SPARC Enterprise M3000/M4000/M5000/
M8000/M9000 Servers
Product Notes for XCP Version 1124
Manual Code: C120-0066-01EN
August 2018, Revision A
Copyright © 2018, Fujitsu Limited. Tous droits réservés.
Oracle et/ou ses sociétés affiliées ont fourni et vérifié des données techniques de certaines parties de ce composant.
Oracle et/ou ses sociétés affiliées et Fujitsu Limited détiennent et contrôlent chacune des droits de propriété intellectuelle relatifs aux produits et
technologies décrits dans ce document. De même, ces produits, technologies et ce document sont protégés par des lois sur le copyright, des brevets,
d’autres lois sur la propriété intellectuelle et des traités internationaux.
Ce document, le produit et les technologies afférents sont exclusivement distribués avec des licences qui en restreignent l’utilisation, la copie, la
distribution et la décompilation. Aucune partie de ce produit, de ces technologies ou de ce document ne peut être reproduite sous quelque forme que ce
soit, par quelque moyen que ce soit, sans l’autorisation écrite préalable d’Oracle et/ou ses sociétés affiliées et de Fujitsu Limited, et de leurs éventuels
bailleurs de licence. Ce document, bien qu’il vous ait été fourni, ne vous confère aucun droit et aucune licence, expresses ou tacites, concernant le produit
ou la technologie auxquels il se rapporte. Par ailleurs, il ne contient ni ne représente aucun engagement, de quelque type que ce soit, de la part d’Oracle ou
de Fujitsu Limited, ou des sociétés affiliées de l’une ou l’autre entité.
Ce document, ainsi que les produits et technologies qu’il décrit, peuvent inclure des droits de propriété intellectuelle de parties tierces protégés par
copyright et/ou cédés sous licence par des fournisseurs à Oracle et/ou ses sociétés affiliées et Fujitsu Limited, y compris des logiciels et des technologies
relatives aux polices de caractères.
Conformément aux conditions de la licence GPL ou LGPL, une copie du code source régi par la licence GPL ou LGPL, selon le cas, est disponible sur
demande par l’Utilisateur final. Veuillez contacter Oracle et/ou ses sociétés affiliées ou Fujitsu Limited.
Cette distribution peut comprendre des composants développés par des parties tierces.
Des parties de ce produit peuvent être dérivées des systèmes Berkeley BSD, distribués sous licence par l’Université de Californie. UNIX est une marque
déposée aux États-Unis et dans d’autres pays, distribuée exclusivement sous licence par X/Open Company, Ltd.
Oracle et Java sont des marques déposées d’Oracle Corporation et/ou de ses sociétés affiliées. Fujitsu et le logo Fujitsu sont des marques déposées de
Fujitsu Limited.
Toutes les marques SPARC sont utilisées sous licence et sont des marques déposées de SPARC International, Inc., aux États-Unis et dans d’autres pays. Les
produits portant la marque SPARC reposent sur des architectures développées par Oracle et/ou ses sociétés affiliées. SPARC64 est une marque de SPARC
International, Inc., utilisée sous licence par Fujitsu Microelectronics, Inc. et Fujitsu Limited. Tout autre nom mentionné peut correspondre à des marques
appartenant à d’autres propriétaires.
United States Government Rights - Commercial use. U.S. Government users are subject to the standard government user license agreements of Oracle
and/or its affiliates and Fujitsu Limited and the applicable provisions of the FAR and its supplements.
Avis de non-responsabilité : les seules garanties octroyées par Oracle et Fujitsu Limited et/ou toute société affiliée de l’une ou l’autre entité en rapport avec
ce document ou tout produit ou toute technologie décrits dans les présentes correspondent aux garanties expressément stipulées dans le contrat de licence
régissant le produit ou la technologie fournis. SAUF MENTION CONTRAIRE EXPRESSÉMENT STIPULÉE DANS CE CONTRAT, ORACLE OU FUJITSU
LIMITED ET LES SOCIÉTÉS AFFILIÉES À L’UNE OU L’AUTRE ENTITÉ REJETTENT TOUTE REPRÉSENTATION OU TOUTE GARANTIE, QUELLE
QU’EN SOIT LA NATURE (EXPRESSE OU IMPLICITE) CONCERNANT CE PRODUIT, CETTE TECHNOLOGIE OU CE DOCUMENT, LESQUELS
SONT FOURNIS EN L’ÉTAT. EN OUTRE, TOUTES LES CONDITIONS, REPRÉSENTATIONS ET GARANTIES EXPRESSES OU TACITES, Y COMPRIS
NOTAMMENT TOUTE GARANTIE IMPLICITE RELATIVE À LA QUALITÉ MARCHANDE, À L’APTITUDE À UNE UTILISATION PARTICULIÈRE
OU À L’ABSENCE DE CONTREFAÇON, SONT EXCLUES, DANS LA MESURE AUTORISÉE PAR LA LOI APPLICABLE. Sauf mention contraire
expressément stipulée dans ce contrat, dans la mesure autorisée par la loi applicable, en aucun cas Oracle ou Fujitsu Limited et/ou l’une ou l’autre de leurs
sociétés affiliées ne sauraient être tenues responsables envers une quelconque partie tierce, sous quelque théorie juridique que ce soit, de tout manque à
gagner ou de perte de profit, de problèmes d’utilisation ou de perte de données, ou d’interruptions d’activités, ou de tout dommage indirect, spécial,
secondaire ou consécutif, même si ces entités ont été préalablement informées d’une telle éventualité.
LA DOCUMENTATION EST FOURNIE « EN L’ÉTAT » ET TOUTE AUTRE CONDITION, DÉCLARATION ET GARANTIE, EXPRESSE OU TACITE, EST
FORMELLEMENT EXCLUE, DANS LA MESURE AUTORISÉE PAR LA LOI EN VIGUEUR, Y COMPRIS NOTAMMENT TOUTE GARANTIE
IMPLICITE RELATIVE À LA QUALITÉ MARCHANDE, À L’APTITUDE À UNE UTILISATION PARTICULIÈRE OU À L’ABSENCE DE
CONTREFAÇON.
Preface xi
1. Software Requirements
1
XCP, Oracle Solaris OS and Required Patches
Obtaining XCP and Oracle Solaris Patches
Web Browser
Existing XCP Firmware Versions and Support Information
1
4
4
5
What’s New in XCP 1124
Notes on Active Directory
Notes on LDAP/SSL
Notes on Power Consumption Monitoring Function (M3000 Server)
Notes on Airflow Indicator
Notes on NTP Server
Notes on the NTP Server Referring to the Local Clock 10
7
8
9
9
9
9
v
Miscellaneous Notes and Limitations 13
Updating to XCP 1124 16
(M4000/M5000/M8000/M9000 Servers) 16
(M8000/M9000 Servers) 16
Oracle Solaris OS Issues and Workarounds 59
Oracle Solaris Issues and Workarounds for All Supported Releases 60
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/11 66
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 9/10 67
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/09 68
vi
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/08 72
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08 76
Servers XSCF User’s Guide 94
Configuring XSCF Routing 94
Updates of the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Identifying Degraded Memory in a System 98
Identifying Different Memory Sizes in a System Board 98
▼
▼
Using the showdevicesCommand 98
Using the prtdiagCommand to Identify Memory Size 99
Identifying Permanent Memory in a Target Board 100
Notes on DVD Drive and Discs 103
Notes on the Use of USB Memory 103
Power Control and Operator Panel Mode Switch 103
Limitation for 1027A-Z/X1027A-Z 104
Limitation for 4447A-Z/X4447A-Z 104
Contents
vii
Hardware RAID Function 105
Requirements for Hardware RAID Support 105
Hardware Documentation Updates 105
Updates of SPARC Enterprise M3000 Server Service Manual 107
Removing the Motherboard Unit 107
Limitation for 4447A-Z/X4447A-Z 111
Patches for Emulex PCI Express (PCIe) Adapters 112
Hardware Issues and Workarounds 112
Sun Crypto Accelerator 6000 112
Hardware Documentation Updates 112
Notes and Limitations 115
Notes on Hardware RAID 116
Notes on Using the Hardware RAID 116
Notes on Building or Deleting the Hardware RAID 116
Notes on Operating the Hardware RAID 116
Notes on raidctl(1M) Command 117
viii
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Updates of SPARC Enterprise M8000/M9000 Servers Service Manual 126
I/O Unit (IOU) Mounting Locations 126
Miscellaneous Information 127
Cabinet Stabilization Measures 127
Method of Securing the Cabinet to the Floor Surface 128
Contents
ix
x
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Preface
These product notes contain important and late-breaking information about the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 servers. It covers hardware,
software, firmware, and documentation, covering the latest XSCF Control Package
(XCP) release.
Note – The M3000/M4000/M5000/M8000/M9000 servers are
based on high-performance and highly-reliable SPARC64 VII+ or
SPARC64 VII processors.
Some references to server names and document names are abbreviated for
readability. For example, if you see a reference to the M9000 server, note that the full
product name is the SPARC Enterprise M9000 server. And if you see a reference to
the XSCF Reference Manual, note that the full document name is the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual.
Note – Once an XCP version newer than the XCP version supported by this manual
is released, only the manuals related to the new XCP version are subsequently
updated. You must therefore check the manuals for the latest XCP version in addition
to those related to the XCP version you are using.
Global Site
Japanese Site
http://www.fujitsu.com/jp/products/computing/servers/unix/sparc-
enterprise/downloads/manual/
xi
This section includes:
■
■
■
■
“Audience” on page xii
“How to Use This Document” on page xii
“Related Documentation” on page xiii
“Documentation Feedback” on page xvi
Audience
This notes is written for experienced system administrators with working knowledge
of computer networks and advanced knowledge of the Oracle Solaris Operating
System (Oracle Solaris OS).
How to Use This Document
This document is intended for all models of the
M3000/M4000/M5000/M8000/M9000 servers. Read the appropriate related items
depending on your server, as shown in the table below.
Chapter of This Document
M3000
O
M4000/M5000
M8000/M9000
Chapter 1 Software Requirements
O
O
O
O
O
O
Chapter 2 XCP 1124 and XCP 1124-related Information
Chapter 3 Information About Software
Chapter 4 Information About Hardware of M3000 Server
O
O
O
Chapter 5 Information About Hardware of M4000/M5000
Servers
O
Chapter 6 Information About Hardware of M8000/M9000
Servers
O
Chapter 1, Chapter 2, and Chapter 3 provide the information related to firmware and
software for all servers. These chapters contain the descriptions which are specific to
particular models, too. The relevant server name is clearly stated.
xii
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Chapter 4, Chapter 5, and Chapter 6 provide the information related to hardware.
Each chapter describes each model. The information which is common in all models
is provided in all of these chapters. As a result, the description may be overlapped
when you retrieve the hardware-related information of multiple models.
Related Documentation
All documents for your server are available online at the following locations:
Manual
Link
(Oracle Solaris OS, and so on)
Fujitsu documents (Global Site)
http://www.fujitsu.com/global/products/
enterprise/downloads/manuals/
Fujitsu documents (Japanese Site)
http://www.fujitsu.com/jp/products/comp
enterprise/downloads/manual/
Note – Information in these product notes supersedes the information in the
M3000/M4000/M5000/M8000/M9000 servers documentation set.
The following table lists titles of related documents.
Related SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers Documents
SPARC Enterprise M3000 Server Site Planning Guide
Manual Code
C120-H030
C120-H015
C120-H014
C120-H016
C120-E536
C120-E345
C120-E323
C120-E537
SPARC Enterprise M4000/M5000 Servers Site Planning Guide
SPARC Enterprise M8000/M9000 Servers Site Planning Guide
SPARC Enterprise Equipment Rack Mounting Guide
*
SPARC Enterprise M3000 Server Getting Started Guide
SPARC Enterprise M4000/M5000 Servers Getting Started Guide*
SPARC Enterprise M8000/M9000 Servers Getting Started Guide*
SPARC Enterprise M3000 Server Overview Guide
Preface
xiii
Related SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers Documents
SPARC Enterprise M4000/M5000 Servers Overview Guide
Manual Code
C120-E346
C120-E324
C120-E633
SPARC Enterprise M8000/M9000 Servers Overview Guide
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers Important
Legal and Safety Information*
SPARC Enterprise M3000 Server Safety and Compliance Guide
SPARC Enterprise M4000/M5000 Servers Safety and Compliance Guide
SPARC Enterprise M8000/M9000 Servers Safety and Compliance Guide
External I/O Expansion Unit Safety and Compliance Guide
SPARC Enterprise M4000 Server Unpacking Guide*
C120-E538
C120-E348
C120-E326
C120-E457
C120-E349
C120-E350
C120-E327
C120-E539
C120-E351
C120-E328
C120-E540
C120-E352
C120-E330
C120-E329
C120-E361
SPARC Enterprise M5000 Server Unpacking Guide*
SPARC Enterprise M8000/M9000 Servers Unpacking Guide*
SPARC Enterprise M3000 Server Installation Guide
SPARC Enterprise M4000/M5000 Servers Installation Guide
SPARC Enterprise M8000/M9000 Servers Installation Guide
SPARC Enterprise M3000 Server Service Manual
SPARC Enterprise M4000/M5000 Servers Service Manual
SPARC Enterprise M8000/M9000 Servers Service Manual
External I/O Expansion Unit Installation and Service Manual
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI Build
Procedure
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers
Administration Guide
C120-E331
C120-E332
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s
Guide
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
Reference Manual
Varies per
release
SPARC Enterprise M4000/M5000/M8000/M9000 Servers Dynamic
Reconfiguration (DR) User’s Guide
C120-E335
C120-E336
C120-E360
SPARC Enterprise M4000/M5000/M8000/M9000 Servers Capacity on Demand
(COD) User’s Guide
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI User’s
Guide
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers Product
Notes
Varies per
release
†
xiv
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Related SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers Documents
Manual Code
SPARC Enterprise M3000 Server Product Notes
Varies per
release
SPARC Enterprise M4000/M5000 Servers Product Notes
SPARC Enterprise M8000/M9000 Servers Product Notes
Varies per
release
Varies per
release
External I/O Expansion Unit Product Notes
C120-E456
C120-E514
C120-H007
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers Glossary
Fujitsu M10/SPARC Servers/SPARC Enterprise/PRIMEQUEST Common
Installation Planning Manual
*
This is a printed document.
† Beginning with the XCP 1100 release.
Documentation CD
For the Documentation CD, please contact your local sales representative.
■
■
■
SPARC Enterprise M3000 Server Documentation CD (C120-E541)
SPARC Enterprise M4000/M5000 Servers Documentation CD (C120-E365)
SPARC Enterprise M8000/M9000 Servers Documentation CD (C120-E364)
Preface
xv
Information Required for Server Operation
The latest information about the support for your server are provided on the
websites.
■
Message
Global Site
http://www.fujitsu.com/global/products/computing/servers/unix/
sparc-enterprise/msg/
Japanese Site
http://www.fujitsu.com/jp/products/computing/servers/unix/spar
c-enterprise/msg/
■
Firmware program
For the information on how to get the latest files of firmware, contact your sales
representative.
The following files or document are provided.
■
Firmware program file (XSCF Control Package (XCP) file)
XSCF extension MIB (OPL-SP-MIB) definition file
■
Note – XSCF Control Package (XCP): XCP is a package which has the control
programs of hardware that configures a computing system. The XSCF firmware and
the OpenBoot PROM firmware are included in the XCP file.
■
Fault Management MIB (SUN-FM-MIB) definition file
Obtain the Fault Management MIB (SUN-FM-MIB) definition file from Oracle.
Documentation Feedback
If you have any comments or requests regarding this document, or if you find any
unclear statements in the document, please state your points specifically on the form
at the following website:
xvi
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
CHAPTER
1
This section contains the software requirements of SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 (M3000/M4000/M5000/M8000/M9000)
servers.
■
■
■
■
“XCP, Oracle Solaris OS and Required Patches” on page 1
“Obtaining XCP and Oracle Solaris Patches” on page 4
“Web Browser” on page 4
“Existing XCP Firmware Versions and Support Information” on page 5
XCP, Oracle Solaris OS and Required
Patches
The supported XCP and Oracle Solaris Operating System (Oracle Solaris OS)
depends on the processors being installed on the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 servers.
This section lists supported XCP, and Oracle Solaris OS and mandatory patches for
the M3000/M4000/M5000/M8000/M9000 servers.
Note – Apply the patches in the following order. For the procedures of CPU
upgrade including the patches, see SPARC Enterprise M8000/M9000 Servers Service
Manual.
