: Your Storage Array Software  
265  
Table 19-3. Additional Status Icons  
Status  
Icon  
Description  
Adding a Storage  
Array  
The Contacting Storage Array icon is shown in  
the Tree view and Table view until the current  
status of each managed storage array is known.  
The Contacting Storage Array icon is shown in  
the EMW Status Bar and the AMW Status Bar  
and the tooltip shows Contacting Storage arrays.  
As each storage array is contacted, its current status  
is obtained and shown in the Tree view and Table  
view. The applicable statuses are the Optimal,  
Needs Attention, Fixing, or Unresponsive.  
Adding a Storage  
Array OK  
No problems were encountered while adding the  
storage array.  
MDSM software continues to check for any  
status change events.  
Adding a Storage  
Array Error  
Appears only when an error occurs.  
NOTE: MDSM may take a few minutes to update a status change to Unresponsive  
or from Unresponsive. A status change from or to Unresponsive depends on the  
network link to the storage array. All other status change updates faster.  
Storage Array Support Data  
You can gather various types of inventory, status, and performance data that  
can help troubleshoot any problem with the storage array. All the files are  
compressed into a single archive in a zipped-file format. You can forward the  
archive file to your Technical Support representative for troubleshooting and  
further analysis.  
To generate the support data report:  
1
In the AMW, perform one of these actions:  
Select Advanced  
Troubleshooting  
Support Data  
Collect.  
Select the Support tab, and click Gather Support Information  
.
The Collect All Support Data window is displayed.  
266  
Troubleshooting: Your Storage Array Software  
   
2
Enter a name for the support data file in Specify filename or click Browse  
to navigate to a previously saved file to overwrite an existing file.  
The suffix .zip is added automatically to the file if you do not specify a  
suffix for the file.  
3
4
Enter the Execution summary  
.
Click Start  
.
After all of the support files have been gathered, they are archived using  
the file name that you specified.  
5
Click OK.  
Automatically Collect the Support Bundle Data  
You can use the Collect Support Bundle option to automatically save a copy of  
the support bundle when the client monitor process detects a critical event.  
During a critical event, the support bundle is saved to the local physical disk  
of the client system in the same area that is used for other recovery  
information. This information is not overwritten for at least 72 hours.  
WARNING: Use this option only under the guidance of your Technical Support  
representative.  
To automatically collect the support bundle data:  
1
In the AMW, select Advanced  
Automatic Settings  
Select Automatically collect support data for critical events  
To change the location of the saved support bundle, click Change  
The Change Folder Location window is displayed, navigate to the relevant  
folder and click OK  
Troubleshooting  
Support Data  
.
2
3
.
.
.
4
5
To reset the default location, click Reset  
Click OK  
.
.
Troubleshooting: Your Storage Array Software  
267  
   
Collecting the Physical Disk Data  
You can use the Collect Physical Disk Data option to collect log sense data  
from all the physical disks on your storage array.  
Log sense data consists of statistical information that is maintained by each  
of the physical disks in your storage array. Your Technical Support  
representative can use this information to analyze the performance of your  
physical disks and for troubleshooting problems that might exist.  
WARNING: Use this option only under the guidance of your Technical Support  
representative.  
To collect physical disk data:  
1
In the AMW, select Advanced  
Disk Data  
The Collect Physical Disk Data window is displayed.  
Troubleshooting  
Collect Physical  
.
2
Enter a name for the physical disk data filename in Specify filename or click  
Browse to navigate to a previously saved file to overwrite an existing file.  
The suffix *.bin is added to the file automatically if you do not specify a  
suffix for the file.  
3
4
Click Start.  
The physical disk data collection is completed and saved at the location  
that you entered.  
Click OK  
.
268  
Troubleshooting: Your Storage Array Software  
   
Event Log  
You can use the Event Log Viewer to view a detailed list of events that occur  
in a storage array. The event log is stored on reserved areas on the storage array  
disks. It records configuration events and storage array component failures.  
The event log can be used as a supplementary diagnostic tool to the Recovery  
Guru for tracing storage array events. Always refer to the Recovery Guru first  
when attempting to recover from component failures in the storage array.  
WARNING: Use this option only under the guidance of your Technical  
Support representative.  
The event log stores approximately 8000 events before it replaces an event  
with a new event. If you want to keep the events, you may save them, and  
clear them from the event log.  
The event log window shows two types of event views:  
Summary view—Shows an event summary in a table form.  
Detail view—Shows details about a selected event.  
To view the event log:  
1
2
In the AMW, select Advanced  
The Event Log is displayed. By default, the summary view is displayed.  
To view the details of each selected log entry, select View details  
Troubleshooting  
View Event Log.  
.
A detail pane is added to the event log that contains detailed information  
about the log item. You can view the details about a single log entry at a time.  
3
To save the event log, click Save As  
The Save Events dialog is displayed, navigate to the relevant folder, enter  
the relevant file name, and click Save  
To erase all log entries from the event log, click Clear All  
To exit the event log, click Close  
.
.
4
5
.
.
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Troubleshooting: Your Storage Array Software  
269  
   
Recovery Guru  
The Recovery Guru is a component of MDSM that diagnoses critical events  
on the storage array and recommends step-by-step recovery procedures for  
problem resolution.  
In the AMW, to display the Recovery Guru, perform one of these actions:  
Click Recovery Guru  
.
In the Support tab, click the Recover from Failure link.  
From the Status pane on the Summary tab, click the Storage Array Needs  
Attention link.  
You can detect a problem using the following indicators:  
Non-Optimal status icons  
Alert notification messages that are sent to the appropriate destinations  
Hardware indicator lights  
The status icons return to Optimal status as problems are resolved.  
Storage Array Profile  
The storage array profile provides a description of all of the components and  
properties of the storage array. The storage array profile also provides the  
option to save the storage array profile information to a text file. You might  
want to use the storage array profile as an aid during recovery or as an  
overview of the current configuration of the storage array. Create a new copy  
of the storage array profile if your configuration changes.  
To open the storage array profile:  
1
In the AMW, perform one of the following actions:  
Select Storage Array  
View  
Profile.  
Select the Summary tab, and click Storage Array Profile in the  
Status area.  
Select the Support tab, and click View Storage Array Profile.  
The Storage Array Profile dialog appears. The Storage Array Profile dialog  
contains several tabs, and the title of each tab corresponds to the subject  
of the information contained.  
270  
Troubleshooting: Your Storage Array Software  
       