1
TABLE 1-1
Supported XCP, Oracle Solaris OS, and Required Patches
Server: Processors
XCP
Oracle Solaris OS
Required Patches
M4000/M5000: SPARC64 VII+ 2.66 GHz 1100 or later • Oracle Solaris 11 11/11
No patches required
No patches required
No patches required
No patches required
All patches contained in
M8000/M9000: SPARC64 VII+ 3.00 GHz
• Oracle Solaris 10 8/11
• Oracle Solaris 10 9/10
• Oracle Solaris 10 10/09
• Oracle Solaris 10 5/09
**
the PTF R10021 or later
• Oracle Solaris 10 10/08
• Oracle Solaris 10 5/08
• Oracle Solaris 10 8/07
All patches contained in
**
the PTF R10021 or later
All patches contained in
**
The PTF R10021 or later
All patches contained in
**
The PTF R10021 or later
M4000/M5000: SPARC64 VII 2.53 GHz 1090 or later • Oracle Solaris 11 11/11
No patches required
No patches required
No patches required
No patches required
All patches contained in
M8000/M9000: SPARC64 VII 2.88 GHz
• Oracle Solaris 10 8/11
• Oracle Solaris 10 9/10
• Oracle Solaris 10 10/09
• Oracle Solaris 10 5/09
**
the PTF R10021 or later
• Oracle Solaris 10 10/08
• Oracle Solaris 10 5/08
• Oracle Solaris 10 8/07
All patches contained in
**
the PTF R10021 or later
All patches contained in
**
the PTF R10021 or later
All patches contained in
**
the PTF R10021 or later
*
M4000/M5000: SPARC64 VII 2.4 GHz
M8000/M9000: SPARC64 VII 2.52 GHz
1070 or later • Oracle Solaris 11 11/11
• Oracle Solaris 10 8/11
• Oracle Solaris 10 9/10
• Oracle Solaris 10 10/09
• Oracle Solaris 10 5/09
• Oracle Solaris 10 10/08
• Oracle Solaris 10 5/08
• Oracle Solaris 10 8/07
No patches required
No patches required
No patches required
No patches required
No patches required
No patches required
137137-09
119254-51 or later
125891-01 or later
127755-01 or later
127127-11
2
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 1-1
Supported XCP, Oracle Solaris OS, and Required Patches (Continued)
Server: Processors
XCP
Oracle Solaris OS
Required Patches
M4000/M5000: SPARC64 VI 2.15 GHz
1040 or later • Oracle Solaris 11 11/11
• Oracle Solaris 10 8/11
• Oracle Solaris 10 9/10
• Oracle Solaris 10 10/09
• Oracle Solaris 10 5/09
• Oracle Solaris 10 10/08
• Oracle Solaris 10 5/08
• Oracle Solaris 10 8/07
• Oracle Solaris 10 11/06
No patches required
No patches required
No patches required
No patches required
No patches required
No patches required
M8000/M9000: SPARC64 VI 2.28/2.4
GHz
137137-09
No patches required
††
118833-36
125100-04 or later
120068-03 or later
123839-07 or later
125424-01 or later
125075-01 or later
125670-02 or later
M3000: SPARC64 VII+ 2.86 GHz
M3000: SPARC64 VII 2.75 GHz
1101 or later • Oracle Solaris 11 11/11
• Oracle Solaris 10 8/11
No patches required
No patches required
No patches required
‡
• Oracle Solaris 10 9/10
1091 or later • Oracle Solaris 11 11/11
• Oracle Solaris 10 8/11
No patches required
No patches required
No patches required
No patches required
All patches contained in
• Oracle Solaris 10 9/10
• Oracle Solaris 10 10/09
• Oracle Solaris 10 5/09
**
The PTF R10021 or later
• Oracle Solaris 10 10/08
All patches contained in
The PTF R10021 or later
**
M3000: SPARC64 VII 2.52 GHz
1080 or
later*
• Oracle Solaris 11 11/11
• Oracle Solaris 10 8/11
• Oracle Solaris 10 9/10
• Oracle Solaris 10 10/09
• Oracle Solaris 10 5/09
• Oracle Solaris 10 10/08
• Oracle Solaris 10 5/08
No patches required
No patches required
No patches required
No patches required
No patches required
No patches required
†
119254-59, 138866-01,
137137-09, 138504-04
*
If 8GB DIMM is installed on a server, update XCP to XCP 1081 or later.
† XCP 1080 is the first XCP release for the M3000 server.
‡ When you install a version earlier than Oracle Solaris 10 9/10, onboard LAN (bge driver) does not work. Be sure to use Oracle Solaris
10 9/10 or later.
** For more information on the PTF, contact a sales representative or a qualified service engineer.
††Apply 118833-36 before 125100-04.
Chapter 1 Software Requirements
3
Note – You cannot boot a domain mounted with the SPARC64 VII processors using
the Oracle Solaris 10 8/07 installation DVD. Use the Oracle Solaris 10 5/08 or later
installation DVD to boot a domain mounted with the SPARC64 VII processors.
For additional Oracle Solaris OS information, see “Oracle Solaris OS Issues and
Workarounds” on page 59.
Obtaining XCP and Oracle Solaris
Patches
For the information on how to get the latest XCP firmware or the latest patch for
Oracle Solaris OS, contact your sales representative.
Web Browser
Many web browsers support the XSCF Web. The browsers in TABLE 1-2 have
demonstrated compatibility with the XSCF Web through testing. For other
information about XSCF Web, see “Notes on XSCF Web” on page 11.
TABLE 1-2
Tested Web Browser Versions
Web Browser Application
Version
Microsoft Internet Explorer
Mozilla Firefox (Windows)
6.0/7.0/8.0/9.0/10.0/11.0
3.6.25/8.0/9.0/ESR10.0/ESR17.0/ESR24.2/
ESR38.4/ESR45/ESR52/ESR60
Mozilla Firefox (Oracle Solaris 10)
Mozilla Firefox (Oracle Solaris 11)
2.0/3.5.9/3.6.22
6.0
4
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Existing XCP Firmware Versions and
Support Information
The following lists the XCP firmware versions that have previously been released for
the SPARC Enterprise series, as well as the main support and update information for
each version. For the support information for XCP 1113 or earlier, see the "What's
New Information" of each version's Product Notes.
TABLE 1-3
Previous XCP Version Releases and Support Information
XCP Version
XCP 1124
XCP 1123
Main Support Information
Security fixes.
• Security fixes.
• Improvement of XCP firmware defects.
XCP 1122
XCP 1121
XCP 1120
Support of SHA256 Web server certificates and TLS 1.2.
Improvement of XCP firmware defects.
• Improvement of XCP firmware defects.
• Security fixes.
XCP 1119
XCP 1118
XCP 1117
Security fixes.
Improvement of XCP firmware defects.
Enhancement of the investigation information at the time of the
occurrence of "XSCF process down detected"
XCP 1116
• Support of the -boption of the sethttps(8) command.
• Enhancement of the diagnostic function for the "FMEM checksum
error" event.
XCP 1115
XCP 1114
• Support of the forcerebootxscf(8) command
• Enhancement of file check and other XSCF diagnostic information
• Enhancement of the investigation information at the time of occurrence
of an XSCF watchdog timeout *
*
If this fix is not applied, the cause of the occurrence of an XSCF watchdog timeout may not be identifiable.
Chapter 1 Software Requirements
5
6
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
CHAPTER
2
XCP 1124 and XCP 1124-related
This section describes the information about XCP 1124 and XCP 1124-related
information.
■
■
■
■
“What’s New in XCP 1124” on page 7
“Notes and Limitations” on page 8
“XCP 1124 Issues and Workarounds” on page 15
“Updating to XCP 1124” on page 16
What’s New in XCP 1124
This release offers the following improvement:
■
Security fixes. (RTIF1-180702-001)
7
Notes and Limitations
This section describes the notes and limitations at the time of this release.
Limitations for SPARC64 VII+ and SPARC64 VII
Processors
You must complete the upgrades to the XCP and to Oracle Solaris OS before
installing the CPU/memory board unit which is mounted with the SPARC 64 VII+
and SPARC64 VII processors on your server.
Notes on Secure Socket Layer (SSL) 3.0
Due to security reasons, SSL 3.0 cannot be used with XCP 1119 or later. Use
Transport Layer Security (TLS) 1.0.
Use Transport Layer Security (TLS) 1.0 for XCP 1121 or earlier.
Use Transport Layer Security (TLS) 1.2, 1.1, or 1.0 for XCP 1122 or later.
Notes on SSH
Due to security reasons, XCP 1123 and later restrict the use of some algorithms at
SSH connection. If the SSH connection to the XSCF fails, consider updating the SSH
client and changing the settings.
Notes on Active Directory
■
While Active Directory is enabled, when you attempt to login to XSCF via the
telnet, you might fail to login due to timeout of the query to secondary alternated
server or later.
■
If the specified timeout is too brief for the configuration, the login process or
retrieval of user privilege settings could fail. In such case, specify larger value for
the timeout and then execute again.
8
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Notes on LDAP/SSL
If the specified timeout is too brief for the configuration, the login process or
retrieval of user privilege settings could fail. In such case, specify larger value for the
timeout and try again.
Notes on Power Consumption Monitoring
Function (M3000 Server)
■
The amount of power consumption might not be indicated correctly in the MIB
information, in the showenvironment power command output, and on the XSCF
Web in the following cases; and you should wait for one minute and check the
value again.
■
During the server powering on or powering off, or for a while after the power-
on or power-off complete
■
During the active replacement of power supply unit, or for a while after the
active replacement complete
■
The figures on the power consumption monitoring function represent the
dissipation power from the server. The figures of peripheral devices are not
included.
Notes on Airflow Indicator
■
The amount of exhaust air might not be indicated correctly in the MIB
information, in the showenvironment aircommand output, and on the XSCF
Web in the following cases; and you should wait for one minute and check the
value again.
■
During the server powering on or powering off, or for a while after the power-
on or power-off complete
■
During the active replacement of power supply unit, or for a while after the
active replacement complete
■
The figures on the airflow indicator represent the volume of air exhausted from
the server. The figures of peripheral devices are not included.
Notes on NTP Server
■
We recommend the domain to use the XSCF Unit as NTP server. In this case, pay
attention to the following points:
■
XSCF must be connected to an external NTP server.
Chapter 2 XCP 1124 and XCP 1124-related Information
9
■
When you connect one or more NTP servers in addition to XSCF, connect the
same NTP server as XSCF is using.
For details on NTP server, contact a service engineer. For details on NTP settings,
refer to the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
User’s Guide.
Notes on the NTP Server Referring to the Local Clock
When the NTP server which XSCF refers to is referring to the server's own system
time (local clock), and when the address of "127.127.1.0" is set to that local clock,
time synchronization in XSCF might fail.
The address of the XSCF's own local clock is fixed to "127.127.1.0." On the other
hand, when the address of the local clock of the NTP server which XSCF refers to is
set to "127.127.1.0," the address of the clock source (refid) has the same value as the
address of the XSCF's own local clock. An NTP server like this is excluded from the
target of XSCF time synchronization.
You can execute the showntp -lcommand to refer to the address of the NTP
server's own clock source which is set in XSCF and the address of the XSCF's own
local clock.
XSCF> showntp -l
remote refid st t when poll reach delay offset jitter
==============================================================================
192.168.1.2 LOCAL(0) 3 u 10 1024 377 0.000 0.000 0.000
*127.127.1.0 .LOCL. 5 l 28 64 377 0.000 0.000 0.008
Of the two NTP server outputs, the upper (192.168.1.2) indicates the NTP server
which is set by using the setntp(8) command. The refid is LOCAL(0), which means
that the local clock which has the address of "127.127.1.0" is set to the clock source of
this NTP server. On the other hand, the lower indicates the XSCF's own local clock.
The address of the XSCF's own local clock is fixed to "127.127.1.0." Due to this, the
NTP server (192.168.1.2) is excluded from the target of XSCF time synchronization;
which results in the XSCF synchronizes with its own local clock.
With any of the following measures to avoid the trouble, time can be correctly
synchronized with the NTP server which is set by using the setntp(8) command.
■
Change the clock source that the NTP server being set in XSCF refers to
Use the showntp -lcommand and check the clock source of the NTP server
which is set in XSCF. An NTP server which indicates the refid of LOCAL(0)in the
output is referring to the local clock which has the address of "127.127.1.0," and
you should change it to refer to another clock source.
When you change the clock source of an NTP server, make sure in advance that it
has no impact on other NTP clients.
10
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
■
Change the address of the local clock of the NTP server
Of the NTP server which XSCF refers to, change the address of the local clock to
"127.127.1.1," "127.127.1.2," or "127.127.1.3." Change /etc/inet/ntp.confof
Oracle Solaris OS. To enable the change, restart of the NTP daemon is required.
When you change the address of the local clock of an NTP server, make sure in
advance that it has no impact on other NTP clients.
■
Change the stratum value of the NTP server
Of the NTP server which XSCF refers to, change the stratum value to "1." An NTP
server which has the stratum value of "1" becomes the most significant clock
source and has no refid. Therefore, there is no chance that it will have the same
address as the XSCF's own local clock.
When you change the stratum value of an NTP server, make sure in advance that
it has no impact on other NTP clients.
■
Change the address of the XSCF's own local clock
By using the setntp -m localaddr=value command, change the address of the
XSCF's own local clock. In value, specify the least significant byte of the clock
address of the local clock 127.127.1.x for value. A numeric from 0 to 3 can be
specified. By specifying either from 1 to 3, the address of an NTP server which is
referring to the local clock does not correspond to the address of the XSCF
internal local clock anymore, and a server which is referring to the local clock can
also be set as the NTP server of XSCF.
Notes on Timezones
The time zones (regions/location names) supported by the XSCF can be changed to
support the latest time zone information.
For any previously set time zone that the system can no longer use, the XSCF
switches from this unusable time zone to Coordinated Universal Time (UTC) and
operates in UTC.
When the set time zone operates in UTC, execute the settimezone-csettz-a
command, and check the settable time zones. If the time zone list does not have a set
time zone, set a time zone again.
Notes on XSCF Web
■
When you use Mozilla Firefox (Windows version) 45ESR as the Web browser with
XCP 1121 or earlier, the XSCF Web menu may not appear.
Chapter 2 XCP 1124 and XCP 1124-related Information
11
■
Using the XSCF Web function, when you change from a SHA1 Web server
certificate to a SHA256 Web server certificate, re-create the certificate with the
sethttps command.
■
■
Under the Windows 7 with Internet Explorer 8.0, 9.0, or 10.0 environment, XSCF
Web does not support the use through the built-in Administrator account.
On Internet Explorer 8.0, when you move the on-screen horizontal frame up or
down to change the frame height and use the [Monitor Msg Show/Hide] button,
the screen background color may turn black. In this case, click Refresh on the
View menu or press [F5] key to refresh the display, to go back to the screen right
after the login.
When you moved the on-screen horizontal frame up or down to change the frame
height, do not use the [Monitor Msg Show/Hide] button.
■
■
M4000/M5000/M8000/M9000 servers:
XSCF Web does not support the External I/O Expansion Unit Manager feature.
The M3000 server does not support External I/O Expansion Unit.
M3000/M4000/M5000 servers:
Using the XSCF Web, when you import XCP or update the firmware, Session ID
error may be displayed on the web browser. And in the Autologout setting, when
you specify the timeout period as over 30 minutes, Internal Server Error might be
displayed when you perform the firmware update. Please close the current
browser and open the new browser to reconnect to XSCF Web.
M8000/M9000 servers:
Using the XSCF Web, when you import XCP or update the firmware, Session ID
error may be displayed on the web browser. And Internal Server Error might be
displayed when you perform the firmware update. Please close the current
browser and open the new browser to reconnect to XSCF Web.
■
When you use the XSCF Web under Internet Explorer 6 or 7, or Firefox 2
environment, remove the plug-in if a plug-in such as the search tool installed with
the browser, or disable the pop-up blocking.
When you use the XSCF Web under Internet Explorer 8, 9, 10, or 11, or Firefox 3,
6, 8, 9, 10, or 17 environment, disable the pop-up blocking.
■
When you use the XSCF Web under Firefox 3, 6, 8, 9, 10, or 17 environment and
attempt to log in to XSCF, you may be prompted by the browser to save the login
ID/password. In this case, do not save the login ID/password. If you save the
login ID/password, the stored data can be displayed unexpectedly on the LDAP,
SMTP and REMCS web pages.
Use one of two settings described below, to disable the browser's ID/password
remember function:
■
Disable the entire browser ID/password remember function.
Click on the [Tools] menu, and select [Preferences]. On the [Security] panel, see
[Passwords] and uncheck the [Remember passwords for sites] checkbox.
12
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
■
Designate the XSCF Web address as an exception to the ID/password save
operation.
Click on the [Tools] menu, and select [Preferences]. On the [Security] panel, see
[Passwords] and check the [Remember passwords for sites] checkbox. Then, on
the dialog for the ID/password saving which appears in the login attempt to
XSCF, click the [Never for This Site] button. As a result, the XSCF Web address
will be registered in the [Exceptions] list to the ID/password save operation,
and the dialog to prompt the ID/password saving will no longer be displayed
in the subsequent login attempts to XSCF.
■
When using FireFox ESR 38.4 as the Web browser with XCP 1120 or earlier, the
display of Domain ID and Status in [Menu] - [Domain Operation] - [Domain
Power] may appear distorted. If the display is distorted, Power ON and other
operations are not available from the above menu. If this problem occurs, use
the XSCF shell.
■
The browser layout may appear distorted in Internet Explorer 9, 10, or 11
environment. If this occurs, enable Compatibility View in Internet Explorer.
Limitations on Audit
■
The log file transfer function of the audit log set with the setauditarchive
command and the setauditdisablecommand is not supported at present.
■
If the policy for writing to an audit record is set to stop temporarily ("suspend"),
degradation due to an error may occur or the XSCF may be reset. In the
policy for the audit trail. For details of this issue, see "RTIF1-141209-001."
Miscellaneous Notes and Limitations
Note – M4000/M5000/M8000/M9000 servers:
For dynamic reconfiguration (DR) and hot-plug issues, see “Oracle Solaris OS Issues
and Workarounds” on page 59.
■
■
The log archive feature by the setarchiving(8) and the showarchiving(8)
commands, and the [Log Archives] menu of XSCF Web is not supported.
M4000/M5000/M8000/M9000 servers:
The following functions displaying the power consumption is not supported on
M4000/M5000/M8000/M9000 servers.
■
poweroperand of the showenvironment(8) command
■
XSCF Web
Chapter 2 XCP 1124 and XCP 1124-related Information
13
Note – Invalid values will be displayed by executing the commands or menu.
SNMP Agent Function obtains invalid values of power consumption.
■
M4000/M5000 servers:
The M4000/M5000 servers are cold service machines. Hot-swapping of the CPU
module (CPUM), memory board (MEMB), I/O unit (IOU), or XSCF unit is not
supported.
■
■
M8000/M9000 servers:
Do not use the internal CD-RW/DVD-RW drive unit and the TAPE drive unit at
the same time.
When you use the external power control interface of the external power
controller, the following notification signals are not supported:
■
The OS panic or the server hardware error signal (*CPUN/RTNU)
■
The server hardware error signal (power fail, temperature error, and fan error)
(*ALARM)
■
M8000/M9000 servers:
On the M8000/M9000 servers with XCP 1050 or later, the dual XSCF Unit feature
is working. Therefore, you can not downgrade M8000/M9000 servers with XCP
1050 or later to XCP 1040 or XCP 1041, which does not support dual XSCF Unit
feature.
■
■
You cannot use the following user account names, as they are reserved for system
use: root, bin, daemon, adm, operator, nobody, sshd, rpc, rpcuser, ldap,
apache, ntp, admin, proxyuser, and default.
XSCF-LAN is compliant with auto-negotiation. When you connect XSCF-LAN
and the network device which has been fixed to the full-duplex mode, according
to the IEEE 802.3 rule, XSCF-LAN communicates in the half-duplex mode. Due to
this, network communication speed may slow down or communication error may
occur. Do not fail to set the network device which connects with XSCF-LAN to the
auto-negotiation mode.
■
M4000/M5000/M8000/M9000 servers:
While you are performing DR operation to a COD board, do not execute the
addcodactivation(8)/addcodlicense(8)/deletecodactivation(8)/
deletecodlicense(8)/setcod(8) command.
■
■
■
At this time the -e, -l, -Poptions of the snapshot(8) command are not
supported.
An XSCF user account user name cannot match an LDAP user name, and an
XSCF user account number (UID) cannot match an LDAP UID number.