2
Perform one of these actions in the Storage Array Profile dialog:  
View detailed information—Go to step 3.  
Search the storage array profile—Go to step 4.  
Save the storage array profile—Go to step 5.  
Close the storage array profile—Go to step 6.  
3
4
Select one of the tabs, and use the horizontal scroll bar and the vertical  
scroll bar to view the storage array profile information.  
You can use the other steps in this procedure to search the storage array  
profile, to save the storage array profile, or to close the storage array profile.  
To search the storage array profile, perform these steps:  
a
b
Click  
.
Type the term that you want to search for in the Find text box.  
If the term is located on the current tab, the term is highlighted in the  
storage array profile information.  
NOTE: The search is limited to the current tab. If you want to search for the  
term in other tabs, select the tab and click the Find button again.  
c
Click the Find button again to search for additional occurrences of  
the term.  
5
To save the storage array profile, perform these steps:  
a
b
c
Click Save As.  
To save all sections of the storage array profile, select the All sections  
.
To save information from particular sections of the storage array  
profile, select the Select sections, and select the check boxes  
corresponding to the sections that you want to save.  
d
Select an appropriate directory.  
Troubleshooting: Your Storage Array Software  
271  
     
e
f
In File Name, type the file name of your choice. To associate the file  
with a particular software application that opens it, specify a file  
extension, such as .txt.  
NOTE: The file is saved as ASCII text.  
Click Save  
.
6
To exit the storage array profile, click Close.  
For more information, see the PowerVault Modular Disk Storage Manager  
online help topics.  
Viewing the Logical Associations  
You can use the Associated Logical Elements option to view the logical  
associations among different virtual disks in a storage array. To view the  
associations for source virtual disks, snapshot virtual disks, and snapshot  
repository virtual disks:  
1
In the AMW, select the Logical tab or in the Topology pane of the  
Mappings tab.  
2
Select View  
Associated Logical Elements.  
or  
Right-click the virtual disk to open a pop-up menu and select View  
Associated Logical Elements  
.
If you select a virtual disk that does not have logical associations with other  
virtual disks, the Associated Logical Elements option is disabled.  
NOTE: The View Associated Logical Elements dialog appears, which  
indicates the logical associations for the selected virtual disk.  
3
To close the View Associated Logical Elements dialog, click Close.  
272  
Troubleshooting: Your Storage Array Software  
     
Viewing the Physical Associations  
You can use the Associated Physical Components option to view the physical  
components that are associated with source virtual disks, snapshot virtual  
disks, snapshot repository virtual disks, disk groups, unconfigured capacity,  
and free capacity in a storage array.  
To view the physical associations:  
1
In the AMW, select a node in the Logical pane of the Logical tab or in the  
Topology pane of the Mappings tab.  
2
Select View  
selected node is a virtual disk, you can right-click the node to open a pop-  
up menu and select View Associated Physical Components. If the  
selected node is a disk group, unconfigured capacity, or free capacity, you  
can right-click the node to open a pop-up menu and select View  
Associated Physical Components  
Associated Physical Components. Alternatively, if the  
.
The View Associated Physical Components dialog appears with blue dots  
next to the physical components that are associated with the selected node.  
3
To close the View Associated Physical Components dialog, click Close.  
Finding Nodes  
You can use the Find option to search for a particular node on the Logical tab,  
the Physical tab, or the Mappings tab of the AMW. The search might be  
based on a particular node name, the RAID level, virtual disk capacity, or  
specific free capacity nodes. The search might be based also on one of these  
combinations:  
The node name and the RAID level  
The node name and the virtual disk capacity  
Troubleshooting: Your Storage Array Software  
273  
       
To find nodes:  
1
2
In the AMW, select View  
Find.  
Based on the type of search, select one of these options, and go to the  
indicated step:  
Search by name—see step 3.  
Search by special criteria—see step 4.  
3
4
Type the name of the node to be found in the Find node text box. See  
step 8.  
Based on the search criteria, select one of these options, and go to the  
indicated step:  
Find all virtual disks with RAID level—Go to step 5.  
Find all virtual disks with capacity—Go to step 6.  
Find all free capacity nodes—Go to step 7.  
5
6
To search for all nodes based on their RAID level, perform these steps:  
a
b
c
Select Find all virtual disks with RAID level  
Select the RAID level from the list.  
Go to step 8.  
.
To search for all nodes based on their virtual disk capacity, perform these  
steps:  
a
b
c
Select Find all virtual disks with capacity  
.
Type the capacity in the GB box.  
Specify that the capacity to be matched is less than, equal to, or  
greater than the capacity entered in the GB box.  
d
Go to step 8.  
7
To search for all Free Capacity nodes with a particular capacity, perform  
these steps:  
NOTE: This option is not available when the Search by name option is  
selected or from the Mappings tab. You must cancel the selection of the  
Search by name option to use this option.  
a
b
Select Find all free capacity nodes.  
Type the capacity in the GB box.  
274  
Troubleshooting: Your Storage Array Software  
         
c
d
Specify that the free capacity to be matched is less than, equal to, or  
greater than the capacity entered in the GB box.  
Go to step 8.  
8
9
Click Find Next.  
To see every node that matches the criteria, click Find Next repeatedly. If  
no matches are found, the Search Failed dialog appears. Click OK, and re-  
enter the search criteria.  
To close the dialog, click Cancel  
.
To continue searching for nodes with the same criteria after the Find  
dialog has been closed, press <F3>.  
Using Go To  
Use the Go To option to quickly jump to an associated snapshot repository  
virtual disk, snapshot virtual disk, source virtual disk, or target virtual disk.  
These virtual disks appear in the Logical pane of the Logical tab.  
The Go To option is available only if the Snapshot premium feature or the  
Virtual Disk Copy premium feature is enabled or if snapshot virtual disks or  
virtual disk copies currently exist on the storage array. The Go To option is not  
accessible from the Mappings tab of the Array Management Window.  
1
On the Logical tab of the AMW, select one of these virtual disks, and go to  
the indicated step:  
Snapshot virtual disk—Go to step 2.  
Snapshot repository virtual disk—Go to step 3.  
Source virtual disk—Go to step 4.  
Target virtual disk—Go to step 5.  
2
3
Select View  
Go To  
Snapshot Virtual Disk.  
The selection jumps to the associated snapshot virtual disk in the  
Logical pane.  
Select View  
Go To  
Snapshot Repository Virtual Disk.  
The selection jumps to the associated snapshot repository virtual disk in  
the Logical pane.  
Troubleshooting: Your Storage Array Software  
275  
         