In case that any failure occurred on a system board and failed to read the FRU
information of system board, XSCF regards the system board as a COD board.
The showboards-vacommand displays the COD status as "y." In this case,
replace the failed system board.
14
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
■
An XSCF internal database failure may be detected following the occurrence of an
XSCF storage failure, etc. If so, the same database failure may be detected again in
the XSCF, or the XSCF may be degraded. The following describes the operation
for this problem in each model and how to restore the XSCF.
■
M3000/M4000/M5000 server:
[XCP 1091 or earlier]
If an XSCF internal database failure is detected, the XSCF restarts and
continues the operation. Consequently, the same database failure may be
detected again. If so, the domain may stop or reboot.
<How to restore>
After shutting down all domains, replace the motherboard unit (MBU) or the
XSCF unit (XSCFU).
[XCP 1092 or later]
If an XSCF internal database failure is detected, the XSCF degrades itself so
that the same database failure is not detected again. In this case, the domain
neither stops nor reboots but instead continues the operation.
<How to restore>
After shutting down all domains, replace the MBU or XSCFU.
■
M8000/M9000 server:
[XCP 1091 or earlier]
If an XSCF internal database failure is detected, the XSCF restarts and
continues the operation. Consequently, the same database failure may be
detected again. If so, the domain neither stops nor reboots but instead
continues the operation.
<How to restore>
Execute the replacefru command to replace the XSCFU.
[XCP 1092 or later]
If an XSCF internal database failure is detected, the XSCF degrades itself so
that the same database failure is not detected again. In this case, the domain
neither stops nor reboots but instead continues the operation.
<How to restore>
XCP 1124 Issues and Workarounds
For known XCP issues and possible workarounds in XCP 1124, see “Known Issues
and Workarounds in XCP 1124” on page 21. For fixed XCP issues in XCP 1124, see
“XCP Issues Fixed in XCP 1123” on page 25.
Chapter 2 XCP 1124 and XCP 1124-related Information
15
Updating to XCP 1124
To update your XCP to XCP 1124, please pay attention to the following points
depending on your current XCP version.
You can upgrade to XCP 1124 from XCP version 1050 or later. Refer to the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide for
instructions.
Importing the XCP File
Once the XCP file is imported, update the firmware immediately. If the XCP file is
merely imported, the versions of the XSCF firmware and OpenBoot PROM firmware
running on the system remain the old ones. If the CPU/memory board unit or
motherboard unit is replaced or the XSCF unit is replaced/switched in this status,
the firmware of the unit cannot be matched with the original firmware version, and
attempts to replace/switch the unit or start the domain will fail.
Updating from a Version Earlier Than XCP 1050
(M4000/M5000/M8000/M9000 Servers)
■
Updating directly to XCP 1071 or a later version is not possible.
First, update your XCP to any version between XCP 1050 and XCP 1070, and then
update to XCP 1071 or later. For the procedure to update to any version between
XCP 1050 and XCP 1070, see the Product Notes of the respective version.
■
Delete any accounts named "admin".
Any accounts named admin must be deleted prior to updating to XCP 1050 or
later. This account name is reserved in XCP 1050 and later. Use the
deleteuser(8) command to delete the account.
Updating from a Version Earlier Than XCP 1070
(M8000/M9000 Servers)
■
After a firmware update from XCP 1061 or earlier to XCP 1070 or XCP 1071, both
the active and standby XSCFs enter the Halt state. To work around this problem,
be sure to reset the XSCFs after the firmware update, by using the rebootxscf(8)
command.
16
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
■
On a domain which has been in operation during the XCP update to 1124, when
you perform Dynamic Reconfiguration (DR) to add or replace the SPARC64 VII
processors, you need to update the OpenBoot PROM firmware. The OpenBoot
PROM firmware is updated as you update the XCP and restart the domain. For
this reason, it is strongly recommended to restart all the domains after you update
the firmware to XCP 1124, regardless of whether you added or replaced the
SPARC64 VII processors.
Chapter 2 XCP 1124 and XCP 1124-related Information
17
18
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
CHAPTER
3
This section describes the special instructions and the issues about the
M3000/M4000/M5000/M8000/M9000 servers software.
■
■
■
■
■
“Notes and Limitations” on page 19
“XCP Issues and Workarounds” on page 20
“Oracle Solaris OS Issues and Workarounds” on page 59
“Software Documentation Updates” on page 87
“Miscellaneous Information” on page 98
Notes and Limitations
Notes on RCI Function
The RCI function is supported on any of the following servers.
■
The server which you purchased in Japan from Fujitsu
■
The server which is bearing the Fujitsu logo on the front surface, regardless of
area of purchase1
When using the RCI function such as RCI build procedure, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers RCI Build Procedure and the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI User's Guide.
1. This refers to the server which has the black-colored chassis and is exclusively bearing the Fujitsu logo on the
front surface.
19
Notes on Remote Maintenance
This section is explained the notes on using the remote maintenance service. For
details on making settings and using the remote maintenance service, see the
manuals about the remote maintenance service in the Enhanced Support Facility User’s
Guide for REMCS.
Before Setting the Remote Maintenance Service
Before using the remote maintenance service on your
M3000/M4000/M50900/M8000/M9000 server, you have to complete the settings for
the REMCS Agent function on XSCF Web.
Be sure to complete the necessary settings for using XSCF Web such as enabling the
HTTPS, on XSCF shell. After completing, specify the settings for REMCS Agent
function on XSCF Web.
For detail on the settings for XSCF Web, see SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
The Timezone Setting for REMCS Agent Function
The REMCS agent uses the current server timezone. So when you change the server
timezone by XSCF, the "Periodical Connection Schedule" setting is required again for
XCP Issues and Workarounds
For known XCP issues and possible workarounds in XCP 1124, see "“Known Issues
and Workarounds in XCP 1124” on page 21." For XCP issues that have been fixed,
see "XCP Issues Fixed" for each version.
20
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Known Issues and Workarounds in XCP 1124
TABLE 3-1 lists known XCP issues and possible workarounds in XCP 1124.
TABLE 3-1
ID
Known Issues and Workarounds in XCP 1124
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070914-025
O
O
When you execute XCP Sync on the No workaround is available.
Firmware Update page, after 15 However, the XCP Sync process has
minutes, the error message "Another been continuously executed. Check
flashupdate is now the XSCF update completion message
processing" or "The page cannot on the monitoring message to confirm
be displayed" may appear.
the completion of Sync process.
RTIF1-
071116-001
DR operations might fail (with a
misleading message regarding the
Do not fail to execute the diagnosis in
the maintenance menu of the
board being unavailable for DR) after addfru(8) or replacefru(8)
the addfru(8) or replacefru(8) command. In case of missing the
command have been used for active diagnostic test in the maintenance
replacement. This happens when the menu of the addfru(8) or the
active replacement is done without replacefru(8) command, execute
the diagnostic test in the maintenance the testsb(8) command or delete the
menu.
CPU/memory board unit using the
deletefru(8) command and then
retry the addfru(8) command.
RTIF1-
O
When the server is being installed,
Turn off the system AC power, then
071126-002
and the mainline switch is turned on turn it on again.
for the first time, these error messages
might be displayed:
When the error log is displayed again,
turn off the system AC power, then
turn it on again. To turn on the system
power that you turned off, wait at
least 30 seconds before power-on.
PSU shortage
XSCF watchdog timeout
XSCFU hang-up is detected
XSCF process down detected
RTIF1-
080725-001
O
O
O
O
O
O
In setsnmpaddv3traphost, when No workaround is available.
the authentication failed due to the
reasons such as the trap host not
working, or the wrong user name or
password, the subsequent SNMP
traps will not be notified.
Confirm that the SNMP trap host is
working and re-execute the
setsnmp(8) command using the
correct user name.
RTIF1-
080725-002
When the SNMPv3 trap has been set, Reset the XSCF.
after the watchdog timeout occurred
in XSCF and XSCF reset, the
subsequent SNMP traps will not be
notified.
Chapter 3 Information About Software
21
TABLE 3-1
ID
Known Issues and Workarounds in XCP 1124 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
081016-002
O
When FRUs are cold-replaced (with No workaround is available.
the system off), (after poweron) the
monitor process might not generate
entries in the monitor message log
identifying the replacement operation
that has occurred.
RTIF1-
081225-001
O
O
O
In the settimezone -c adddst
command, when you set eight or
more letters to the abbreviation of
time zone and the name of Daylight
Saving Time, execution of the
showlogscommand induces a
segmentation fault and results in an
error.
Specify the abbreviation of time zone
and the name of Daylight Saving Time
in seven letters or less.
RTIF1-
090108-003
O
O
O
O
O
O
Date and time of the Last Update
which displayed on the XSCF Web
may not be updated.
Please refresh the display.
RTIF1-
090427-005
When you use the
No workaround is available.
Use XSCF Web.
setpasswordpolicy(8) command
and set "9999999999" (10 digits) in
Different Characters, the
showpasswordpolicy(8) command
displays "-1."
RTIF1-
090731-001
O
O
While a DIMM correctable error (CE) This error log can be safely ignored.
has been occurred, when XSCF
failover is started, a message "XSCFU
hang-up is detected"might
appear in the error log.
RTIF1-
100331-002
O
O
O
O
When you start up a domain from a No workaround is available.
disk that is connected to the 8 Gbps
fibre channel card (SE0X7F21F,
SE0X7F22F) and if it is unable to start
the domain due to causes such as disk
failure, an error log of "FRU:
/UNSPECIFIED" might be registered.
RTIF1-
O
If setpasswordpolicy-ris set to 0, Do not execute the command
110222-001
existing XSCF users are unable to
change their password and get the
error message, password:
"setpasswordpolicy-r0".
Operationfailed.
22
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-1
ID
Known Issues and Workarounds in XCP 1124 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
O
After regenerating XSCF host public XSCF reboot is required after
120329-001
key with setssh-cgenhostkey,
other connected XSCF users may
encounter a “permissiondenied”
error when executing the XSCF
command line.
generating the XSCF host key.
RTIF1-
120329-003
O
O
The fault information that is held in If a new failure occurs after replacing
the domain, and that can be checked the XSCF unit or MBU, check the
with the Oracle Solaris OS fmadm
failure occurrence time and UUID in
faultycommand is sent to the XSCF the fmdump -Voutput to judge
every time the domain is restarted or whether the failure has been resolved
the XSCF is reset. On the other hand, in the past or is a new failure.
the XSCF holds a history of the fault
information it has received in the past
from the domain, and exercises
control to ensure that the same failure
is not registered in the error log.
Also, if any fault information on
failures already resolved remains in
the domain, execute the Oracle Solaris
OS fmadm repaircommand during
maintenance to reset the fault
However, replacing the XSCF unit on information.
the M4000/M5000 server or replacing
the motherboard unit (MBU) on the
M3000 server restores the failure
history information held by the XSCF
to its initial values. As a result,
failures resolved in the past may be
registered again as new failures in the
error log.
Chapter 3 Information About Software
23
TABLE 3-1
ID
Known Issues and Workarounds in XCP 1124 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
141209-001
O
O
O
Suppose that the audit system is
configured so that writing to audit
Execute the setaudit-pcount
command to configure the audit
records stops temporarily when the system to delete new audit records
audit trail reaches its full capacity. An when the audit trail has reached its
XSCF shell or XSCF Web operation at full capacity. (Default).
this time may lead to an XSCF reset
due to the mistaken detection of one
of the following errors: "Hardware
accesserror." or "SCFpanic
detected."
[How to restore]
Perform the following procedure.
1. Perform login authentication using
the default user account.
2. Execute the restoredefaults
xscfcommand.
This problem occurs even when the
XSCF shell is automatically run
regularly to monitor the logical
domain status by using
3. Execute the setaudit-pcount
command to configure the audit
system to delete new audit records
when the audit trail has reached its
full capacity. (Default)
PRIMECLUSTER software.
You can check the contents of the
currently set audit system by using
the showaudit(8) command. In the
following example, "Policy on
full trail" is "suspend,"
indicating that writing to the audit
record has been set to stop
temporarily. Also, "Audit space
free" is "0," indicating that the audit
trail has reached its full capacity.
[Example]
XSCF> showaudit all
Auditing: enabled
Audit space used: 4194304
(bytes)
Audit space free: 0 (bytes)
Records dropped: 0
Policy on full trail:
suspend
omitted
24
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
XCP Issues Fixed in XCP 1124
TABLE 3-2 lists XCP issues fixed in XCP 1124.
TABLE 3-2
ID
XCP Issues Fixed in XCP 1124
M4000 M8000
Workaround
RTIF1-
180702-001
O
O
O
Security fixes.
Update the XCP firmware to XCP 1124
or newer.
XCP Issues Fixed in XCP 1123
TABLE 3-3 lists XCP issues fixed in XCP 1123.
TABLE 3-3
ID
XCP Issues Fixed in XCP 1123
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
170605-001
O
O
O
When the XSCF is specified as an NTP Set an NTP server other than the
server that Oracle Solaris refers to, no XSCF of the current system as an NTP
polling response can be obtained from server that Oracle Solaris refers to.
the XSCF, and synchronization with
the NTP server might fail.
RTIF1-
170609-001
O
O
O
Security fixes.
Update the XCP firmware to XCP 1123
or newer.
(CVE-2016-6304, CVE-2016-2182,
CVE-2016-2183, CVE-2016-6302, CVE-
2016-6306, CVE-2016-6515, CVE-2017-
3731, CVE-2016-7431, CVE-2016-7429,
CVE-2016-7433)
For detail, see the Oracle website
related to the following:
Critical Patch Updates
Chapter 3 Information About Software
25
XCP Issues Fixed in XCP 1122
TABLE 3-4 lists XCP issues fixed in XCP 1122.
TABLE 3-4
ID
XCP Issues Fixed in XCP 1122
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
161006-002
O
O
O
The reporting of an error or the
remote reporting of a hardware
setting change notification may not
reach the REMCS center.
No workaround is available.
XCP Issues Fixed in XCP 1121
TABLE 3-5 lists XCP issues fixed in XCP 1121.
TABLE 3-5
ID
XCP Issues Fixed in XCP 1121
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIS1-
151212-001
O
O
O
If the SNMP Agent Status setting was After changing the SNMP Agent
changed when in the enable state, Status setting once to disable with the
SNMP Agent Status becomes disable. setsnmp command, set Agent Status
to enable with the setsnmp command.
You can check Agent Status from the
Agent Status display of the
showsnmp command.
RTIF1-
161006-004
O
O
O
Security fixes.
Update the XCP firmware to XCP 1121
or newer.
(CVE-2015-3183, CVE-2015-2808)
For detail, see the Oracle website
related to the following:
Critical Patch Updates
26
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
XCP Issues Fixed in Releases Earlier Than XCP
1121
TABLE 3-6 lists XCP issues that have been fixed in releases earlier than XCP 1121.
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070418-004
O
O
O
O
All domains must be powered off
before upgrading the XCP.
Power off domains before using the
flashupdate(8) command to
upgrade XCP firmware.
RTIF1-
If you log in to the XSCF while it is Log out of the bash$prompt and
070418-005
still booting, you may get a bash$
prompt instead of the XSCF>prompt,
and be unable to perform most
operations.
wait for the XSCF to finish booting.
RTIF1-
O
O
O
While XSCF is running, a process may M8000/M9000 servers:
070418-009
go down, a watchdog timeout may
Check that XSCF is started. If there is
occur, or a hang-up may occur. After XSCF unit which is not started,
this, XSCF may reset.
execute the replacefru(8)
command, and then uninstall the
XSCF unit and install it again. Or
stop all the domains and then execute
the system power off/on (AC
OFF/ON). To turn on the system
power that you turned off, wait at
least 30 seconds before power-on.
M4000/M5000 servers:
Check that XSCF is started. If not
started, stop all the domains and then
execute the system power off/on (AC
OFF/ON). To turn on the system
power that you turned off, wait at
least 30 seconds before power-on.
M3000 server:
Check that XSCF is started. If not
started, stop the domain and then
execute the system power off/on (AC
OFF/ON). To turn on the system
power that you turned off, wait at
least 30 seconds before power-on.
Chapter 3 Information About Software
27
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070418-010
O
O
The showdomainstatus-a
command shows domain status as
Powered Off, but the showboards-a
command shows the domain is
testing.
Use the showboardscommand to
check the status of domain power.
The showdomainstatuscommand
takes a longer time to show the
correct status.
RTIF1-
070418-011
O
O
O
O
O
O
Some commands that update
configuration data take a relatively
long time to execute.
Do not cancel set*commands. They
appear to hang, but eventually
complete in about 30 seconds.
RTIF1-
070418-012
The fault (memory.block.ue) is
encountered and reported
periodically.
An uncorrectable error exists in a
DIMM and the DIMM should be
replaced.
RTIF1-
070418-020
When using the XSCF Web to import a Use the command getflashimage
firmware image, if the image is -dto delete the corrupted image. If
corrupted (for example, if the browser necessary, reboot the XSCF Unit, then
window is closed during import), the use the flashupdatecommand
flashupdatecommand might later again to clear the internal error.
report an internal error.
CR ID 6537996 is similar.
RTIF1-
070418-021
O
O
O
O
Attempting to move a COD board
using the moveboard(8) command addboard(8) commands instead of
might fail.
Use the deleteboard(8) and
the moveboard(8) command.
RTIF1-
The XSCF firmware monitors itself
Allow the XSCF Unit to finish
070418-022
and if it detects any abnormalities, it resetting. It will return to normal
will force a reset.
operation within approximately 5
minutes.
RTIF1-
070418-023
O
O
O
O
Using the rebootxscf(8) command Ignore this event.
may result in a process down error,
and possibly an FMA event with MSG
ID SCF-8005-NE.
RTIF1-
showaudit all shows a long list of Update the database with the
070418-025
defaults in the policy section after the following:
database is cleared.
setaudit -a opl=enable
setaudit -a opl=default
RTIF1-
O
O
When you have updated the
No workaround is available.
070528-001
/etc/ttydefsfile of Oracle Solaris
OS to disable the console flow control,
you cannot disable the console flow
control on the telnet via a server.
28
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070528-002
O
O
O
While XSCF is running, watchdog
timeout may occur and XSCF may
reboot.
M4000/M5000/M8000/M9000
servers:
Check that XSCF is started. If not
started, stop all the domains and then
execute the system power off/on (AC
OFF/ON). To turn on the system
power that you turned off, wait at
least 30 seconds before power-on.