4
5
Select View  
Go To  
Source Virtual Disk.  
The selection jumps to the associated source virtual disk in the  
Logical pane.  
Select View  
Go To  
Target Virtual Disk.  
NOTE: If the source virtual disk has more than one associated target virtual  
disk, select the target virtual disk that you want from the list, and click OK.  
The selection jumps to the associated target virtual disk in the Logical pane.  
Recovering From an Unresponsive Storage Array  
Condition  
A storage array can have an Unresponsive status for several reasons. Use the  
procedure in this topic to determine a possible cause and solution.  
MDSM can take up to five minutes to detect that a storage array has become  
unresponsive or becomes responsive again. Before completing this procedure,  
ensure that you wait some time before you decide that the storage array is still  
unresponsive.  
To recover from an unresponsive storage array:  
1
2
Check the Tree View in the EMW to see if all storage arrays are unresponsive.  
If any storage arrays are unresponsive, check the storage management  
station network connection to ensure that it can reach the network.  
3
Ensure that the RAID controller modules are installed and that there is  
power to the storage array.  
4
5
If there a problem with the storage array, correct the problem.  
Perform one of these actions, depending on how your storage array is managed:  
Out-of-band managed storage array—Go to step 6.  
In-band managed storage array—Go to step 12.  
6
For an out-of-band managed storage array, ensure that the RAID controller  
modules are network accessible by using the ping command to ensure that  
the RAID controller module can be reached. Type one of these commands,  
and press Enter.  
ping <host-name>  
ping <RAID controller module-IP-address>  
276  
Troubleshooting: Your Storage Array Software  
         
7
8
If the verification is successful, see step 8, if not, see step 9.  
Remove the storage array with the Unresponsive status from the EMW,  
and select Add Storage Array to add the storage array again.  
9
If the storage array does not return to Optimal status, check the Ethernet  
cables to ensure that there is no visible damage and that they are  
securely connected.  
10 Ensure the appropriate network configuration tasks have been performed.  
For example, ensure that IP addresses have been assigned to each RAID  
controller module.  
11 If there is a cable or network accessibility problem, see step 20, if not  
step 12.  
12 For an in-band managed storage array, ensure that the host is network  
accessible by using the ping command to verify that the host can be  
reached. Type one of these commands, and press <Enter>.  
ping <host-name>  
ping <RAID controller module-IP-address>  
13 If the verification is successful, go to step 14. If the verification is not  
successful, go to step 15.  
14 Remove the host with the Unresponsive status from the EMW, and select  
Add Storage Array to add the host again.  
15 If the host does not return to Optimal status, go to step 16.  
16 Ensure that the host is turned on and operational and that the host  
adapters have been installed.  
17 Check all external cables and switches or hubs to ensure that no visible  
damage exists and that they are securely connected.  
18 Ensure the host context agent software is installed and running.  
If you started the host system before you were connected to the RAID  
controller module in the storage array, the host context agent software is not  
able to detect the RAID controller modules. If this is the case, ensure that the  
connections are secure, and restart the host context agent software.  
19 If you have recently replaced or added the RAID controller module, restart  
the host context agent software so that the new RAID controller module  
is recognized.  
Troubleshooting: Your Storage Array Software  
277  
           
20 If the problem still exists, make the appropriate host modifications, check  
with other administrators to see if a firmware upgrade was performed on  
the RAID controller module from another storage management station.  
If a firmware upgrade was performed, the EMW on your management  
station might not be able to locate the new AMW software needed to  
manage the storage array with the new version of the firmware.  
21 If the problem persists contact your Technical Support representative.  
22 Determine if there is an excessive amount of network traffic to one or  
more RAID controller modules.  
This problem is self-correcting because the EMW software periodically retries  
to establish communication with the RAID controller modules in the storage  
array. If the storage array was unresponsive and a subsequent try to connect to  
the storage array succeeds, the storage array becomes responsive.  
For an out-of-band managed storage array, determine if management  
operations are taking place on the storage array from other storage  
management stations. A RAID controller module-determined limit exists  
to the number of Transmission Control Protocol/Internet Protocol  
(TCP/IP) connections that can be made to the RAID controller module  
before it stops responding to subsequent connection attempts. The type of  
management operations being performed and the number of management  
sessions taking place together determine the number of TCP/IP  
connections made to a RAID controller module. This problem is self-  
correcting because, after some TCP/IP connections terminate, the RAID  
controller module then becomes responsive to other connection attempts.  
23 If the storage array is still unresponsive, a problem might exist with the  
RAID controller modules. Contact your Technical Support representative.  
278  
Troubleshooting: Your Storage Array Software  
 
Locating a Physical Disk  
You can use the Locate Physical Disk option to physically locate and identify  
one or more of the physical disks in an expansion enclosure by activating  
physical disk LEDs.  
To locate the physical disk:  
1
2
3
Select the Physical tab.  
Select the physical disks that you want to locate.  
Select Physical Disk  
Blink  
Physical Disk.  
The LEDs on the selected physical disks blink.  
4
When you have located the physical disks, click OK.  
The LEDs stop blinking. If any other blink operations (Blink Disk Group,  
Blink Storage Array, Blink Physical Disk Ports, or Blink Expansion  
Enclosure) are currently being invoked from another storage management  
station, these LEDs also stop blinking.  
5
6
In the rare case that the LEDs on the physical disks do not stop blinking,  
in the AMW, select Storage Array  
If the LEDs successfully stop blinking, a confirmation message appears.  
Click OK  
Blink  
Stop All Indications.  
.
Locating an Expansion Enclosure  
You can use the Blink option to physically locate and identify an expansion  
enclosure in the storage array.  
The LED activation varies according to the type of expansion enclosure that  
you have.  
If you have an expansion enclosure with a white LED, the Blink Expansion  
Enclosure operation causes the white LED on the expansion enclosure to  
come on. The LED does not blink.  
If you have any other types of expansion enclosures, this operation causes  
the appropriate LED on all of the physical disks in the expansion enclosure  
to blink.  
Troubleshooting: Your Storage Array Software  
279  
       