M3000 server:
Check that XSCF is started. If not
started, stop the domain and then
execute the system power off/on (AC
OFF/ON). To turn on the system
power that you turned off, wait at
least 30 seconds before power-on.
RTIF1-
070802-001
O
O
O
O
When connected to telnet, the login or Disconnect the telnet session and try
the password prompt doesn't appear. again.
RTIF1-
Using the XSCF Web, when you
To specify the value which exceeds
070823-001
selected [SSH] on the snapshot screen, the maximum number of character
the maximum number of character
input for Host, Directory, ID, and
Password doesn't correspond to the
maximum number of character input
on the XSCF Shell.
input for the XSCF Web, use XSCF
Shell.
RTIF1-
070823-002
O
O
O
O
When you display the Logical tree on On the Menu, select System Board
the XSCF Web, there may be multiple Configuration and refer to the domain
displays of the same domain,
depending on the domain
configuration.
configuration from XSB Status (All).
Or use the showboards(8) command
to refer to the domain configuration.
RTIF1-
When you display the Logical tree on On the Menu, select Device Status to
070823-003
the XSCF Web, the hardware
configuration of CPU or memory
refer to the domain hardware
configuration. Or use the
which assigned to the domain appears showdevices(8) command to refer to
differently from the actual domain
configuration.
the domain hardware configuration.
Chapter 3 Information About Software
29
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
O
M8000/M9000 servers:
M8000/M9000 servers:
070824-001
When remote power control mode of After XSCF Unit is replaced,
interlocking mechanism for power
supply to domain, is enabled, after
XSCF Unit is replaced, the
configures the RCI again and sets the
remote power control mode.
interlocking for power supply by RCI
cannot work.
M3000/M4000/M5000 servers:
After the motherboard unit is
replaced, configures the RCI again
and sets the remote power control
mode.
M3000/M4000/M5000 servers:
When remote power control mode of
interlocking mechanism for power
supply to domain, is enabled, after the
motherboard unit is replaced, the
interlocking for power supply by RCI
cannot work.
RTIF1-
070824-002
O
O
O
O
On the XSCF Web, when you select
Domain Mode Configuration to
perform various settings, the pop-up settings. Or once terminate the XSCF
screen may not appear but
"Undefined" may be displayed on
the screen.
Select Domain Mode Configuration
one more time and perform the
Web and then perform the settings.
RTIF1-
070824-003
On the XSCF Web, while the XCP
import is in process from the
[Firmware Update] screen, when you
execute the [REFRESH] button, the
pop-up of "Uploading Now!"
disappears and the XCP import
terminates abnormally.
None available at this time.
RTIF1-
070824-004
O
O
O
O
On the XSCF Web, on the [Domain
Status] screen, when you select an
XSB displayed on the Domain
Component List, and in case the
selected XSB is not yet mounted or is
Uni-XSB, the pop-up screen displays
no data.
None available at this time.
RTIF1-
On the XSCF Web, when you changed Ignore the pop-up and close the
070824-005
the Refresh Interval value of the
Monitor Message Frame, the invalid
pop-up "Confirmation Domain
mode" may appear.
screen.
30
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
On the tab browser, to the same host, When you use the tab browser, do not
070824-006
when you perform multiple log-in
with the user accounts of different
user privileges, the user privilege of
the last log-in user account will be
applied to those pages which you've
already logged in.
perform multiple log-in to the same
host.
RTIF1-
070824-007
O
O
O
When you change configuration on
From the Menu, select the original
the XSCF Web, if you select cancel on configuration page again.
the confirmation dialog or perform
forced close, the original
configuration page may be
unavailable.
RTIF1-
070824-008
O
O
On the [Domain Configuration]
screen, when you select an undefined
Domain ID, the Configuration Policy
remains as the content which
previously displayed.
None available at this time.
RTIF1-
070824-009
On the [System Board Configuration] None available at this time.
screen, when you execute Detail, it is
displayed as if the CPUM and the
memory of MBU are all mounted on
the PSB#0 side. And in the detail
display of PSB#0, memory is
displayed as "22 + 2."
RTIF1-
070824-011
O
O
While using Firefox 2, in the
From the Menu, select the [Domain
Configuration] page again.
Configuration policy setting on the
[Domain Configuration] screen, when
you specify a domain which is in
operation, an error display pop-up
appears. When you click on the [Back]
button on this error display pop-up
and click on the [Cancel] button on
the inquiry screen to re-display the
data, the system remains in the error
message screen.
RTIF1-
070904-001
O
CLIs executed in Standby XSCF
should display
Only the following CLIs can be
executed on the Standby XSCF:
snapshot(8), switchscf(8).
"Permission denied".
Do not attempt to run any other CLI
on the Standby XSCF. Such attempts
will report various errors.
Chapter 3 Information About Software
31
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070904-002
O
O
O
O
When the snapshot CLI attempts to Do not attempt to use write-protected
write to a USB stick that has write
protect set results in an I/O error.
USB devices for collecting snapshot.
RTIF1-
An incorrect domain state is reported. No workaround is available. This is
070904-003
After the command sendbreak(8) to expected behavior of the
domain is issued,
sendbreak(8) operation.
showdomainstatus(8) continues to
show the state as "Running" when the
domain is actually at "ok" prompt.
RTIF1-
O
O
Time can't be synchronized with the Check the stratum value of the NTP
070904-005
NTP server.
server.
The stratum value of XSCF is "5." The
NTP server which the XSCF refers to
must be a server on which the stratum
value is smaller than 5. When you
changed the NTP server to refer to,
reboot the XSCF.
When the stratum value has been set
correctly and the time can't be
synchronized, use the showntp(8)
command to check the jitter value to
be displayed.
If this value is large, please reboot the
XSCF.
RTIF1-
070904-006
O
O
O
O
While executing the domain power- Execute the domain reset one more
on, domain reset or DR, in case the time, or power off the domain and
XSCF reset occurred, the process may then power on again.
be aborted in some or all of the XSB.
RTIF1-
If an invalid SMTP server is
Wait for the CLI to complete. The rest
070912-001
configured, a subsequent attempt to of the system will function normally
disable e-mail service (using the
setemailreportCLI) may block for
up to 30 minutes.
during this time.
• The CLI can also be aborted by ^C.
Note that the operation (disabling
e-mail report) is completed, even if
^C is used.
• showemailreportcan be used to
confirm that the e-mail service has
been disabled.
32
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070914-001
O
O
While executing the firmware update Do not execute the import of XCP by
by using the flashupdate(8) using the getflashimage(8)
command or BUI, when the import of command or BUI, while executing the
XCP executed by using the
firmware update.
getflashimage(8) command or BUI,
the firmware update aborts.
RTIF1-
070914-002
O
O
O
O
O
O
When the XSCF switching performed, Execute the showhardconf(8)
the monitor message doesn't show the command to refer to the state of XSCF
number of XSCF unit which turned
active.
unit.
RTIF1-
070914-003
When the XSCF switching performed, No workaround is available.
there may be "SCF: Board control
error (DMA timeout)" stored in
the error log.
This message can be safely ignored.
RTIF1-
070914-005
On the XSCF Web, when using
Internet Explorer 7, on the [User
Accounts] screen, User Local
Accounts doesn't show the Status of
each user.
None available at this time.
RTIF1-
070914-006
O
O
O
O
When you set the XSCF user account Use up to 31 characters to set the
name to the maximum 32 characters, XSCF user account name.
you can log in, but then, when you
execute the XSCF Shell or operate the
XSCF Web, "Permission denied"
occurs.
RTIF1-
070914-007
On the XSCF Web, when you use
Internet Explorer and select [Log
Archives], the [Status of Log
Archiving] screen doesn't show the
status information.
None available at this time.
RTIF1-
070914-008
O
O
O
O
On the XSCF Web, when you select
[POP] or [SMTP-auth] on the [SMTP]
screen to enter Password, the input
value doesn't appear as "*".
None available at this time.
None available at this time.
RTIF1-
070914-009
On the XSCF Web, when selected
[POP] or [SMTP-auth] on the [SMTP]
screen, the setting can be done while
the input field remains blank. And
despite the setting done, the past
setting data appears.
Chapter 3 Information About Software
33
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070914-010
O
O
On the XSCF Web, on the [SNMPv3 When you set the SNMP User with 16
Security Settings] screen, when you digits or more, use the
select [Add User] or [Copy User] of showsnmpusm(8) command.
the [SNMP User-Based Security
Model (USM)], the [Create or Copy
User] screen appears as a pop-up. On
this screen, when you set the SNMP
User with 16 digits or more, it will be
registered correctly but the web
browser screen displays up to 15
digits.
RTIF1-
070914-011
O
O
O
O
O
O
On the XSCF Web, on the [SNMPv3 Do not fail to set Confirm Password
Security Settings] screen, when you
select [Change Password] of the
[SNMP User-Based Security Model
(USM)], the [Change Password] screen
appears as a pop-up. On this screen,
despite the [Confirm Password] is a
field which must be entered, it is not
displayed with the note "*
which is a field must be entered.
Indicates require field."
RTIF1-
070914-012
On the XSCF Web, on the [SNMP-
Properties] screen, when you don't
select [Trap Host] and execute the
[Remove] button, the invalid message
"The trap host will be
None available at this time.
removed" appears as a pop-up. And
when you select [OK] on the pop-up
screen, the invalid message "Remove
trap host failed. setsnmp:
invalid trap type" appears.
RTIF1-
On the XSCF Web, on the [Audit-
Set the correct user name or execute
070914-013
Properties] screen, when you execute the [Cancel] button to close the pop-
the [Add Policy] button, the [Set User up screen.
Policy] appears as a pop-up. On this
screen, when you specified a user
name which is not registered in XSCF
and an input error resulted, the value
entered won't be cleared though you
execute the [Reset] button to clear the
specified user name.
34
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070914-014
O
O
On the XSCF Web, when you use a
user account without an appropriate
privilege to select [Snapshot], set the
parameters of [Start time] or [End
time] and execute the [Download]
button, it doesn't result in a privilege
error but the parameter error will be
displayed.
None available at this time.
RTIF1-
070914-015
O
O
O
O
On the XSCF Web, when you select
[User Manager] to add the user
account, the user name to be specified adduser(8) command.
To specify a user name of 20
characters or more, use the
is limited up to 20 characters.
RTIF1-
070914-016
On the XSCF Web, when you select
[LDAP] to perform the LDAP
Since "0" is an invalid value in the
[LDAP Search Timeout], do not fail to
configuration, if you execute [Save]
while the [LDAP Search Timeout]
remains as its default value "0," it
results in "LDAP Configuration
Update Failed. Save LDAP
Configuration Data failed,"
and the setting can't be stored.
set a value other than 0.
RTIF1-
070914-017
O
O
O
O
On the XSCF Web, when you select
None available at this time. The
[LDAP] and input the [LDAP Search maximum 2147483647 seconds will be
Timeout] with a value which exceeds set to the [LDAP Search Timeout].
the maximum 2147483647 seconds, it
doesn't result in parameter error.
RTIF1-
On the XSCF Web, when you use a
The configuration is not deleted. To
070914-018
user account without an appropriate show the configuration information
privilege to select [LDAP] and execute again, select [LDAP] from the Menu.
the [Delete] button, it results in a
privilege error but the configuration
information which has been displayed
will be cleared and seems as if
deleted.
Chapter 3 Information About Software
35
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
070914-019
O
O
The CLI 'showldap -c' (which
displays current LDAP certificates)
will show proper data only when
Use the same user account for all
LDAP display or configuration
operations, for both CLI and XSCF
used with the same user account that Web.
had originally provided certificate
information using 'setldap -c'. Any
other user account will generate
"Permission denied" error.
Similarly, the XSCF Web’s [LDAP
Configuration] pop-up screen will
display no data, when a different user
account is used.
RTIF1-
070914-020
O
O
O
O
On the User Account setting page on To change the password, select [User
the [User Manager] screen, after the Manager] on the Menu again.
password change resulted in "Change
Password Error," when you click
on the [REFRESH] button, there
appears the error message "No user.
The request is aborted."
RTIF1-
070914-021
During the OpenBoot PROM process, This error log can be safely ignored.
when you power off the domain, the
error log of Domain hang-up
detected (level3)may be
registered.
RTIF1-
070914-023
O
O
O
O
When you specify the domain ID or Specify the available domain ID or
XSB number which are not supported XSB number on the server.
on the server, there appears the
parameter error message.
RTIF1-
070914-024
When you display the physical tree on None available at this time.
the XSCF Web, the warning mark is
displayed on the link card for External
I/O expansion unit.
RTIF1-
070914-026
O
O
O
O
The XSCFU cannot act as a reliable
NTP source for domains.
All domains should be configured to
use a NTP source other than the
XSCFU.
RTIF1-
070915-001
On the [Domain Configuration]
screen, when you newly define the
Domain Component List, the
When you define the Domain
Component List, use CLI.
parameter error may be displayed.
36
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
When the server is being installed,
Turn off the system AC power, then
071011-001
and the mainline switch is turned on turn it on again.
for the first time, these error messages
might be displayed:
XSCFU hang-up is detected
XSCF process down detected
DB synchronization timeout
RTIF1-
071102-001
O
O
O
The XSCF firmware monitors itself
and if it detects any inconsistencies, it the XSCF Unit to finish rebooting. It
forces an XSCF reboot.
No workaround is available. Allow
returns to normal operation within
approximately 5 minutes.
RTIF1-
071102-002
O
O
O
The snmp daemon might quit.
To restart the snmp daemon, issue the
command setsnmp enable.
RTIF1-
071116-001
DR operations might fail (with a
misleading message regarding the
Do not fail to execute the diagnosis in
the maintenance menu of the
board being unavailable for DR) after addfru(8) or replacefru(8)
the addfru(8) or replacefru(8) command. In case of missing the
command have been used for active diagnostic test in the maintenance
replacement. This happens when the menu of the addfru(8) or the
active replacement is done without replacefru(8) command, execute
the diagnostic test in the maintenance the testsb(8) command or delete the
menu.
CPU/memory board unit using the
deletefru(8) command and then
retry the addfru(8) command.
RTIF1-
071116-003
O
O
O
O
Using the XSCF Web, when you
selected [COD], codusagedetails
cannot be displayed correctly.
Use showcodusage(8) command to
display the cod usage.
RTIF1-
When Internet Explorer 7 browser is Use deletecodlicense(8)
071116-004
used, the License key deletion cannot command to delete a license key. Or
be executed on the BUI COD page.
use other browsers:
• Microsoft Internet Explorer 6.0
• Firefox 2.0 or later
• Netscape Navigator 7.1 or later
RTIF1-
071116-005
O
While system power on, when the
XSCF switching performed by
No workaround is available. This
message can be safely ignored.
switchscf(8) command, PANIC
might be generated in XSCF before it
switches, and "SHUTDOWN timeout"
may be stored in the error log.
Chapter 3 Information About Software
37
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
071116-006
O
While XSCF is running, error message Check that XSCF is started. If not
of "hang-up is detected" might started, use the rebootxscf(8)
be displayed to XSCF console, and
XSCF may reboot.
command to restart XSCF, or stop all
the domains and then execute the
system power off/on (AC OFF/ON).
To turn on the system power that you
turned off, wait at least 30 seconds
before power-on.
RTIF1-
O
O
If the following message is displayed Replace the XSCF Unit.
071129-001
on XSCF console, the XSCF might not
start:
dbs[xxx]: ERR:
scdb_init_all(): xx, No space
left on device
In addition, this error log might be
registered:
XSCF process down detected
RTIF1-
071129-002
O
O
O
O
O
O
When performing XSCF Shell
None available at this time.
commands remotely using SSH on the
remote-control device, the following
messages might be displayed:
stty: standard input: Invalid
argument
RTIF1-
071129-003
An error log cannot be registered in No workaround is available.
XSCF when the following message is
displayed on Oracle Solaris OS by the
I/O error:
SUNW-MSG-ID: SUN4-8000-4P,
TYPE: Fault
RTIF1-
071129-004
The following messages are displayed Confirm the OpenBoot PROM
and you might not boot Oracle Solaris environment variable. If the variable
OS:
has errors, set the variable again.
• Abnormal OpenBoot
environment variable Boot
process failed
• ERROR: Invalid token ''
FATAL: NVRAM contents
corrupt; Reinitializing
NVRAM parameters.
38
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
In XSCF, write of date and time may Execute the rebootxscf(8) command
071227-001
become an error. When the domain
powered on, the following message
may appear and the domain may fail
to be powered on.
to restart XSCF.
Poweron canceled due to
invalid system date and
time.
RTIF1-
071227-002
O
O
O
O
When the showhardconf(8)
Press Ctrl-C to terminate the
command executed in an environment showhardconf(8) command and
with the External I/O Expansion Unit, execute the showhardconf(8)
the showhardconf(8) command may command again.
appear as if hang up.
RTIF1-
When a non-existent device name
Specify the correct device name to the
071227-003
specified to the Boot device path and Boot device path.
resulted in the OS Boot error, the
status of I/O unit may become
"Degraded."
When the status of I/O unit became
"Degraded," replace the I/O unit.
Chapter 3 Information About Software
39
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
080313-001
O
O
When you set XSCF as the NTP server Perform either one of the following
of the domain, following message workarounds depending on the mode
may appear on the domain console at which has been set in the xntpd of the
the domain start and you may fail to domain.
perform time synchronization.
Note - In case of step mode, the time
ntpdate[xx]: no server
correction induces the time step.
suitable for synchronization When the time step occurred, it may
affect the applications running on the
domain.
• In case of step mode:
Specify the -woption to execute the
ntpdate(1M) command. Until it
succeeds in synchronizing with the
NTP server, the ntpdatecommand
retry continues at regular intervals.
In the retries, the message which
indicates that it failed to
synchronize with the NTP server
can be safely ignored. When the
synchronization completed, the
time will be corrected to the time of
the NTP server.
• In case of slew mode:
Execute the ntpdate(1M)
command without specifying the -
woption. In slew mode, the time
variance from the NTP server will
be corrected at 0.5 ms/second
intervals. The message which
indicates that it failed to
synchronize with the NTP server
can be safely ignored. Before the
time completely corrected, there
remains a time variance between
the NTP server and the domain.
40
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
M4000/M5000 servers:
M4000/M5000 servers:
080325-001
When you turn on AC power, or
Before turning on AC power, or
when you execute the rebootxscf(8) executing the rebootxscf(8)
command, you may become unable to command, please confirm that XSCF
log in to XSCF and your press on the LAN has been correctly connected
POWER switch on the operator panel and linked up.
may fail to turn on the power.
In case the situation in question
occurred, please turn off the power of
all domains and then execute the
system power off/on. To turn on the
power that you turned off, wait at
least 30 seconds before power-on.
M8000/M9000 servers:
M8000/M9000 servers:
When you turn on AC power, or
when you execute the rebootxscf(8)
or switchscf(8) command, either of
the following situations may occur.
Before turning on AC power, or
executing the rebootxscf(8) or
switchscf(8) command, please
confirm that XSCF LAN has been
correctly connected and linked up.
• Following error message gets
registered in the error log, the
status becomes Faulted, and the
standby XSCF becomes unable to
start.
Device error on serial
interconnection between
XSCFUs
In case the situation in question
occurred, please take either of the
following actions.
• Use the replacefru(8) command
to reattach the standby XSCF unit.
• Turn off the power of all domains
and then execute the system power
off/on. To turn on the power that
you turned off, wait at least 30
seconds before power-on.
• You become unable to log in to
XSCF and your press on the
POWER switch on the operator
panel can't turn on the power.
RTIF1-
080404-001
O
O
Following a dynamic reconfiguration The MSG-ID assigned to the error
operation using the XSCF commands displayed and the actual behavior
deleteboard(8) and addboard(8), you may not be consistent, and reset or
might see I/O channel degradation, degradation may not occur. Contact a
resulting in error messages and
qualified service engineer and give
entries in the corresponding ereport. the error code that displayed by using
the showlogs error command.
If you run into this problem, the
fmdump(8) command will report the
following error by mistake:
ereport.chassis.SPARCEnterprise.
asic.ioc.ch.leaf.fe
Chapter 3 Information About Software
41
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
080404-002
O
O
When you use the command
setsnmpusm passwdto set a
password, if you set a password of
fewer than eight characters, a
segmentation fault occurs.
Always set a password of at least
eight characters.
RTIF1-
O
O
When you specify "localhost" to the Do not specify "localhost" to the
080512-001
hostname of the sethostname(8)
command and reset XSCF by using
the applynetwork(8) and the
rebootxscf(8) commands, a process
goes down in XSCF.
hostname of the sethostname(8)
command.
RTIF1-
O
When the NTP setting are enabled, a This message can be safely ignored.
080512-002
message as follows may appear on the
XSCF console.
No workaround is available.
Output sample:
-xx.xxx.xxx.xxx mode 4 code
1 auth 0 clock_filter: n 8
off -0.000310 del 0.001300
dsp 0.001158 jit 0.000172,
age 0 clock_update: at
637535 assoc 2 local_clock:
assocID 54190 offset -
0.000309657 freq -9.765 state
4
...
RTIF1-
080512-003
O
O
To a domain which is in Quad-XSB
configuration, when you perform
Do not perform active addition of PCI
cards. When you failed in active
active addition of PCI cards and then addition of PCI cards, stop the
restart the domain, active addition of domain to add on the cards.
PCI cards may become unavailable
afterwards.
RTIF1-
080520-001
O
O
BUI Domain Operation page mistakes Do not perform active addition of PCI
the operation demand for Domain 8 cards. When you failed in active
or 9, and it issues it to DID#00
wrongly.
addition of PCI cards, stop the
domain to add on the cards.
RTIF1-
O
When the system is stressed with
Reboot the service processor using the
080526-001
many faults, the fmd process on the XSCF command rebootxscf(8).
service processor might hang. Once
this happens, fma commands on the
service processor can fail or hang.
42
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
080620-001
O
O
O
O
The SNMP trap not notified in the
event that the power cable of the
server got unplugged accidentally.
No workaround is available.
RTIF1-
The initial value that set with the
In the OpenBoot PROM prompt (ok),
080725-003
setdomparamset-defaultsoption execute the set-defaultscommand
differs from the initial value of
OpenBoot PROM.
to restore the OpenBoot PROM
environmental variables to the initial
value.
parameters
current value
expected value
diag-level none(0x00)
min(0x20)
auto-boot? false(0x00)
true(0xff)
RTIF1-
080725-004
O
O
O
O
O
After set the Daylight Saving Time by No workaround is available.
using the XSCF shell, XSCF Web does
not show the correct time in the Logs
menu.
Use the showlogs(8) command of the
XSCF shell.
RTIF1-
080808-001
On the cluster system using the
PRIMECLUSTER software, when
there are 10 or more RCI units, the
Refer to the manual of the
PRIMECLUSTER software to check
the cluster status. If no problem found
RCI busy status may be temporarily on the status, this message can be
unreleased and output the following safely ignored. If any problem, follow
syslog message:
the instructions on the manual to
solve the problem.
SA_pprcir.so to test host
host_name failed
RTIF1-
081006-001
O
O
O
O
O
O
The error log "XSCF FMEM write
error" recorded and the firmware
update might fail.
Power off (AC OFF) the system, and
power on (AC ON) again. Then, re-
execute the firmware update.
RTIF1-
In the setemailreport(8)
Do not specify over 255 characters in
081006-002
command, when you specified over the SMTP address.
255 characters in the SMTP address,
an error results.
RTIF1-
O
During the domain operation, after
Replace the motherboard unit only.
081006-003
the XSCF reset due to the XSCF unit When no failure found on the
failure, the hardware status shows
Deconfigurein the status of CPU
and memory.
memory, you do not need to replace
the memory.
Chapter 3 Information About Software
43
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
081006-004
O
O
O
During the firmware update, there
might be the following output
message and a XSCF panic might
occur.
Reset XSCF and use the
flashupdate(8) command to retry
the firmware update.
kernel BUG in
jffs2_do_read_inode at
fs/jffs2/readinode.c:XXX!
RTIF1-
081006-005
O
O
O
The network configuration on the
XSCF Web does not support the
function equivalent to the
Use the setnetwork -rcommand
on the XSCF shell.
setnetwork -rcommand. And
when you specified localhost or
localdomain to the host name or the
domain name, the error message
"SessionID has expired"
appears.
RTIF1-
081006-006
O
O
O
O
O
O
The panic log on the XSCF Web might When the output is insufficient,
not be displayed from the top of the execute the showlogspanic
message.
command on the XSCF shell.
RTIF1-
The password(8) command indicates No workaround is available.
081006-007
that the user operand is optional but
will fail if a user operand is not
included when other options are
specified.
Specify the user operand to execute
the password(8) command when you
specify other options.
RTIF1-
081006-011
O
O
O
SNMP trap host configuration
changes are not valid until setsnmp
disableand setsnmp enable.
Modify the SNMP setting:
XSCF> setsnmp disable
XSCF> setsnmp enable
RTIF1-
081016-001
O
O
O
O
O
O
Power failure at the commercial AC No workaround is available.
supply connector to the UPS does not
send notification/send trap.
RTIF1-
In Internet Explorer 6 or 7, clicking on Log back into the browser interface
081016-003
the [Reset] button then the [OK]
button from the [Settings]-[Audit]-
and use the [Backspace] key to clear
text in the [User] text box of the pop-
[Add Policy] pop-up screen will log up screen instead of using the [Reset]
the user out with message:
Error
button.
Session Error
Session ID has been expired
44
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
The information related to the defined Use the showdcl(8), addboard(8),
081030-001
LSB cannot be displayed on XSCF
Web [Domain Configuration] page
with an account which has single
domain privilege larger than DID15.
In addition, the operation for such
LSB by [XSB Config...] button cannot
be performed.
deleteboard(8), moveboard(8)
command on the XSCF shell.
RTIF1-
081030-002
O
O
O
When the timezone other than three Use the showlogs(8) command on
characters has been set, the error logs the XSCF shell.
cannot be displayed on XSCF Web
[Error Log] page.
In addition, XSCF Web [Panic Log]
and [IPL Message Log] pages display
the date on the table with "---".
RTIF1-
081104-001
O
O
O
O
O
O
The monitor message log might not be No workaround is available.
registered when a PCI slot error
detected.
Use the showlogs errorcommand
or the fmdumpcommand to check the
fault information of PCI slot.
RTIF1-
090108-001
The domain console may display this This message can be safely ignored.
message:
ipsec_check_inbound_policy:
Policy Failure for the
incoming packet (not secure)
RTIF1-
O
O
O
In the dual power feed system, when You need to remove and insert the
090108-002
the power failure and the power
recovery occurred repeatedly on one
line, all domains will be forcibly
powered off. And in the power
power cable.
recovery from the forced power-off, a
PSU illegal configuration might be
registered in the error log and the
power recovery might not be started.
Chapter 3 Information About Software
45
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
090108-004
O
When you replaced the XSCF unit by Turn off all mainline switches and
using the replacefru(8) command, then turn on again.
there may be the DB
synchronization timeouterror
log and the replacement may fail.
Or, just after the XSCF reset using the
rebootxscf(8) command, execute
the replacefru(8) command and
extract and insert the XSCF unit
which failed in replacement. If time
passed after the XSCF reset, the
replacefru(8) command may fail
again. In such a case, re-execute the
rebootxscf(8) command and the
subsequent processes.
RTIF1-
090115-001
O
O
O
When you execute the settelnet - After you stopped the Telnet service,
c disablecommand, the Telnet execute the rebootxscf(8) command
service will be stopped immediately. to reset XSCF.
However, unless you reset XSCF by
using the rebootxscf(8) command,
you might fail in the subsequent
restart of the Telnet service.
RTIF1-
090122-001
O
O
O
O
O
O
"Power recovery" message is
output before a PSU has recovered.
After the "Power recovery"
message is output, wait 60 seconds
before removing power from another
PSU.
RTIF1-
090220-001
In a system connecting multiple hosts All the RCI hosts shall be powered on.
and one or more I/O units with RCI,
the power-on operation to one of
those RCI hosts may not power on
any of the RCI I/O units.
46
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
O
M3000 server:
M3000 server:
090220-002
After replaced the motherboard unit, If you have set the power interlocking
the RCI power interlocking setting
restores to its default value.
to a value other than the default,
replace the motherboard unit and
then use the setpwrmode(1M)
command to set the power
interlocking setting again.
M4000/M5000/M8000/M9000
servers:
M4000/M5000/M8000/M9000
servers:
After replaced the XSCF unit, the RCI
power interlocking setting restores to
its default value.
If you have set the power interlocking
to a value other than the default,
replace the XSCF unit and then use
the setpwrmode(1M) command to set
the power interlocking setting again.
RTIF1-
090427-001
O
O
O
O
O
O
Despite the normal status of ppp
network interface for the XSCF-to-
Oracle Solaris OS communication,
SNMP MIB notifies abnormal (down)
status.
No workaround is available.
RTIF1-
After the XSCF reset or switching, you No workaround is available.
090427-002
cannot login to XSCF using the LDAP
server.
Log in to active XSCF with initial
account and execute the following
command to re-import the certificate
chain.
'setldap-c<remote file>'
RTIF1-
O
After changed the SNMP
No workaround is available.
090427-003
configuration, when the XSCF
switching occurred, the SNMP
function may be disabled.
In case this event occurred, the
following command may restore the
function.
setsnmpdisable
setsnmpdisable (run "setsnmp
disable" twice)
setsnmpenable
RTIF1-
O
O
O
In the setting of the LDAP server, the To set 129 characters or more, use
090427-004
number of characters which can be
specified differs between XSCF Web
and XSCF Shell. On XSCF Web, you
can enter up to 128 characters.
XSCF Shell.
Chapter 3 Information About Software
47
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
090430-001
O
O
O
O
O
O
After you set https (using the self
certificate authority) by XSCF and
created a web server certificate, https
is not enabled.
To set the self certificate authority, do
not specify a blank character.
RTIF1-
In a domain mounted with the 10
The entire output can be safely
090508-001
Gigabit Ethernet card (SE0X7HE1F), ignored.
when you set the OpenBoot PROM
environmental variable diag-
To avoid these outputs, execute the
following command at the ok prompt
and set the OpenBoot PROM
environmental variable diag-
switch?to false.
switch?to true, the following
warning message appears on the
console; and at the same time, "Msg:
Device error (FCode informed
error)" is recorded in the error log.
setenvdiag-switch?false
WARNING: /pci@7,700000: FCODE
map-in doesn't match decoded
register type;
And when you execute the
showstatus(8) command,
"Degraded" might be indicated to the
FRU which has installed the relevant
card.
RTIF1-
090729-001
O
O
O
O
O
O
When you use the sethttps(8)
command to create a self-signed web parameter and execute the
Specify up to 100 characters in the
server certificate, if the number of
characters specified in a parameter
exceeds 100, it might result in an
internal error and you might fail to
create the certificate.
sethttps(8) command again.
RTIF1-
The XSCF might go down and require Use only user accounts with a user ID
090824-001
a reboot, or you might see console
error messages and a core dump
(UID) value between 100 and 60000.
This is the range of auto-assigned
(ereport.chassis.software.cor UIDs for the XSCF command
e) when one of these conditions
adduser(8).
occurs:
• A local account has been created
with a user ID explicitly assigned to
a value larger than 65536 (adduser
-u uid).
• An LDAP account has been used
that has a UID value larger than
65536.
48
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
091021-001
O
O
O
While Active Directory is enabled, if No workaround is available.
you log in to XSCF using the user
name that registered in the Active
Directory server, the
showconsolepath(8) command
displays not the log-in user name but
"proxyuser."
RTIF1-
091105-001
O
O
O
O
O
O
O
O
O
The loadcertconsoleoperand is For how to use the loadcert
missing in the usage of the setad(8) consoleoperand of the setad(8)
command and the setldapssl(8)
command and the setldapssl(8)
command.
command, see the man page.
RTIF1-
091109-001
Once domain detected a failure
related to DIMM or PCI, the error log
about this detected failure is
This error log can be safely ignored.
registered every time you reset XSCF.
RTIF1-
091109-002
In XSCF, a host route cannot be set.
When you use the setroute(8)
command and set the netmask to
255.255.255.255, it returns an invalid
parameter.
To set a host route, do not specify the
netmask (-moption).
RTIF1-
091109-003
O
O
O
When you use the setnameserver(8) To delete all the DNS servers or the
command or the DNS setting of XSCF search paths which are same, specify
Web to register the DNS server or the just one corresponding address or
search path, you can specify two or
more same IP address values or
domain names.
domain name.
In doing so, all the DNS servers or the
search paths which are same will be
deleted. Register the data again as
needed.
However, when you specify two or
more same IP address values or
domain names to delete all the DNS
servers or the search paths which are
same, it returns an invalid
parameter.
Chapter 3 Information About Software
49
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
091109-004
O
O
O
When you start up a domain from a No workaround is available.
disk connected to the 8-Gbps fibre
channel card (SE0X7F21F, SE0X7F22F),
if you fail to specify the correct device
path, an error log with no message
might be registered.
In the same way, when you start up a
domain from a disk connected to the
8-Gbps fibre channel card (SE0X7F21F,
SE0X7F22F), even if the domain
cannot be started due to reasons such
as disk failure, the error log might not
be registered.
RTIF1-
100331-001
O
O
O
O
O
O
When you use the sethttps(8)
command to create the self CA
When you create the self CA or a CSR,
do not specify only spaces in the e-
(Certificate Authority) or a CSR with mail operand.
specifying only spaces (for example, "
") in the e-mail operand, the HTTPS
function might be disabled after XSCF
reset.
RTIF1-
100713-001
While the Oracle Solaris 10 OS is
running, if the occurrence of RED
State Exception or watchdog reset or
the execution of the reset xir
command lead to the domain CPU
reset, the following message might be
displayed on the console and the
domain might hang up.
No workaround is available.
When this problem occurred, turn off
the domain power and then turn on.
Failed to complete trap
processing. mid=0xXX
RTIF1-
100713-002
O
O
Though the mainline switch is on, the No workaround is available.
showhardconf(8) command might
wrongly display the PSU status as
"Breaker off" or "Input fail."
RTIF1-
O
O
When you start up a domain from a No workaround is available.
101019-001
hard disk that is connected to the
8Gbps fibre channel card (SE0X7F21F,
SE0X7F22F) and if you specified an
erroneous value after "@" of the disk
node, the message of "Fast Data
Access MMU Miss" might be
displayed on the console and an error
log of the TT=0x68 trap might be
registered.
When this event occurred, change the
environmental variable 'auto-boot?'
setting to falseand use the reset-
allcommand to reset the domain.
Then, specify the correct value after
"@" of the disk node and execute the
bootcommand.
50
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
When you use the setroute(8)
To set the gateway address, confirm
101029-001
command to set the gateway address that the specified address is correct.
and if you mistake the setting or the
gateway has been stopped, XSCF
hangs up.
Before resetting XSCF by using the
rebootxscf(8) command, confirm
that the gateway is operating
properly.
RTIF1-
101102-001
O
When the RCI of XSCF#0 and XSCF#1 No workaround is available.
both got the communication error, the
RCI asynchronous monitoring of
PRIMECLUSTER does not stop
normally but gets the monitoring
timeout, which delays the switching
of the clusters after the error detected.
RTIF1-
101215-001
O
O
O
O
While XSCF is running, a process may No workaround is available.
go down. After this, XSCF may reset.
This error log can be safely ignored.
RTIF1-
When you execute the
No workaround is available.
110216-001
showenvironmentpowercommand
on the M3000 server with SPARC64
VII+ processors, an incorrect value of
"Permitted AC power consumption"
will be displayed with 470 W.
The correct value is 505 W for AC
power supply model and is 510 W for
DC power supply model.
RTIF1-
110222-002
O
O
O
XSCF sends Rst packets without AH This message can be safely ignored.
header when dscp disconnects,
resulting in false security
vulnerability alarms.
The domain console may display this
message:
ipsec_check_inbound_policy:
PolicyFailurefortheincoming
packet(notsecure).
RTIF1-
110523-001
O
O
O
O
O
O
From the OpenBoot PROM prompt
(ok), when you use the nvalias
command and try setting a devalias already defined in the environmental
to the environmental variable variable nvramrc, by using the
nvramrc, it may fail with the message nvaliascommand. Then, in order to
In case this event occurred, redo the
settings of any devaliaswhich is
"Can't create new alias
because nvramrc is full."
set the failed devalias, execute the
nvaliascommand once again.
RTIF1-
Invoking ' setsnmp' without options To view 'setsnmp' usage options, use
110526-001
results in segmentation fault.
'setsnmp-h'
Chapter 3 Information About Software
51
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
M4000/M5000 servers:
M4000/M5000 servers:
111005-001
When the system board is mounted No workaround is available.
with the onboard I/O device, if you
use the addboard(8) or
moveboard(8) command to add or
move it by DR, operation may fail.
M8000/M9000 servers:
No workaround is available.