To locate the expansion enclosure:  
1
2
3
Select the Physical tab.  
Select a physical disk in the expansion enclosure that you want to locate.  
Select Physical Disk  
The LED or LEDs on the expansion enclosure or physical disks come on.  
When you have located the expansion enclosure, click OK  
Blink  
Expansion Enclosure.  
4
.
The LEDs stop blinking. (If you have an expansion enclosure with a blue  
LED, the LED goes off). If any other blink operations (Blink Storage Array,  
Blink Disk Group, Blink Physical Disk Ports, Blink Expansion Enclosure, or  
Blink Physical Disk) are currently being invoked from another storage  
management station, these LEDs also stop blinking.  
5
6
In the rare case that the LEDs on the expansion enclosure do not stop  
blinking, from the Array Management Window, select Storage Array  
Blink Stop All Indications  
If the LEDs successfully stop blinking, a confirmation message appears.  
.
Click OK  
.
Capturing the State Information  
Use the Troubleshooting  
Capture State Information option to capture  
information about the current state of your storage array and save the  
captured information to a text file. You can then send the captured  
information to your Technical Support representative for analysis.  
Potential to cause an unresponsive storage array – The State Capture option  
can cause a storage array to become unresponsive to both the host and the  
storage management station. Use this option only under the guidance of your  
Technical Support representative.  
1
2
3
From the AMW, select Advanced  
Troubleshooting  
Capture  
State Information  
.
Read the information in the Confirm State Capture dialog, and type yes  
to continue.  
In the Specify filename text box, enter a name for the file to be saved, or  
browse to a previously saved file if you want to overwrite an existing file.  
280  
Troubleshooting: Your Storage Array Software  
   
Use the convention filename.dmp for the name of the file. The suffix  
.dmp is added to the file automatically if you do not specify a suffix for the  
file.  
4
5
Click Start.  
NOTE: Each test shows a status of Executing while it is in progress. The test  
then shows Completed when it successfully finishes. If any of the tests cannot  
be completed, a status of Failed appears in the Execution summary window.  
Monitor the progress and completion status of all of the tests. When they  
finish, click OK to close the State Capture dialog.  
Clicking Cancel stops the state capture process, and any remaining tests  
do not complete. Any test information that has been generated to that  
point is saved to the state capture file.  
NOTE: See the PowerVault Modular Disk Storage Manager online help topics  
for more information on troubleshooting, and recovering from failures.  
SMrepassist Utility  
SMrepassist (replication assistance) is a host-based utility for Windows  
platforms. This utility is installed with MDSM. Use this utility before and  
after you create a virtual disk copy on a Windows operating system to ensure  
that all the memory-resident data for file systems on the target virtual disk is  
flushed and that the driver recognizes signatures and file system partitions.  
You can also use this utility to resolve duplicate signature problems for  
snapshot virtual disks.  
From a command prompt window on a host running Windows, navigate to  
C:\Program Files\Dell\MD Storage Manager\util and run the following  
command:  
SMrepassist -f <filesystem-identifier>  
where -f flushes all the memory-resident data for the file system indicated by  
<filesystem-identifier>, and <filesystem-identifier> specifies a unique file  
system in the following syntax:  
drive-letter: <mount-point-path>  
The file system identifier may consist of only a drive letter, for example:  
SMrepassist -f E:  
Troubleshooting: Your Storage Array Software  
281  
   
NOTE: In Windows, the mount point path is a drive letter.  
An error message appears in the command line when the utility cannot  
distinguish between the following:  
Source virtual disk and snapshot virtual disk (for example, if the snapshot  
virtual disk has been removed).  
Standard virtual disk and virtual disk copy (for example, if the virtual disk  
copy has been removed).  
Unidentified Devices  
An unidentified node or device occurs when MDSM cannot access a new  
storage array. Causes for this error include network connection problems, the  
storage array is turned off, or the storage array does not exist.  
NOTE: Before beginning any recovery procedure, ensure that the host context  
agent software is installed and running. If you host is started before it is connected  
to the storage array, the host context agent software will not be able to find the  
storage array. If so, ensure that the connections are tight, and restart the host  
context agent software.  
If a storage array is managed by using both out-of-band management and  
in-band management using the same host, a management network  
connection problem might prevent direct communication with the storage  
array. However, you might still be able to manage the storage array over the  
in-band connections. The opposite situation can also occur.  
If a storage array is managed through more than one host, it is possible  
that the storage array might become unresponsive to communication over  
the connections given by one host. However, you might still be able to  
manage the storage array over the connections provided by another host.  
Recovering From an Unidentified Storage Array  
1
2
3
Ensure that the network connection to the storage management station  
is operating.  
Ensure that the controllers are installed and that the power is turned on to  
the storage array. Correct any existing problems before continuing.  
If you have an in-band storage array, use the following procedure. Click  
Refresh after each step to check the results:  
282  
Troubleshooting: Your Storage Array Software  
       
a
b
Ensure that the host context agent software is installed and running. If  
the host is started before it is connected to the controllers in the  
storage array, the host context agent software will not be able to find  
the controllers. If so, ensure that the connections are tight, and restart  
the host context agent software.  
Ensure that the network can access the host by using the ping  
command in the following syntax:  
ping <host-name-or-IP-address-of-the-host>  
.
If the network can access the host, continue to step c. If the network  
cannot access the host, skip to step d.  
c
Remove the host with the unresponsive status from the MDSM, and  
add that host again.  
If the host returns to optimal status, you have completed this procedure.  
d
e
f
Ensure that the power to the host is turned on and that the host  
is operational.  
If applicable, ensure that the host bus adapters have been installed in  
the host.  
Examine all external cables and switches or hubs to ensure that you  
cannot see any damage and that they are tightly connected.  
g
If you have recently replaced or added the controller, restart the host  
context agent software so that the new controller is found.  
If a problem exists, make the appropriate modifications to the host.  
4
If you have an out-of-band storage array, use the following procedure. Click  
Refresh after each step to ensure the results:  
a
Ensure that the network can access the controllers by using the ping  
command. Use the following syntax:  
ping <controller-IP-address>.  
If the network can access the controllers, continue to step b. If the  
network cannot access the controllers, skip to step c.  
b
Remove the storage array with the unresponsive status from MDSM,  
and add that storage array again.  
Troubleshooting: Your Storage Array Software  
283  
 