M8000/M9000 servers:
When the system board is mounted
with one or more PCI cards, if you use
the addboard(8) or moveboard(8)
command to add or move it by DR,
operation may fail.
However, when it is mounted with
the IOUA card only, this event does
not occur.
In either case, the following message
appears on the destination domain
console of the system board addition
or move:
interpreter[xxxx]: ABORT: Bus
Error: Fault Addr: 0xXXXXXXXX
efdaemon[xxx]: Wait pid: xxxx
status: xxx
WARNING: IKP: Unable to
probe PCI leaf (x-x-x)
RTIF1-
111028-001
O
O
O
In the setemailreport(8),
setsmtp(8) or setupplatform(8)
command, any of the following
characters in the e-mail address will
result in an error.
Do not specify these characters in the
e-mail address.
!#$%&'*+/=?^`{|}~
52
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
111227-001
O
O
In case of connecting an External I/O No workaround is available.
Expansion Unit, a message
"watchdogtimeout" might be
registered in an error log under the
following conditions:
• Executes the switchscf(8)
command on an M8000/M9000
server.
• Executes the rebootxscf(8)
command on an M4000/M5000
server.
RTIF1-
120216-001
O
O
O
O
Using PCI hot-plug, when you
perform active addition of a PCI card addition of a PCI card or before the
or perform active maintenance or
active addition of an external I/O
expansion unit, operation may fail.
Stop the target domain before the
maintenance or addition of an
external I/O expansion unit.
RTIF1-
120319-001
O
O
An internal module of the fmd
process (case-close module) turns
disable.
No workaround is available.
If you continue the operation in a
situation that the case-close module is
disabled, an fmd process failure
occurs and resets XSCF.
RTIF1-
120322-001
O
O
O
While starting the XSCF, SSH service Wait for the XSCF finish starting, and
might be disabled and cannot perform then retry SSH connection. If SSH
SSH connection.
service has been disabled, reset the
XSCF using the rebootxscf(8)
command.
RTIF1-
120326-001
O
O
O
O
Executing the setldap -pcommand No workaround is available.
with a password that contains a colon
(:) causes a problem with the LDAP
configuration.
To recover, execute
restoreconfig(8) to restore the last
working LDAP configuration.
RTIF1-
120329-004
O
O
When the XSCF unit is monitored by Disable Ops Center monitoring of the
Ops Center, the service processor can XSCF unit.
sometimes reset or failover
unexpectedly.
RTIF1-
There is a low probability that FRU
No workaround is available.
120424-001
replacement, which is performed by
the replacefru(8) command, fails
during the replacement of XSCF
boards in reaction to an XSCF board
failure.
Chapter 3 Information About Software
53
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
120611-001
O
O
When an External I/O Expansion Unit No workaround is available.
error occurs with the following error
code, the showlogs(8) command may
fail to display an error message or the
component that needs to be replaced.
Code: XXXXXXXX-00XXXXXX-
11XXXXXXXXXXXXXX
RTIF1-
120703-001
O
O
O
O
There is a low probability that the
XSCF restart fails, when the XSCF
reset occurs due to an XSCF
malfunction.
No workaround is available.
If the defect occurred, contact a
qualified service engineer.
RTIF1-
121112-001
O
At the XSCF start, when the machine No workaround is available.
administration of Enhanced Support
Facility (ESF) which operates in the
M4000/M5000/M8000/M9000
domain detected an error, process
servers:
failure may occur which leads to the
Check that XSCF is started.
XSCF reset.
If not started, stop all the domains
and then execute the system power
off/on (AC OFF/ON).
To turn on the system power that you
turned off, wait at least 30 seconds
before power-on.
M3000 server:
Check that XSCF is started.
If not started, stop the domain and
then execute the system power off/on
(AC OFF/ON).
To turn on the system power that you
turned off, wait at least 30 seconds
before power-on.
RTIF1-
O
When a temporary blackout occurred No workaround is available.
121112-002
on the M3000 server which is
mounted with the SPARC64 VII+
processor, the "Power subsystem
failure(detector=2)"
When this event occurred, power off
and wait at least 30 seconds before
power-on.
If not recovered with the above
workaround, contact a qualified
service engineer.
error is detected erroneously and the
domain may be suspended.
54
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
121112-003
O
O
O
After the getflashimage -v
command execution, the URI of the
send file and the receive file in
transfer from 'A' to 'B' is displayed
vice versa.
No workaround is available.
Read the URI display of the send file
and the receive file in reverse order.
RTIF1-
121112-004
O
O
O
After the fmdump(8) command
execution, the FRU information of
each event may fail to display the last
one character in part, serial, and
revision.
No workaround is available.
When the FRU at the time of event
occurrence is mounted, you can use
the showhardconf(8) command to
confirm the FRU information.
RTIF1-
121112-005
O
O
In the following configuration, active In the relevant configuration, stop the
replacement or active addition of a domain and perform cold addition or
PCI card may fail. On this occasion, cold replacement and then start the
Ap_Id format and devices may not be domain.
shown correctly by cfgadm(1M).
To add a PCI card to the
M4000/M5000 servers, perform the
active addition of a PCI card to a slot
other than PCIE#1.
M4000/M5000 servers:
• Active addition of a PCI card to
PCIE#1
To restore, reboot the domain.
• Active addition of a PCI card to a
PCI slot on the External I/O
Expansion Unit
• Active replacement or active
addition of a link card on PCIE#1
which is connected with the
External I/O Expansion Unit
M8000/M9000 servers:
• Active addition of a PCI card to a
PCI slot on the External I/O
Expansion Unit
RTIS1-
130218-004
O
O
O
O
After updating the firmware, the Fault Execute the following command or
Management MIB Trap might not be reset the XSCF.
reported.
XSCF> setsnmp disable
XSCF> setsnmp enable
RTIF1-
M9000 server:
No workaround is available.
120329-002
When using the expansion cabinet,
the depleted life (lifetime) of the
energy storage module (ESM) on a
Sun Flash Accelerator F20 PCIe card
might be inaccurately reported or
cannot be displayed. A depleted ESM
might not be reported.
Chapter 3 Information About Software
55
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
130404-001
O
O
O
If the domain POR reset is executed Once the domain shutdown
while a domain shutdown is being processing is completed, execute the
executed, the domain status may be domain POR reset.
erroneously displayed as Running
(Waiting for OS Shutdown)after
the domain is restarted.
To restore, restart the domain.
RTIF1-
O
O
O
In an environment in which RCI is
If the power status display is incorrect
130404-002
used, the setrci-c statecommand for a server, first turn off the power to
may not correctly display the power the server. If the power to the server
status of the server.
cannot be turned off, disconnect the
RCI cable of the server, together with
the T-branch connector, and then turn
off the power.
If the above status occurs in a
configuration with multiple servers
and I/O units connected by RCI, the
I/O unit power may erroneously be After turning off the power to a server
turned off when the power to one
server is off.
that does not correctly display the
power status, restore the system by
first disconnecting the server and then
restarting the input power (AC
OFF/ON).
RTIF1-
O
O
O
In a configuration with multiple servers To connect a powered-on I/O unit to a
130404-003
and I/O units connected by RCI,
connecting a powered-on I/O unit to a
powered-off server with an RCI cable
may erroneously turn off the I/O unit
power.
server with an RCI cable, either connect
the I/O unit before turning on the server
input power (AC ON), or connect the I/O
unit to the server when the server power
is on.
Connecting with the power off in cases
where the server RCI settings are
initialized does not cause a problem.
RTIF1-
130719-001
O
O
If a memory failure occurs after the
following error is detected, POST
diagnosis may hang up without
isolating the faulty memory.
No workaround is available.
Msg: DIMM serious error
Msg: Unexpected Trap type
(tt=0x32)
RTIF1-
130801-001
O
O
O
The following error log may be registered This error log may be ignored.
when the restoredefaults(8) command is
This trouble can be worked around by
executed on XCP 1115:
re-setting the XSCF configuration.
XSCFselfdiagnosiswarning
detection.
As a permanent solution, upgrade to
XCP 1116 or newer.
56
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
O
O
While XSCF is running, "XSCF panic M3000/M4000/M5000 servers:
130801-002
detected" or "XSCF Watchdog
timeout" may be detected, and XSCF
may reset.
Check that XSCF is started.
If not started, stop all the domains and
then execute a system power off/on (AC
OFF/ON).
To turn on the system power that you
turned off, wait at least 30 seconds before
power-on
M8000/M9000 servers:
Check that XSCF is started.
If there is an XSCF unit that has not
started, execute the replacefru(8)
command, and then uninstall the
XSCF unit and install it again. Or stop
all the domains and then execute a
system power off/on (AC OFF/ON).
To turn on the system power that you
turned off, wait at least 30 seconds
before power-on.
RTIF1-
140131-001
O
O
In a configuration with even
numbered domains, if the following explained in "Explanation", please
operation is executed when half of the execute any of the following
Before executing the operations,
domains have become inactive
("Degraded" state) due to failures or
configuration violations, either all the
running domains stop or the
procedures.
• Perform maintenance to fix the
malfunctions and configuration
violations.
restarting of such domains fails
• Stop all domains.
[How to restore]
[Operation]
1. Remove all XSBs of the domains
which have become inactive (by
executing the "deleteboard -c
unassign" command).
Perform maintenance to fix the
malfunctions and configuration
violations. Restart the domain after
that.
2. Execute poweron or reboot on any
of the domains which has not become
inactive.
RTIF1-
140611-001
O
O
O
O
O
O
Security fixes. (CVE-2014-0224)
Update the XCP firmware to XCP 1118
or newer.
For detail, see the Oracle website
related to the following:
Critical Patch Updates
RTIF1-
141002-001
Security fixes (CVE-2014-7169. See
MOS Doc Id 1934739.1)
Update the XCP firmware to XCP 1118
or newer.
Chapter 3 Information About Software
57
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
141117-001
O
O
O
O
Security fixes. (CVE-2014-3566)
Update the XCP firmware to XCP 1119
or newer.
For details, see the Oracle website
related to the following:
Critical Patch Updates
RTIF1-
150605-001
When using the addfru(8) or
replacefru(8) command for the
maintenance of the CPU/memory
board unit, executing the system
board diagnostic test may cause the
XSCF process to fail.
No workaround is available.
Update the XCP firmware to XCP 1120
or newer.
RTIF1-
O
O
O
It may be temporarily impossible to No workaround is available.
150605-002
log in to the XSCF shell.
Wait a few hours before attempting to
log in to the XSCF shell again.
However, at this time, it is still
possible to use the ICMP response
and the XSCF network functions such
as XSCF Web.
Even if it is possible to log in to the
XSCF shell, executing the console
command may end with an error and
the following message displayed:
"console: error while loading
shared libraries:
libgcc_s_nof.so.1: cannot
open shared object file
Input/output error"
RTIF1-
150605-003
O
If the active XSCF unit fails, and
active replacement is performed for replacement of the active XSCF unit,
that failed XSCF, synchronous turn off the power to all domains and
If this issue occurs during active
processing of logs between the active then turn off and on the input power
and standby XSCF units may not be (AC OFF/ON).
performed normally, and the log data
below will not be inherited by the
replacement XSCF unit.
To turn on the input power that you
turned off, wait at least 30 seconds
before power-on.
• showlogs power
• showlogs event
58
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-6
ID
XCP Issues Fixed in Releases Earlier Than XCP 1121 (Continued)
M4000 M8000
M3000 M5000 M9000 Description
Workaround
RTIF1-
O
If you specify the USB device with
Use the remote host instead of the
150608-001
XCP 1119 and execute the XSCF shell USB device and execute the command
commands below, or execute XSCF
Web menu items corresponding to the
XSCF shell commands below, the
system displays "Unable to mount
USB device" and the process fails.
or operate XSCF Web.
Or, after updating the firmware to
another version other than XCP 1119,
execute the command or operate
XSCF Web.
• dumpconfig
• getflashimage
• restoreconfig
• setad
• setldapssl
• showcodactivationhistory
• snapshot
RTIF1-
150707-001
O
O
O
Security fixes. (CVE-2013-5704, CVE- Update the XCP firmware to XCP 1120
2014-3570, CVE-2014-3572, CVE-2014- or newer.
8275, CVE-2015-0204, CVE-2015-0235)
For detail, see the Oracle website
related to the following:
Critical Patch Updates
Oracle Solaris OS Issues and
Workarounds
This section contains information about Oracle Solaris OS issues. TABLE 3-7, TABLE 3-8,
might encounter, depending upon which Oracle Solaris OS release you are using.
Oracle Solaris Issues and Workarounds for All
Supported Releases
TABLE 3-7 lists Oracle Solaris OS issues that you might encounter in any supported
release of Oracle Solaris OS.
Chapter 3 Information About Software
59
TABLE 3-7
Oracle Solaris OS Issues and Workarounds for All Supported Releases
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
4816837
O
O
O
System hangs when executing parallel No workaround is available.
hot plug operation with SP DR in
suspend phase.
6449315
The Oracle Solaris OS cfgadm(1M)
command does not unconfigure a
DVD drive from a domain on a
M8000/M9000 server.
Disable the Volume Management
Daemon (vold) before unconfiguring
a DVD drive with the cfgadm(1M)
command.
To disable vold, stop the daemon by
issuing the command
/etc/init.d/volmgt stop. After
the device has been removed or
inserted, restart the daemon by
issuing the command
/etc/init.d/volmgt start.
6459540
O
O
The DAT72 internal tape drive on a
server may time out during tape
operations. The device might also be
identified by the system as a QIC
drive.
Add the following definition to
/kernel/drv/st.conf:
tape-config-list=
"SEAGATE DAT
DAT72-000",
"SEAGATE_DAT____DAT72-000",
"SEAGATE_DAT____DAT72-000";
SEAGATE_DAT____DAT72-000=
1,0x34,0,0x9639,4,0x00,0x8c,0
x8c,
0x8c,3;
There are four spaces between
SEAGATE DAT and DAT72-000.
6466617
O
O
Performing a hot plug operation with Pause a few seconds between the
the PCI-Express slot too quickly
interrupts a PCI leaf reset and occurs
the following error:
issue of each cfgadm -ccommand.
cfgadm: Component system is
busy error
60
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-7
Oracle Solaris OS Issues and Workarounds for All Supported Releases (Continued)
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
6481002
O
O
O
Installing the Oracle Solaris OS from If you are using a Sun PCI-E Dual
the network using certain PCI-Express Gigabit Ethernet MMF Adapter or a
cards may cause a panic.
Sun PCI-E Dual Gigabit Ethernet UTP
Adapter, do not install the Oracle
Solaris OS using either of these cards.
Instead, use other network devices,
such as the onboard Gigabit Ethernet
or another network device.
6515648
O
O
"Replumb Failed" error appears
when dr@0:SB1::memoryfails.
Once the DR operation is complete, it
can be plumbed up manually.
Example steps to re-plumb the
interface manually:
# ifconfiginterface plumb
xxx.xxx.xxx.xxx netmask+
broadcast+up
# ifconfig interface group group-
name
# ifconfiginterface addif
xxx.xxx.xxx.xxx -failover
deprecatedup
This workaround assumes that the
/etc/hostname.interface file is
correctly configured for the IPMP
group and does not need any
modification. The IP addresses used
in the example above should match
what was previously used and what
matches the
/etc/hostname.<interface> file.
6516135
6519290
O
O
O
O
Ap_Id format and devices may not be Use the following operations to
shown correctly by cfgadm(1M).
display all of the PCI slots.
1) devfsadm(at Oracle Solaris OS
prompt)
2) cfgadm
O
Large amounts of I/O on swap
devices can cause the system to
appear hung by overwhelming the
I/O system. The amount of I/O
required can be generated through a
number of ways, eg memory shortage,
heavy use of /tmpetc.
Set the following to /etc/system
and then reboot the domain:
setmaxfastscan=0x2000
Chapter 3 Information About Software
61
TABLE 3-7
Oracle Solaris OS Issues and Workarounds for All Supported Releases (Continued)
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
6522017
O
O
DR and ZFS may not be used in the This has been fixed in Oracle Solaris
same domain.
11.1 SRU3.4.
Reduce the amount of kernel memory
that ZFS can allocate by setting the
zfs_arc_max parameter in the
/etc/system file. The following
example sets the maximum size to 512
Mbytes.
setzfs:zfs_arc_max =
0x20000000
6529714
6530753
O
O
O
O
Warning messages occur while trying No workaround is available.
to configure more than four PCIe
Quad-port Gigabit Ethernet UTP Low
Profile Adapter (X4447A-Z) or PCIe
Dual 10 Gigabit Ethernet Fiber XFP
Low Profile Adapter (X1027A-Z) into
one I/O Boat.
Some of the PCI slots in the External Use one of the following operations to
I/O Expansion Unit PCI slots are not display all of the PCI slots.
displayed during a normal boot
operation.
• boot -r (at open boot prompt)
• devfsadm -C (at Oracle Solaris
OS prompt)
• cfgadm (twice at Oracle Solaris OS
prompt)
6531036
6532215
O
O
O
O
O
O
The error message network
initialization failedappears message can be safely ignored.
repeatedly after a boot net
installation.
No workaround is available. This
volfs or dscp service may fail when Restart the service if the failure is
domain is booted.
observed. To avoid the problem, issue
the following commands.
svc:/platform/sun4u/dscp:defa
ult: Method
"/lib/svc/method/svc-dscp
start" failed with exit
status 95.
# svccfg -s dscp setprop
start/timeout_seconds=count:
300
# svccfg -s volfs setprop
start/timeout_seconds=count:
300
svc:/system/filesystem/volfs:
default: Method or service
exit timed out. Killing
contract 59.
# svcadm refresh dscp
# svcadm refresh volfs
62
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-7
CR ID
Oracle Solaris OS Issues and Workarounds for All Supported Releases (Continued)
M4000 M8000
M3000 M4000 M9000 Description
Workaround
6565553
6611966
6674266
O
O
DR deleteboard(8) and
moveboard(8) operations might fail.
No workaround is available.
Try DR operations again.
Example for messages on domain:
drmach: WARNING: Device
driver failure: /pci
dcs: <xxxx>
config_change_state:Hardware
specific failure: unconfigure
SB1: Device driver failure:
/pci
6588650
6589644
O
O
O
On occasion, the system is unable to No workaround is available.
DR after an XSCF failover or XSCF
reboot.
After added the system board by DR, The console can be recovered by
when the switching occurred on the pressing Ctrl-q(the [Ctrl] key and
redundant XSCF Units, the domain
console may hang up.
the [q] key).
6592302
6619344
O
O
O
O
Unsuccessful DR operation leaves
memory partially configured.