If the storage array returns to optimal status, you have completed  
this procedure.  
c
d
Examine the ethernet cables to ensure that you cannot see any  
damage and that they are tightly connected.  
Ensure that the applicable network configuration tasks have been done  
(for example, the IP addresses have been assigned to each controller).  
5
Ensure that the controller firmware is compatible with MDSM on your  
management station. If the controller firmware was upgraded, the MDSM  
might not have access to the storage array. A new version of MDSM might  
be needed to manage the storage array with the new version of the  
controller firmware.  
If this problem exists, see the Dell support website at support.dell.com.  
6
7
Look to see if there is too much network traffic to one or more controllers.  
This problem corrects itself because the MDSM tries to re-establish  
communication with the controllers in the storage array at regular times. If  
the storage array was unresponsive and a subsequent try to connect to the  
storage array succeeds, the storage array becomes responsive.  
For an out-of-band storage array, look to see if management operations  
are taking place on the storage array from other storage management  
stations. The type of management operations being done and the  
number of management sessions taking place together establish the  
number of TCP/IP connections made to a controller. When the  
maximum number of TCP/IP connections have been made, the  
controller stops responding. This problem corrects itself because after  
some TCP/IP connections complete, the controller then becomes  
responsive to other connection tries.  
8
If the storage array is still unresponsive, problems might exist with  
the controllers.  
If these problems exist, see the Dell support website at support.dell.com.  
284  
Troubleshooting: Your Storage Array Software  
 
Starting or Restarting the Host Context Agent  
Software  
The host context agent software module is the software component that  
resides on the server or management station that communicates with the  
MD3600f series storage arrays. The SMagent software automatically starts  
after you reboot the host.  
Windows  
To restart the SMagent software in Windows:  
1
Click Start  
Settings  
Control Pane  
Administrative Tools  
Services  
.
or  
Click Start  
Administrative Tools  
Services.  
2
3
In the Services dialog, select Modular Disk Storage Manager Agent  
.
If the modular disk storage manager agent is running, click Action  
Stop  
,
then wait approximately 5 seconds.  
4
Click Action  
Start.  
Linux  
To start or restart the host context agent software in Linux, enter the  
following command at the prompt:  
SMagent start  
The SMagent software might take a little time to initialize. The cursor is  
shown, but the terminal window does not respond. When the program starts,  
the following message is displayed:  
SMagent started.  
After the program completes the startup process, text similar to the following  
messages is displayed:  
Modular Disk Storage Manager Agent, Version  
90.02.A6.14  
Built Wed Feb 03 06:17:50 CST 2010  
Troubleshooting: Your Storage Array Software  
285  
   
Copyright (C) 2009-2010 Dell, Inc. All rights  
reserved.  
Checking device <n/a> (/dev/sg10): Activating  
Checking device /dev/sdb (/dev/sg11): Skipping  
Checking device <n/a> (/dev/sg3): Activating  
Checking device <n/a> (/dev/sg4): Activating  
Checking device <n/a> (/dev/sg5): Activating  
Checking device <n/a> (/dev/sg6): Activating  
Checking device <n/a> (/dev/sg7): Activating  
Checking device <n/a> (/dev/sg8): Activating  
Checking device <n/a> (/dev/sg9): Activating  
286  
Troubleshooting: Your Storage Array Software  
Troubleshooting: Your Storage Array Software  
287  
288  
Troubleshooting: Your Storage Array Software  
20  
Troubleshooting: Your Array  
Safety First—For You and Your Array  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
Troubleshooting Storage Array Startup Failure  
If your system halts during startup, check if:  
The array emits a series of beeps.  
The array fault LEDs are lit. See "RAID Controller Modules" on page 31.  
There is a constant scraping or grinding sound when you access the hard  
drive. See "Getting Help" on page 303.  
Troubleshooting Loss of Communication  
For information about troubleshooting loss of communication, see  
"Troubleshooting Array and Expansion Enclosure Connections" on page 297.  
Troubleshooting an SFP Transceiver  
CAUTION: When handling static-sensitive devices, take precautions to avoid  
damaging the product from static electricity.  
NOTE: Before proceeding with this section see "Working with SFP Modules" and  
"Fiber Optic Cables" in the Dell PowerVault MD3600f Deployment Guide.  
Equip yourself with antistatic protection and a replacement small form-factor  
pluggable (SFP) transceiver before replacing an SFP transceiver in the RAID  
controller module. Also, see the initial setup information for the enclosure to  
verify LED definitions.  
Troubleshooting: Your Array  
289  
                 
1
2
3
If possible, use the storage management software to create, save, and print  
a new storage array profile.  
If the Recovery Guru directed you to replace a failed RAID controller module,  
go to step 3, else run the Recovery Guru to identify the failed component.  
Unpack the new SFP transceiver.  
NOTE: Ensure that the SFP transceiver is replaced by the same type.  
NOTE: Set the new SFP transceiver on a dry, level surface near the RAID  
controller module, the RAID controller module enclosure, or the  
expansion enclosure.  
NOTE: Save all the packing materials in case you need to return the  
SFP transceiver.  
4
To locate a failed SFP transceiver, view the FCIN (host channel) speed  
has failed, both FCIN speed LEDs for a particular port are off.  
If both FCIN speed LEDs are off for a particular port with a known  
good cable, the SFP transceiver must be replaced. Go to step 6.  
If at least one FCIN speed LED is on for a particular port, the SFP  
transceiver is functional. The FCIN speed LEDs indicate a channel  
speed of 2 Gbps, 4 Gbps, or 8 Gbps (see "Host Channel LED  
Link/Rate Indications" on page 34). Ensure that the speed indicated  
by the LEDs is what is expected.  
WARNING: Data processing environments can contain equipment transmitting  
on system links with laser modules that operate at greater than Class 1 power  
levels. Never look into the end of an optical fiber cable or open receptacle.  
CAUTION: To prevent degraded performance, do not twist, fold, pinch, or step on  
fiber optic cables. Do not bend the fiber optic cables tighter than a 5 cm (2") radius.  
5
6
If present, disconnect the fiber optic cable from the failed SFP transceiver.  
Remove the failed SFP transceiver from the interface port.  
290  
Troubleshooting: Your Array  
   
Figure 20-1. Removing an SFP Module  
7
Install the new SFP transceiver into the interface port. Ensure that the  
transceiver is properly seated.  
8
9
Reconnect the fiber optic cable.  
View the FCIN speed LEDs and the Physical Disk Channel speed LEDs.  
Based on the LED status, perform one of these actions:  
At least one of the FCIN LEDs for each port is on. Go to step 11.  
Both the FCIN LEDs on an MD storage array RAID controller module  
Check that the SFP transceiver has been installed correctly. Reinstall  
the SFP transceiver if necessary.  
NOTE: If the storage array is not in an Optimal status, click the Recovery Guru  
toolbar button in the Array Management Window to determine if any other  
actions are required.  
10 If the problem is resolved, go to step 11. If the problem persists, see  
"Getting Help" on page 303.  
11 Create, save, and print a new storage array profile.  
Troubleshooting: Your Array  
291  
 