Try deleteboard(8) again.
The Sun Crypto Accelerator (SCA)
6000 might not work if hot-plug
configured into slot 1.
To avoid the defect, do not hot-plug
this card in slot 1.
6625734
Systems with large number of
processors in a single domain
environment may have suboptimal
Use processor sets to bind application
processes or LWPs to groups of
processors. Refer to the psrset(1M)
performance with certain workloads. man page for more information.
Chapter 3 Information About Software
63
TABLE 3-7
Oracle Solaris OS Issues and Workarounds for All Supported Releases (Continued)
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
6660168
O
O
O
If a ubc.piowbeue-cpu error
If FMA service fails, issue the
occurs on a domain, the Oracle Solaris following command on the domain to
Fault Management cpumem-diagnosis recover:
module might fail, causing an
interruption in FMA service.
# svcadm clear fmd
If this happens, you will see the
following output in the console log:
Then restart cpumem-diagnosis:
# fmadm restart cpumem-
diagnosis
SUNW-MSG-ID: FMD-8000-2K,
TYPE: Defect, VER: 1,
SEVERITY: Minor
EVENT-TIME: Fri Apr 4
21:41:57 PDT 2008
PLATFORM: SUNW,SPARC-
Enterprise, CSN: 2020642002,
HOSTNAME: <hostname>
SOURCE: fmd-self-diagnosis,
REV: 1.0
EVENT-ID: 6b2e15d7-aa65-6bcc-
bcb1-cb03a7dd77e3
DESC: A Solaris Fault
Manager component has
experienced an error that
required the module to be
disabled. Refer to
http://sun.com/msg/FMD-8000-
2K for more information.
AUTO-RESPONSE: The module has
been disabled. Events
destined for the module will
be saved for manual
diagnosis.
IMPACT: Automated diagnosis
and response for subsequent
events associated with this
module will not occur.
REC-ACTION: Use fmdump -v -u
<EVENT-ID> to locate the
module. Use fmadm reset
<module> to reset the
module.
6745410
O
O
O
Boot program ignores the Kadb option Use kmdb instead of kadb.
which causes the system not to boot.
64
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-7
Oracle Solaris OS Issues and Workarounds for All Supported Releases (Continued)
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
6765239
O
If a SAS device containing multiple
SAS targets is connected to the
Use a FUJITSU SAS card
(SE0X7SA1F/SE0X7SA1X).
onboard external SAS interface, it will Alternatively, check for the
not work properly. The enumeration availability of a patch for this defect.
of target IDs within the device may
change across reboots.
7009469
7135497
18055846
O
Creating a RAID configuration using No workaround is available.
the raidctl(1M) command generates
a warning message on initial process.
Subsequent RAID configurations do
not.
O
O
On systems with the RAID-capable
IOUA card and running Oracle Solaris
11, after a RAID volume is deleted
using raidctl -d, the format
utility does not display the disk.
Execute cfgadm(1M):
cfgadm -c unconfigure cn where cn
is the controller number.
O
When adding a PCI Expansion Unit There is no effective workaround.
by PCI hot plugging (PHP), Oracle
Solaris may panic by producing the
following message:
Do not hot plug a PCI Expansion
Unit. There is no problem in hot
replacing in respect to the same slot.
panic[cpuX]/thread=
XXXXXXXXXX: Fatal error has
occurred in: PCIe
fabric.(0xX)(0xXX).
Chapter 3 Information About Software
65
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
8/11
TABLE 3-8 lists issues that have been fixed in Oracle Solaris 10 8/11 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 8/11.
TABLE 3-8
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/11
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
6344060
O
A lot of following ereports are
This has been fixed in patch 144563-
outputted when both interfaces bge1 02.
and bge2 are in use.
ereport.io.device.badint_limi
t
[Workaround]
No workaround is available.
ereport.io.service.lost
This issue is caused on the M3000
server with SPARC64 VII+ processors.
6794630
O
An attempt to use the GUI to install This has been fixed in patch 124630-
Oracle Solaris OS in a domain larger 45.
than 2TB might fail.
[Workaround]
Use the Command Line interface to
install Oracle Solaris OS.
6966086
7009474
O
O
bge interface can not enable Jumbo
Frame function.
This has been fixed in patch 144563-
02.
This issue is caused on the M3000
server with SPARC64 VII+ processors.
[Workaround]
No workaround is available.
O
O
After applying 144563-01 or 144563-02 This has been fixed in patch 144500-
patch, bge interface does not display 19.
link up/down message on console
and syslog when
plugging/unplugging.
[Workaround]
Check the link status by using the
"link_state" parameter from the
kstat(1M) output.
66
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
9/10
TABLE 3-9 lists issues that have been fixed in Oracle Solaris 10 9/10 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 9/10.
TABLE 3-9
Oracle Solaris OS Issues Fixed in Oracle Solaris 10 9/10
M4000 M8000
CR ID
M3000 M4000 M9000 Description
Workaround
6668237
O
O
O
After DIMMs are replaced the
This has been fixed in patch 143527-
corresponding DIMM faults are not 01.
cleared on the domain.
[Workaround]
Use the following commands:
# fmadm repair fmri|uuid
# fmadm rotate
6872501
O
O
O
If 139555-08 has been applied to the This has been fixed in patch 143359-
domain, cores are not offlined when 02.
requested by the XSCF. This CR
effects only Oracle Solaris 10 5/09,
Oracle Solaris 10 10/09, and an earlier
version of Oracle Solaris OS with
patch 139555-08.
[Workaround]
Use fmdump(8) with its -voption on
the Service Processor to identify the
faulty core. Once identified, use
psradm(1M) on the domain to offline
the core.
6888928
O
O
O
IPMP interface fails since probe
packets are not sent through that
interface. Problem occurs with
M3000/M4000/M5000/M8000/M9000
servers running any Oracle Solaris
release running IPMP with patch
141444-09 installed.
This has been fixed in patch 142900-
02.
[Workaround]
Disable probe-based failure detection.
Chapter 3 Information About Software
67
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
10/09
TABLE 3-10 lists issues that have been fixed in Oracle Solaris 10 10/09 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 10/09.
TABLE 3-10 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/09
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6572827
O
O
O
The prtdiag -vcommand reports This has been fixed in patch 141444-
PCI bus types incorrectly. It reports
"PCI" for PCI-X leaf devices and
"UNKN" for legacy PCI devices.
09.
[Workaround]
No workaround is available.
6724307
O
Scheduler decisions systems are
occasionally unbalanced.
This has been fixed in patch 141414-
01.
Sometimes two threads will be on one
core (causing both to run at about half
speed) while another core is idle. For
many OpenMP and similar parallel
applications, the application
[Workaround]
Use processor sets to prevent uneven
threads to core assignment.
performance is limited by the speed of
the slowest thread.
Uneven scheduling is not common,
perhaps 1 in 50 or 1 in 100 decisions.
But if there are 128 threads running,
then the application might have at
least one uneven schedule event.
6800734
O
O
The deleteboard(8) command hang This has been fixed in patch 141444-
in a domain.
09.
[Workaround]
No workaround is available.
68
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-10 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/09 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6816913
O
O
The XSCF showdevices(8) command Use the prtdiag(1M) command on
displays the incorrect processor cache the domain to report processor
size (MB) for fractional processor
cache sizes, such as displaying "5"
when the correct display would be
"5.5."
information.
6821108
O
O
DR and showdevices(8) command This has been fixed in patch 140589-
do not work after XSCF reset.
02.
[Workaround]
Reset the XSCF service processor
twice. Half the SAs (Security
Associations) are deleted the first
time, half are deleted the second time,
so the second addition succeeds and
IPsec communication is reestablished.
6827340
O
O
O
DR and Memory patrol might fail due This has been fixed in patch 142344-
to command error.
01.
[Workaround]
No workaround is available.
Chapter 3 Information About Software
69
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
5/09
TABLE 3-11 lists issues that have been fixed in Oracle Solaris 10 5/09 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 5/09.
TABLE 3-11 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/09
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6588555
O
O
M8000/M9000 servers:
This has been fixed in patch 139555-
XSCF failover during DR operation to 08.
the permanent memory might cause
domain panic.
[Workaround]
M8000/M9000 servers:
M4000/M5000 servers:
Do not start an XSCF failover while a
DR operation is running. Wait for a
DR operation to finish before starting
the failover. If you start the failover
first, wait for the failover to finish
before starting the DR operation.
XSCF reset during DR operation to
the permanent memory might cause
domain panic.
M4000/M5000 servers:
Do not start an XSCF reset while a DR
operation is running. Wait for a DR
operation to finish before starting the
reset.
6623226
O
O
O
The Oracle Solaris OS command
This has been fixed in patch 140336-
lockstat(1M) or the dtrace lockstat 01.
provider might cause a system panic.
[Workaround]
Do not use the Oracle Solaris OS
lockstat(1M) command or the
dtrace lockstat provider.
70
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-11 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/09 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6680733
O
O
O
Sun Quad-port Gigabit Ethernet
This has been fixed in patch 139570-
Adapter UTP (QGC) & Sun Dual 10 01.
GigE Fiber XFP Low Profile Adapter
(XGF) NICs might panic under high
[Workaround]
load conditions.
If possible, use the card in x8 slot.
Otherwise, no workaround is
available.
6689757
O
O
O
Sun Dual 10 GigE Fiber XFP Low
This has been fixed in patch 139570-
Profile Adapter (XGF) with a single or 01.
improperly installed XFP optical
transceivers might cause the following
error to show on the console:
[Workaround]
Check and make sure that both XFP
optical transceivers are firmly seated
in the housing. Do not mix INTEL and
Sun XFP optical transceivers in the
same Adapter.
The XFP optical transceiver
is broken or missing.
Do NOT plumb a port with the
ifconfig command if the port does not
contain an XFP optical transceiver or
it contains one but the transceiver is
not in use.
6725885
O
cfgadmwill display non-existent
This has been fixed in patch 140401-
M3000 system boards (SB1 to SB15). 01.
[Workaround]
The cfgadmoutput for SB1-SB15 can
be ignored.
Chapter 3 Information About Software
71
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
10/08
TABLE 3-12 lists issues that have been fixed in Oracle Solaris 10 10/08 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 10/08.
TABLE 3-12 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/08
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6511374
O
O
Unexpected error messages may be
This has been fixed in patch 137137-
displayed on console after changing 09.
the system configuration.
Example:
[Workaround]
WARNING: Translation error
source /LSB0/B0/0, PA
3c000000000, target
/LSB0/B0/20000000
This message can be safely ignored.
6533686
O
O
When XSCF is low on system
resources, DR operations involving
relocating permanent memory, such as
deleteboard(8) or moveboard(8),
might fail with one or more of these
errors:
This has been fixed in patch 138397-
01.
[Workaround]
Retry the DR operation at a later time.
SCF busy
DR parallel copy timeout
This applies only to Quad-XSB
configured system boards hosting
multiple domains.
6535018
O
In Oracle Solaris domains that include This has been fixed in patch 137111-
SPARC64 VII processors, workloads 01.
that make heavy use of the Oracle
Solaris kernel might not scale as
expected when you increase the
thread count to a value greater than
256.
[Workaround]
For Oracle Solaris domains that
include SPARC64 VII processors, limit
domains to a maximum of 256
threads.
6608404
O
Hot-plug of the PCIe Quad-port
Gigabit Ethernet UTP Low Profile
Adapter (X4447A-Z) in slot 1 might
cause other network devices to fail.
To avoid the defect, do not install this
adapter in slot 1.
72
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-12 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6614737
O
O
The DR deleteboard(8) and
For Oracle Solaris 10 5/08 or earlier,
moveboard(8) operations might hang this has been in patch 137111-01.
if any of the following
conditions exist:
• A DIMM has been degraded.
Avoid performing DR operations if
• The domain contains system boards any of the listed conditions exist.
with different memory size.
To determine whether the system
contains degraded memory, use the
XSCF showstatus(8) command. For
sample output, see “Identifying
Degraded Memory in a System” on
To determine whether the domain
contains system boards with different
memory sizes, display the list of
memory sizes using the XSCF
showdevices(8) command or the
prtdiag(8) command on the domain.
For sample of output, see “Identifying
Different Memory Sizes in a System
Board” on page 98
If a DR command hangs, reboot the
domain to recover.
6619224
O
For Oracle Solaris domains that
include SPARC 64 VII processors, a
single domain of 256 threads (128
cores) or more might hang for an
extended period of time under certain
unusual situations. Upon recovery, the
uptime command will show
This has been fixed in patch 137111-
03.
[Workaround]
For Oracle Solaris domains that
include SPARC 64 VII processors, do
not exceed a domain size of 256
virtual CPUs in a single Oracle Solaris
domain. This means a maximum of 32
CPU chips in a single domain
extremely high load averages.
configuration (maximum
configuration for an M8000 server).
Chapter 3 Information About Software
73
TABLE 3-12 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6632549
O
O
fmd service on domain might fail to This has been fixed in patch 138050-
maintenance mode after DR
operations.
01.
[Workaround]
If fmd service fails, issue the
following commands on the domain
to recover:
# svcadm clear fmd
6660197
O
O
DR might cause the domain to hang if This has been fixed in patch 138397-
either of the following conditions
exist:
01.
• A domain contains 256 or more
CPUs.
[Workaround]
1. Set the following parameter in the
system specification file
(/etc/system):
• Memory error occurred and the
DIMM has been degraded.
set
drmach:drmach_disable_mcopy=
1
1. Reboot the domain.
6679370
O
O
O
The following message may be output This has been fixed in patch 137137-
on the console during the system
booting, the External I/O Expansion
Unit adding by hotplug, or the
FMEMA operating by DR.
09.
Note - Before adapting the patch,
delete the following setting from
/etc/system:
set pcie_expected_ce_mask =
0x2001
SUNW-MSG-ID: SUN4-8000-75,
TYPE: Fault, VER: 1,
SEVERITY: Critical
[Workaround]
...
Add the following to /etc/system
and then reboot the domain.
DESC:
A problem was detected in
the PCI-Express subsystem.
set pcie_expected_ce_mask =
0x2001
Refer to
http://sun.com/msg/SUN4-8000-
75 for more information.
...
74
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-12 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 10/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6720261
O
O
O
M4000/M5000/M8000/M9000
servers:
This has been fixed in patch 137137-
09.
If your domain is running one of the
following versions of Oracle Solaris
OS, the system might panic/trap
during normal operation:
[Workaround]
Set the following parameter in the
system specification file
(/etc/system):
• Oracle Solaris 10 5/08 OS
• An earlier version of Oracle Solaris
10 OS with patch ID 127127-11
set heaplp_use_stlb=0
Then reboot the domain.
M3000 server:
If your domain is running Oracle
Solaris 10 5/08 OS, the system might
panic/trap during normal operation:
Chapter 3 Information About Software
75
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
5/08
TABLE 3-13 lists issues that have been fixed in Oracle Solaris 10 5/08 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 5/08.
TABLE 3-13 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
5076574
O
A PCIe error can lead to an invalid
fault diagnosis on a large
M8000/M9000 domain.
For Oracle Solaris 10 8/07 or earlier,
this has been fixed in patch 127127-11.
[Workaround]
Create a file
/etc/fm/fmd/fmd.confcontaining
the following lines;
setprop client. buflim 40m
setprop client. memlim 40m
6402328
6472153
O
O
System panic might happen due to be For Oracle Solaris 10 8/07 or earlier,
over the interrupt number limit when this has been fixed in patch 127127-11.
one domain has seven or more IOUA
cards.
[Workaround]
Use six or less IOUA cards on one
domain.
O
If you create an Oracle Solaris OS
Flash archive on a non-
For Oracle Solaris 10 8/07 or earlier,
this has been fixed in patch 137046-01.
M4000/M5000/M8000/M9000 sun4u
server and install it on a
[Workaround]
M4000/M5000/M8000/M9000 sun4u
server, the console's TTY flags will not
be set correctly. This can cause the
console to lose characters during
stress.
Just after installing Oracle Solaris OS
from an Oracle Solaris Flash archive,
telnet into the
M4000/M5000/M8000/M9000 server
to reset the console's TTY flags a
follows:
# sttydefs -r console
# sttydefs -a console -i
"9600 hupcl opost onlcr
crtscts" -f "9600"
This procedure is required only once.
76
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-13 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6505921
O
Correctable error on the system PCIe For Oracle Solaris 10 8/07 or earlier,
bus controller generates an invalid
fault.
this has been fixed in 120011-14 and
125369-05.
[Workaround]
Contact a qualified service engineer.
6522433
6527811
O
O
O
O
After the CPU hardware error
occurred, the fmdump(1M) command this has been fixed in patch 127127-11.
on the domain may display an
incorrect faulty component.
For Oracle Solaris 10 8/07 or earlier,
[Workaround]
Check system status on XSCF.
The showhardconf(8) command on For Oracle Solaris 10 8/07 or earlier,
the XSCF cannot display PCI card
information that is installed in
External I/O Expansion Unit, if the
External I/O Expansion Unit is
configured using PCI hotplug.
this has been fixed in patch 128346-01.
[Workaround]
No workaround is available.
Chapter 3 Information About Software
77
TABLE 3-13 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6536564
O
O
showlogs(8) and showstatus(8)
command on XSCF might report
wrong I/O component due to wrong
diagnosis by Oracle Solaris Fault
management Architecture when faults
in I/O devices occur.
For Oracle Solaris 10 8/07 or earlier,
this has been fixed in patch 125369-05.
[Workaround]
To avoid this problem, issue the
following commands on the domain.
# cd
/usr/platform/SUNW,SPARC-
Enterprise/lib/fm/topo/plugin
s
# mv ioboard.so
ioboard.so.orig
# svcadm restart fmd
If the following messages are
displayed on the domain, contact a
service engineer.
Example:
SUNW-MSG-ID: SUNOS-8000-1L,
TYPE: Defect, VER: 1,
SEVERITY: Minor
EVENT-TIME: Sun May 6
18:22:24 PDT 2007
PLATFORM: SUNW,SPARC-
Enterprise, CSN: BE80601007,
HOSTNAME: sparc
6545143
O
O
There is a low probability that a
system panic can occur during trap
processing of a TLB miss for a user
stack address. The problem can occur
if the user stack is unmapped
concurrently with the user process
executing a flush windows trap (ta
3). The panic message will contain the
following string:
For Oracle Solaris 10 8/07 or earlier,
this has been fixed in patch 127111-08.