Troubleshooting External Connections  
Verify that the cables are connected to the correct ports before  
troubleshooting any external devices. For the location of the back-panel  
connectors on your array, see Figure 3-1.  
Ensure that all the cables are securely attached to the external connectors  
on your array.  
For information on cabling, see the Dell PowerVault MD3600f  
Deployment Guide  
.
Troubleshooting Power Supply/Cooling Fan  
Module  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
CAUTION: It is recommended that you turn off the host server before turning off  
the array to prevent loss of data.  
1
Locate the faulty power supply and determine the status of the LEDs.  
If the AC power LED is not lit, check the power cord and power source  
into which the power supply is plugged.  
Connect another device to the power source to verify if it is working.  
Connect the cable to a different power source.  
Replace the power cable.  
If the problem is not resolved, see "Getting Help" on page 303.  
If the DC power LED is not lit, verify that the power switch is turned  
on. If the power switch is turned on, see step 2.  
If the power supply’s fault indicator is lit, see "Getting Help" on page 303.  
292  
Troubleshooting: Your Array  
       
CAUTION: Power supply/cooling fan modules are hot-swappable. The array can  
operate on a single power supply; however both modules must be installed to  
ensure proper cooling. A single power supply/cooling fan module can be removed  
from a powered-on array for a maximum period of five minutes. Beyond that time,  
the array may automatically shut down to prevent damage.  
2
Reseat the power supply by removing and reinstalling it. See "Power Supply  
and Cooling Fan Features" on page 29.  
NOTE: After installing a power supply, allow several seconds for the array to  
recognize the power supply and to determine if it is working properly.  
If the problem is not resolved, see "Getting Help" on page 303.  
Troubleshooting Array Cooling Problems  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
Ensure that none of the following conditions exist:  
Array cover or drive blank is removed.  
Ambient temperature is too high. See “Technical Specifications” in the  
Getting Started Guide  
.
External airflow is obstructed.  
The power supply/cooling fan module is removed or has failed. See  
"Troubleshooting Power Supply/Cooling Fan Module" on page 292.  
If the problem is not resolved, see "Getting Help" on page 303  
.
Troubleshooting: Your Array  
293  
     
Troubleshooting Expansion Enclosure  
Management Modules  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
CAUTION: It is recommended that you turn off the host server before turning off  
the expansion enclosure array to prevent loss of data.  
If the EMM status LED is solid or blinking amber (2 or 4 times per sequence):  
a
b
Turn off the server.  
Remove the EMM and verify that the pins on backplane and EMM are  
not bent. See Removing an EMM in the MD1200 and MD1220  
Storage Enclosures Hardware Owner's Manual  
.
c
Reseat the EMM module and wait for 30 seconds. See Removing an  
EMM in the MD1200 and MD1220 Storage Enclosures Hardware  
Owner's Manual  
.
d
e
f
Check the EMM status LED.  
If the LED does not turn Green, replace the EMM.  
If the problem is not resolved, step c.  
If EMM status LED is blinking amber (5 times per sequence), update the  
firmware to the latest supported firmware on both the EMMs. For more  
information about downloading the latest firmware, see "Management:  
Firmware Downloads" on page 247.  
If the link status LEDs are not green:  
a
b
c
Turn off the server.  
Reseat the cables on the expansion array and the server.  
Turn on the expansion arrays and then the storage array and wait until  
the system is fully booted.  
294  
Troubleshooting: Your Array  
   
d
e
Turn on the server.  
Check the link status LED. If the link status LED is not green, replace  
the cables.  
If the problem is not resolved, see "Getting Help" on page 303.  
Troubleshooting RAID Controller Modules  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
CAUTION: In the case of non-redundant configurations, it is recommended that  
you turn off the host server before turning off the array to prevent loss of data.  
If the array status LED is solid or blinking amber:  
a
In the AMW, select the Summary tab, and click on Storage Array  
needs attention. Follow the listed procedures in the recovery guru(s)  
"Recovery Guru" on page 270.  
b
complete the following procedure to further troubleshoot the array.  
c
d
Turn off the host server as appropriate.  
Remove the RAID controller module and verify that the pins on  
backplane and RAID controller module are not bent. See "Removing a  
RAID Controller Module Blank" on page 229.  
e
Reinstall the RAID controller module and wait for 30 seconds. See  
"Installing a RAID Controller Module" on page 232.  
f
Check the RAID controller module status LED.  
Replace the RAID controller module.  
Turn on the host server.  
g
h
If the problem is not resolved, see "Getting Help" on page 303.  
Troubleshooting: Your Array  
295  
 
If both LEDs for any given FCIN port are unlit, see "Troubleshooting Array  
and Expansion Enclosure Connections" on page 297.  
a
b
Turn off the server, storage arrays, and expansion enclosures.  
Reseat the RAID controller module and reconnect cables on the  
storage array and the server.  
c
d
e
Restart the storage array and wait until the array is fully booted.  
Turn on the server.  
Recheck the LEDs of the affected port(s). Replace the fiber optic  
cables of any port(s) where both LEDs are unlit.  
NOTE: The SAS OUT status LED should be green.  
If the problem is not resolved, see "Getting Help" on page 303.  
Troubleshooting Hard Drives  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
1
2
Check the storage array profile to ensure that the most current version of  
the firmware is installed. For more information see the Support Matrix at  
support.dell.com/manuals  
.
Remove the hard drive from the system. See "Removing a Hard Drive" on  
page 224.  
3
Check the hard drives and the backplane to ensure that the connectors are  
not damaged.  
4
5
Reinstall the hard drive.  
If the problem is not resolved, replace the failed hard drive.  
If the problem persists, see "Troubleshooting Loss of Communication" on  
page 289, "Troubleshooting an SFP Transceiver" on page 289, or "Getting  
Help" on page 303.  
296  
Troubleshooting: Your Array  
   
Troubleshooting Array and Expansion Enclosure  
Connections  
1
Verify that the SAS OUT status LED is green and that for each FCIN port  
connected to a cable, at least one of the paired LEDs is on. If this is not  
the case, see "Planning: RAID Controller Modules" on page 31.  
2
Ensure that all the cables are attached correctly according to array mode  
you selected.  
3
4
Turn off the server, storage array, and expansion enclosures.  
Reseat the RAID controller module and reconnect cables on the storage  
array and the server.  
5
Turn on the expansion arrays and then the storage array and wait until the  
system is fully booted.  
6
7
Turn on the server.  
Check the LEDs of all cabled ports. Replace the fiber optic cables of any  
port(s) where both LEDs are unlit.  
If the problem is not resolved, see "Getting Help" on page 303.  
Reboot the host server.  
8
NOTE: You must turn off the host server before resetting the cables on  
the storage array or expansion enclosure.  
If the problem is not resolved, see "Getting Help" on page 303.  
Troubleshooting: Your Array  
297  
   
Troubleshooting a Wet Storage Array  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
1
2
Turn off the array and disconnect all the cables.  
Remove the following components from the array. See "Management:  
Installing Array Components" on page 221.  
Hard drives  
RAID controller modules  
Power supply/cooling fan modules  
Control panel  
Backplane  
3
4
5
Let the system dry thoroughly for at least 24 hours.  
Reinstall the components you removed in step 2.  
Connect all the cables and turn on the array.  
If the array does not start properly, see "Getting Help" on page 303.  
298  
Troubleshooting: Your Array  
     
Troubleshooting a Damaged Array  
CAUTION: Many repairs may only be done by a certified service technician.  
You should only perform troubleshooting and simple repairs as authorized in  
your product documentation, or as directed by the online or telephone service  
and support team. Damage due to servicing that is not authorized by Dell is not  
covered by your warranty. Read and follow the safety instructions that came  
with the product.  
1
Ensure that the following components are properly installed:  
Hard drives  
RAID controller modules  
Power supply/cooling fan modules  
Control panel  
Backplane  
2
3
Ensure that all the cables are properly connected and that there are no  
damaged pins in the connectors.  
Run diagnostics available in Dell PowerVault Modular Disk (MD) Storage  
Manager software. In the AMW, select a component in the Physical pane of  
the Physical tab. Select Advanced  
Troubleshooting  
Run Diagnostics.  
If the test fails, see "Getting Help" on page 303.  
Troubleshooting: Your Array  
299  
   
Troubleshooting RAID Controller Modules  
Conditions  
Certain events can cause a RAID controller module to fail and/or shut down.  
Unrecoverable ECC memory or PCI errors, or critical physical conditions can  
cause lockdown. If your RAID storage array is configured for redundant access  
and cache mirroring, the surviving controller can normally recover without  
data loss or shutdown.  
Typical controller failures are detailed in the following sections.  
Invalid Storage Array  
The RAID controller module is supported only in a Dell-supported storage  
array. After installation in the storage array, the controller performs a set of  
validation checks. The array status LED is lit with a steady amber color while  
the RAID controller module completes these initial tests and the controllers  
are booted successfully. If the RAID controller module detects a non-Dell  
supported storage array, the controller aborts startup. The RAID controller  
module does not generate any events to alert you in the event of an invalid  
array, but the array status LED is lit with a flashing amber color to indicate a  
fault state.  
For full details on the LEDs and their interpretation, see "Back Panel Features  
and Indicators" on page 27.  
ECC Errors  
RAID controller firmware can detect ECC errors and can recover from a  
single-bit ECC error whether the RAID controller module is in a redundant  
or nonredundant configuration. A storage array with redundant controllers  
can recover from multi-bit ECC errors as well because the peer RAID  
controller module can take over, if necessary.  
The RAID controller module failsover if it experiences up to 10 single-bit  
errors, or up to three multi-bit errors.  
300  
Troubleshooting: Your Array  
   
PCI Errors  
The storage array firmware can detect and only recover from PCI errors when  
the RAID controller modules are configured for redundancy. If a virtual disk  
uses cache mirroring, it fails over to its peer RAID controller module, which  
initiates a flush of the dirty cache.  
Critical Conditions  
The storage array generates a critical event if the RAID controller module  
detects a critical condition that could cause immediate failure of the array  
and/or loss of data. The storage array is in a critical condition if one of the  
following occurs:  
More than one fan has failed  
Any backplane temperature sensors in the critical range  
Backplane/power supply failure  
Two or more temperature sensors are unreadable  
Failure to detect or unable to communicate with peer port  
NOTE: If both RAID controller modules fail simultaneously, the enclosure cannot  
issue critical or noncritical event alarms for any enclosure component.  
When the enclosure is under critical condition, its enclosure status LED  
blinks amber.  
Noncritical Conditions  
A noncritical condition is an event or status that does not cause immediate  
failure, but must be corrected to ensure continued reliability of the storage  
array. Examples of noncritical events include the following:  
One power supply has failed  
One cooling fan has failed  
One RAID controller module in a redundant configuration has failed  
A battery has failed or has been removed  
A physical disk in a redundant virtual disk has failed  
When the enclosure is under noncritical condition, its enclosure status LED  
blinks amber.  
Troubleshooting: Your Array  
301  
302  
Troubleshooting: Your Array  
21  
Getting Help  
Contacting Dell  
For customers in the United States, call 800-WWW-DELL (800-999-3355).  
NOTE: If you do not have an active Internet connection, you can find contact  
information on your purchase invoice, packing slip, bill, or Dell product catalog.  
Dell provides several online and telephone-based support and service options.  
Availability varies by country and product, and some services may not be  
available in your area. To contact Dell for sales, technical support, or  
customer service issues:  
1
2
Visit support.dell.com.  
Verify your country or region in the Choose A Country/Region drop-down  
menu at the bottom of the page.  
3
4
5
Click Contact Us on the left side of the page.  
Select the appropriate service or support link based on your need.  
Choose the method of contacting Dell that is convenient for you.  
Getting Help  
303  
         
304  
Getting Help  
Index  
A
Access Virtual Disk, 68  
Advanced Feature  
Using Snapshot and Disk Copy  
Together, 5 7  
133  
Segment Size of a Virtual  
Disk, 107  
Virtual Disk Modification  
Advanced Features, 53  
Snapshot Repository Virtual  
Disk, 55  
Snapshot Virtual Disks, 54  
Advanced Path, 148  
Array Management Types  
In-Band Management, 68  
Out-of-Band Management, 67  
Physical Disk Type, 109  
Configuring  
Host Access, 87  
Hot Spare Physical Disks, 118  
B
backplane  
installing, 245  
removing, 242  
Configuring Alert Notifications  
SNMP, 81  
Contacting Dell, 303  
contacting Dell, 303  
Battery Settings, 83  
control panel  
installing, 241  
removing, 239  
C
Change  
Copy Manager, 167  
Controller Ownership of the  
Virtual Disk, 127  
I/O Type, 108  
Index  
305  
 
D
Defining a Host, 88  
Dell  
contacting, 303  
Disk Group  
Creating, 100  
Expansion, 135  
Export, 137  
Enclosure Loss Protection, 121  
Exporting, 138  
Import, 138  
Locating, 102  
Migration, 137  
Disk Group and Virtual Disk  
Expansion, 135  
Disk Group Operations, 48  
Defragmentation, 49  
Expansion, 49  
Failed RAID Controller  
Limit, 49  
Features and Indicators  
RAID Level Migration, 48  
Segment Size Migration, 48  
Virtual Disk Capacity  
Expansion, 49  
View, 259  
Disk Groups and Virtual Disks  
Creating, 99  
Free Capacity, 136  
Download  
installing, 222  
removing, 222  
NVSRAM Firmware, 250  
Physical Disk Firmware, 253  
RAID Controller and NVSRAM  
Firmware, 248  
RAID Controller and NVSRAM  
Packages, 247  
RAID controller module  
Firmware, 255  
H
hard drive  
drive carrier, 227  
installing, 226  
Index  
306  
removing, 224  
backplane, 245  
drive blank, 224  
power supply/cooling fan  
Hard-Drive Indicator  
Patterns, 28  
Hardware Features  
Back panel features, 27  
Front panel features, 24  
features, 29  
Host Group  
Adding, 91  
Create, 91  
L
Moving a Host, 92  
Removing a host, 91  
Host Topology, 93  
Host-to-Virtual Disk  
Mapping, 124  
Host-to-Virtual Disk Mappings  
Creating, 124  
Modifying and Removing, 126  
Removing, 128  
Identifiers, 95  
MDSM, 20  
Hot Spare  
Media Errors and Unreadable  
Drive Protection, 121  
Global Hot Spares, 120  
Operation, 120  
Changing settings, 140  
Suspending, 141  
Hot Spares and Rebuild, 120  
Microsoft  
Virtual Disk Service, 261  
Volume Shadow-Copy  
Service, 261  
I
I/O Data Path Protection, 94  
Inside the box, 20  
Microsoft Services  
Index  
307  
Virtual Disk Copy, 55  
Preferred RAID Controller  
Module Ownership, 167  
Monitoring Performance, 59  
Preparing Host Servers  
Multi-Path  
Preferred and Alternate  
Controllers and Paths, 57  
Multi-Path Software, 57  
R
RAID, 43  
group, 133  
N
Non-Exportable  
RAID 0, 43  
RAID 1, 44  
RAID 5, 44  
Usage, 43  
Components, 137  
O
Other Information, 21  
P
phone numbers, 303  
Remote Replication  
Activation, 184  
Host Access, 177  
Encrypting Disk, 110  
Physical Disk States, 40  
Physical Disks, 40  
Erasing Secure, 117  
Unlocking Secure, 117  
Virtual Disks, 178  
removing  
Physical Disks, Virtual Disks,  
and Disk Groups, 39  
backplane, 242  
control panel MD1200, 239  
drive blank, 223  
EMM, 231  
EMM blank, 229  
front bezel, 222  
hard drive, 224  
Power Indicator Codes, 30  
Power Supply and Cooling Fan  
Features, 29  
Index  
308  
hard drive from a drive carrier, 227  
power supply/cooling fan  
module, 237  
Context Agent, 93  
Storage Array  
Removing Copy Pairs, 173  
Removing Host Access, 90  
Restricted Mappings, 130  
Storage Array Media Scan, 139  
S
Safety, 19  
safety, 289  
Security Key  
Changing, 114  
Creating, 112  
Saving, 115  
Segment Size, 45  
Setting a Password, 73  
Setting Copy Priority, 169  
Simple Path, 145  
Troubleshooting  
SMART, 41  
Information, 280  
Collecting the Physical Disk  
Finding Nodes, 273  
SMrepassist Utility, 281  
Snapshot Repository  
Capacity, 154  
Snapshot Virtual Disk  
Creating using advanced  
path, 145  
Locating an Expansion  
Enclosure, 279  
Creating using simple path, 144  
Recovering from an Unidentified  
Storage Array, 282  
Recovering from an Unresponsive  
Storage Array Condition, 276  
Recovery Guru, 270  
Snapshot Virtual Disks  
Disabling, 158  
Re-create, 160  
Re-creating, 157  
Index  
309  
Starting or Restarting the  
Host-Agent Software, 285  
Start-Up Routine, 263  
Storage Array Profile, 270  
Storage Array Support Data, 266  
Unidentified Devices, 282  
Viewing the Logical  
Background Initialization, 46  
Copy and Modification  
Associations, 272  
Viewing the Physical  
Associations, 273  
troubleshooting, 289  
connections, 297  
cooling problems, 293  
damaged enclosure, 299  
external connections, 292  
hard drives, 296  
loss of communication, 289  
module, 292  
startup failure, 289  
wet enclosure, 298  
Foreground Initialization, 46  
Recopying, 170  
Recovery, 56  
Stopping copy, 170  
Storage Array Performance, 169  
Virtual Disk Copy  
Virtual Disk Expansion, 136  
Roaming  
Disk Migration, 51  
Disk Roaming, 53  
U
Unconfigured Capacity, 137  
User Interface  
AMW, 65  
Virtual Disk Operations, 45  
EMW, 64  
Overview, 63  
Virtual Disk Operations  
Limit, 47  
Using Go To, 275  
Virtual Disk Ownership, 58  
Virtual Disk States, 42  
V
Virtual DiskCopy  
Target, 5 6  
Virtual Disk  
Index  
310  
Virtual Disks and Disk  
Groups, 41  
Z
Zoning  
Guidelines, 182  
Switch Configurations, 182  
Index  
311  
312  
Index  

Hotpoint 6505MKII User Manual
Haier HTR20 User Manual
Gateway FPD1775W User Manual
Frigidaire FPEC3685KS User Manual
Clarion OHM643 User Manual
Cisco Systems UC520W8UK9J User Manual
Cadco OST 34A User Manual
Audiovox DT 911 User Manual
Asus Computer Monitor 27T1EH User Manual
American Megatrends APOLLO MAN 772 User Manual