[Workaround]
No workaround is available.
bad kernel MMU trap at TL 2
78
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-13 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6545685
O
O
If the following message displayed on For Oracle Solaris 10 8/07 or earlier,
the OS console, memory degradation this has been fixed in patch 127111-08.
or XSB deconfiguration may occur in
the subsequent reboot.
[Workaround]
Add the following to /etc/system
and then reboot the domain:
Example:
mc-opl: WARNING: mc-opl
rewrite timeout on /LSB0/B0
set mc-opl:
mc_max_rewrite_loop = 20000
6546188
O
O
The system panics when running
hotplug (cfgadm(1M)) and DR
operations (addboard(8) and
deleteboard(8)) on the following
adapters:
For Oracle Solaris 10 8/07, this has
been fixed in patch 127741-01.
No workaround is available.
• PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (X4447A-
Z)
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (X1027A-
Z)
6551356
O
O
The system panics when running
For Oracle Solaris 10 8/07, this has
hotplug (cfgadm(1M)) to configure a been fixed in patch 127741-01.
previously unconfigured adapter. The
message "WARNING: PCI
Expansion ROM is not
[Workaround]
Perform cfgadm -c disconnectto
accessible" will be seen on the
completely remove the card. After
console shortly before the system
waiting at least 10 seconds, the card
panic. The following cards are
may be configured back into the
affected by this defect:
domain using the cfgadm -c
configurecommand.
• PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (X4447A-
Z)
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (X1027A-
Z)
Chapter 3 Information About Software
79
TABLE 3-13 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6559504
O
O
Messages of the form nxge: "NOTICE: For Oracle Solaris 10 8/07, this has
nxge_ipp_eccue_valid_check:
been fixed in patch 127741-01.
rd_ptr =
nnn wr_ptr = nnn" will be
observed on the console with the
following adapters:
[Workaround]
These messages can be safely ignored.
• PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (X4447A-
Z)
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (X1027A-
Z)
6564934
O
O
Performing a DR deleteboard(8)
For Oracle Solaris 10 8/07, this has
operation on a board which includes been fixed in patch 127741-01.
Permanent Memory when using the
following adapters results in broken
connections:
[Workaround]
Re-configure the affected network
•
PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (X4447A-
Z)
interfaces after the completion of the
DR operation. For basic network
configuration procedures, refer to the
ifconfigman page for more
information.
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (X1027A-
Z)
6568417
O
O
After a successful CPU DR
deleteboard(8) operation, the
system panics when the following
adapters are in use:
For Oracle Solaris 10 8/07 or earlier,
this has been fixed in patch 127111-02.
[Workaround]
• PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (X4447A-
Z)
Add the following line to
/etc/system and reboot the system:
set ip:ip_soft_rings_cnt=0
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (X1027A-
Z)
6571370
O
O
Use of the following adapters have
been observed to cause data
corruption in stress test under
laboratory conditions:
For Oracle Solaris 10 8/07, this has
been fixed in patch 127741-01.
[Workaround]
• PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (X4447A-
Z)
Add the following line in
/etc/system and reboot the system:
set
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (X1027A-
Z)
nxge:nxge_rx_threshold_hi=0
80
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-13 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 5/08 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6584984
O
On the M8000/M9000 servers,
For Oracle Solaris 10 8/07 or earlier,
busstat(1M) command may cause this has been fixed in patch 127127-11.
rebooting of domains.
No workaround is available. Do not
use busstat(1M) command.
Check for the availability of a patch
for this defect.
6589546
O
O
prtdiag(8) command does not show For Oracle Solaris 10 8/07 or earlier,
all I/O devices of the following
adapters:
this has been fixed in patch 127127-11.
• Sun StorageTek Enterprise Class
4Gb Dual-Port Fibre Channel PCIe
HBA (XSEFC402AF)
[Workaround]
Use prtdiag -v for full output.
• Sun StorageTek Enterprise Class
4Gb Single-Port Fibre Channel PCIe
HBA (XSEFC401AF)
Oracle Solaris OS Issues Fixed in Oracle Solaris 10
8/07
TABLE 3-14 lists issues that have been fixed in Oracle Solaris 10 8/07 OS. You might
encounter them in supported releases earlier than Oracle Solaris 10 8/07.
Note – CR ID 6534471: Improper handling of large page in kernel memory may
cause random panics. Implement the workaround for CR ID 6534471 or check for the
availability of a patch and install it immediately. This bug has been fixed by 125100-
06 and Oracle Solaris 10 8/07.
Chapter 3 Information About Software
81
TABLE 3-14 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/07
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6303418
O
A M9000 server with a single domain For Oracle Solaris 10 11/06, this has
and 11 or more fully populated
system boards might hang under
heavy stress.
been fixed in patch 120011-01.
[Workaround]
Do not exceed 170 CPU threads.
Limit the number of CPU threads to
one per CPU core by using the Oracle
Solaris psradmcommand to disable
the excess CPU threads. For example,
disable all odd-numbered CPU
threads.
6416224
O
O
System performance can degrade
using a single NIC card with more
than 5,000 connections.
For Oracle Solaris 10 11/06, this has
been fixed in patch 120011-08.
[Workaround]
Use multiple NIC cards to split
network connections.
6441349
6485555
O
O
O
O
The system may hang if there is an
I/O error in the system.
For Oracle Solaris 10 11/06, this has
been fixed in patch 120011-07.
No workaround is available.
On-board Gigabit Ethernet NVRAM For Oracle Solaris 10 11/06, this has
corruption could occur due to a race been fixed in patch 120011-08.
condition.
The window of opportunity for this
race condition is very small.
No workaround is available.
82
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-14 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/07 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6496337
O
O
The "cpumem-diagnosis" module may For Oracle Solaris 10 11/06, this has
fail to load after uncorrectable
error(UE) panic.
been fixed in patch 125369-05.
Systems will function correctly but
events normally automatically
diagnosed by FMA using this module
will require manual diagnosis.
[Workaround]
If problem occurred, implement the
following
workaround:
Example:
1. Remove the following file.
SUNW-MSG-ID: FMD-8000-2K,
TYPE: Defect, VER: 1,
SEVERITY: Minor
# rm /var/fm/fmd/ckpt/cpumem-
diagnosis/cpumem-diagnosis
1. Restart fmd service.
EVENT-TIME: Thu Feb 15
15:46:57 JST 2007
# svcadm restart fmd
PLATFORM: SUNW,SPARC-
To avoid this problem in advance,
add "rm -f
Enterprise, CSN: BE80601007,
HOSTNAME: col2-ff-em7-d0
/var/fm/fmd/ckpt/cpumem-
diagnosis/cpumem-diagnosis" in
/lib/svc/method/svc-dumpadm
file as below.
#
# We haven't run savecore on
a dump device yet
#
savedev=none
rm -f
/var/fm/fmd/ckpt/cpumem-
diagnosis/cpumem-diagnosis
#
6498283
O
O
Using the DR deleteboard(8)
For Oracle Solaris 10 11/06, this has
command while psradmoperations been fixed in patch 120011-07.
are running on a domain might cause
a system panic.
[Workaround]
No workaround is available.
Chapter 3 Information About Software
83
TABLE 3-14 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/07 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6499304
O
O
CPU isn't offlined and unexpected
For Oracle Solaris 10 11/06, this has
message is displayed on console when been fixed in patch 125369-05.
many correctable error(CE) occurs.
Example:
[Workaround]
SUNW-MSG-ID: FMD-8000-11,
TYPE: Defect, VER: 1,
Check CPU status on XSCF.
SEVERITY: Minor EVENT-TIME:
Fri Feb 2 18:31:07 JST 2007
PLATFORM: SPARC-Enterprise,
CSN: BE80601035, HOSTNAME:
FF2-35-0
6502204
O
O
Unexpected error messages may be
For Oracle Solaris 10 11/06, this has
displayed on console on booting after been fixed in patch 125369-05.
CPU UE panic.
Example:
[Workaround]
SUNW-MSG-ID: FMD-8000-11,
TYPE: Defect, VER: 1,
SEVERITY: Minor EVENT-TIME:
Tue Jan 9 20:45:08 JST 2007
If you see unexpected messages, use
the XSCF command
showdomainstatus(8) to check
system status on XSCF.
PLATFORM: SUNW,SPARC-
Enterprise, CSN: 2030636002,
HOSTNAME: P2-DC1-16-d0
6502750
6508432
O
O
O
O
Notification message for inserted or For Oracle Solaris 10 11/06, this has
removed card by PCI hot plug may
not output.
been fixed in patch 120011-08.
No workaround is available.
Many correctable errors (CE) may
occur, and despite these are the
correctable errors, domain may panic.
For Oracle Solaris 10 11/06, this has
been fixed in patch 120011-08.
[Workaround]
Set the following to /etc/system
and then reboot the domain:
set pcie:pcie_aer_ce_mask =
0x2001
6508434
O
O
The domain may panic when an
For Oracle Solaris 10 11/06, this has
additional PCI-X card is installed or a been fixed in patch 120011-08.
PCI-X card is replaced by using PCI
hot plug.
[Workaround]
Do not insert a different type of PCI-X
card on the same PCI slot by using
PCI hot plug.
84
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-14 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/07 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6509337
O
O
O
O
s10s_u3 wanboot fails - The server
returned 416: Requested Range Not
Satisfiable.
No workaround is available.
6510861
When Dual-Channel Ultra320 SCSI
Card (SE0X7SC2F, SE0X7SC2X) is
mounted, correctable errors(CE) occur
and system may panic.
For Oracle Solaris 10 11/06, this has
been fixed in patch 120011-08.
[Workaround]
Add the following entry to the
/etc/system file and then reboot
the system:
set pcie:pcie_aer_ce_mask =
0x31c1
6520990
6527781
O
O
O
Domain may cause a panic when
deleteboard(8) command for kernel been fixed in patch 120011-08.
board by using Dynamic
Reconfiguration (DR).
For Oracle Solaris 10 11/06, this has
[Workaround]
To mask this error, add the following
entry to the /etc/system file.
set drmach:fmem_timeout = 30
The cfgadmcommand fails while
moving the DVD/Tape drive unit
between two domains.
For Oracle Solaris 10 11/06, this has
been fixed in patch 125081-06.
No workaround is available. To
reconfigure DVD/Tape drive unit,
execute reboot -rfrom the domain
exhibiting the problem.
6530178
6530288
O
O
O
O
DR addboard(8) command can hang. For Oracle Solaris 10 11/06, this has
Once problem is observed, further DR been fixed in patch 120011-07.
operations are blocked. Recovery
requires reboot of the domain.
No workaround is available.
Ap_Id format may not be shown
For Oracle Solaris 10 11/06, this has
correctly by cfgadm(1M) command. been fixed in patch 120011-07.
No workaround is available.
Chapter 3 Information About Software
85
TABLE 3-14 Oracle Solaris OS Issues Fixed in Oracle Solaris 10 8/07 (Continued)
M4000 M8000
CR ID
M3000 M5000 M9000 Description
Workaround
6534471
O
O
Systems may panic/trap during
normal operation.
For Oracle Solaris 10 11/06, this has
been fixed in patch 125100-06.
If a patch is not available, disable the
kernel large page sTLB programming.
In the file /etc/system,
change the variable
heaplp_use_stlbto 0:
set heaplp_use_stlb=0
6535564
O
O
PCI hot plug to PCI slot #0, #1 or
For Oracle Solaris 10 11/06, this has
External I/O Expansion Unit may fail been fixed in patch 120011-08.
on XSB added by DR.
[Workaround]
Use DR instead of PCI hot plug if
need to add or remove PCI card on
the XSB.
6539084
6539909
O
O
O
O
Under Oracle Solaris 10 11/06, there is No workaround is available.
a low probability of a domain panic
during reboot when the PCIe Quad-
port Gigabit Ethernet UTP Low
Profile Adapter (X4447A-Z) is present
in a domain.
Under Oracle Solaris 10 11/06, do not Use an alternate type of adapter or
use the following adapters for
onboard network device to install the
network access when you are using Oracle Solaris OS via the network.
the boot net installcommand to
install the Oracle Solaris OS:
• PCIe Quad-port Gigabit Ethernet
UTP Low Profile Adapter (4447A-
Z/
X4447A-Z)
• PCIe Dual 10 Gigabit Ethernet Fiber
XFP Low Profile Adapter (1027A-
Z/
X1027A-Z)
6542632
O
O
Memory leak in PCIe module if driver For Oracle Solaris 10 11/06, this has
attach fails.
been fixed in patch 120011-09.
No workaround is available.
86
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Software Documentation Updates
This section contains late-breaking software information that became known after
the documentation set was published and corrections in the
M3000/M4000/M5000/M8000/M9000 servers software documentation.
The corrections for SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers
XSCF Reference Manual, if not otherwise specified, also apply to the man pages which
XSCF provides. And they supersede the information on the man pages.
Chapter 3 Information About Software
87
TABLE 3-15 lists known documentation updates.
TABLE 3-15 Software Documentation Updates
Title
Page Number
Update
SPARC Enterprise
44
The description, "The HTTPS service starts immediately after
being enabled.", in step 3 of section "To Enable or Disable the
Service Processor HTTPS Service" will be changed as follows:
M3000/M4000/M5000/M
8000/M9000 Servers
Administration Guide
With disable, the HTTPS service stops immediately. With enable,
the HTTPS service starts after the XSCF is reset by execution of
the rebootxscf(8) command.
73
The description, "If you create a second IM volume before the
first IM volume completes its synchronization, the first volume's
RAID status will indicate SYNC, and the second volume's RAID
status will indicate OPTIMAL." in Section 5.2.1, "Creating a
Hardware Mirrored Volume" will be changed as follows:
If you create a second IM volume before the first IM volume
completes its synchronization, the first volume's RAID status
will indicate SYNC, and the second volume's RAID status will
indicate DEGRADED
SPARC Enterprise
2-17
The host name can include "." (period).
M3000/M4000/M5000/M
8000/M9000 Servers
XSCF User's Guide
The description, "A label element can contain alphanumeric
characters (a to z, A to Z, 0 to 9), "-" (hyphen) and "." (period).
Each label element must always begin with an alphabetic
character and end with an alphanumeric character. However,
you cannot use a "." (period) in a host name." in "Host
name/domain name" of TABLE 2-2 Network Configuration will
be changed as follows:
A label element can contain alphanumeric characters (a to z, A to
Z, 0 to 9), "-" (hyphen) and "." (period). Each label element must
always begin with an alphabetic character and end with an
alphanumeric character.
2-28
"Configuring XSCF Routing" in Section 2.2.1 Network
Configuration will be changed. See “Configuring XSCF Routing”
on page 94 for details.
88
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-15 Software Documentation Updates (Continued)
Title
Page Number
Update
SPARC Enterprise
2-33
In the step 2 of "Configuring IP Packet Filtering Rules for XSCF
Network",
M3000/M4000/M5000/M
8000/M9000 Servers
XSCF User's Guide
The description,
<Example 1> Permit the IP address 192.168.100.0/255.255.255.0
to go through.
XSCF> setpacketfilters -y -c add -i xscf#0-lan#0
-s 192.168.100.0/255.255.255.0
is updated as follows:
<Example 1> Permit the IP address 192.168.100.0/255.255.255.0
to go through.
XSCF> setpacketfilters -y -c add -i xscf#0-lan#0
-s 192.168.100.0/255.255.255.0 -j ACCEPT
2-43
The description,
"<Example> Specify 3 for the retry count, an eight-character
password containing at least two digits, 60 days for the
expiration period, and 15 days for the advance notice of
expiration.
XSCF>setpasswordpolicy -y 3 -m 8 -d 2 -u 0 -l 0
-o 0 -M 60 -w 15",
is updated as follows:
<Example> Specify 3 for the retry count, 8 for the minimum
password length, 2 for the maximum credit by digits contained
in a password, 60 days for the period of validity, and 15 days for
the advance notice of validity expiration warning.
XSCF> setpasswordpolicy -y 3 -m 8 -d 2 -u 0 -l 0
-o 0 -M 60 -w 15
In this case, since the maximum credit by digits contained in a
password is set to 2, the number of numeric characters used in
the password (2 at maximum) is subtracted from the minimum
password length (8 characters); and the actual minimum
password length is 6 to 8 characters.
However, regardless of the setting of credit, the actual minimum
password length will never be less than 6.
For details, see TABLE 2-4 User Account Administration, and the
setpasswordpolicy(8) man page or the XSCF Reference
Manual.
Chapter 3 Information About Software
89
TABLE 3-15 Software Documentation Updates (Continued)
Title
Page Number
Update
SPARC Enterprise
2-177
You cannot set the memory mirror mode for a PSB in
M8000/M9000 servers.
M3000/M4000/M5000/M
8000/M9000 Servers
XSCF User's Guide
The PSB#02 data is included erroneously for the procedures
example of M4000/M5000 servers in "Setting the Memory Mirror
Mode for a PSB" of Section 2.2.14 System Board Configuration.
The example will be changed as follows:
3. Use the showfru(8) command to confirm the setting of
memory mirror mode.
XSCF>showfru -a sb
Device Location XSB Mode Memory Mirror Mode
sb
sb
00
01
Quad
Uni
Yes
No
3-4
The description, "When you return to XSCF Shell console
without logging out from the domain, the return causes
automatically logging out from the domain. At this time, a
background program is forced to quit. To avoid the background
program force-quit, return to XSCF Shell console with logging
out from the domain. " in "Domain console (RW console)" of
TABLE 3-1 Types of Terminals Connected With XSCF will be
changed as follows:
If you return to the XSCF Shell from the domain console, or if
you terminate the XSCF Shell, both without logging out of the
domain, you will be automatically logged out from the domain.
At the same time, a termination signal might be sent to any
program that is running in the background on the domain
console.
90
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-15 Software Documentation Updates (Continued)
Title
Page Number
Update
SPARC Enterprise
8-12
8.1.10 Firmware Update Procedure, "Updating XCP From the
Network",
M3000/M4000/M5000/M
8000/M9000 Servers
XSCF User's Guide
In - Command operation, step 2 and - Web browser operation,
step4, the following sentence into the top of each step are added,
- Command operation
2. Import XCP
Perform next step 3, "Perform the firmware update" immediately,
after the XCP import ends.
- Web browser operation
4. Import XCP.
Perform next step 5, "Perform the firmware update"
immediately, after the XCP import ends.
And - Web browser operation, step5 is modified as follows,
5. If complete message is displayed, the XCP importing has
ended. Perform the firmware update.
to
4 c) If complete message is displayed, the XCP importing has
ended. Confirm the imported version.
5. Perform the firmware update.
SPARC Enterprise
console(8)
command
M3000/M4000/M5000/
M8000/M9000 Servers
XSCF Reference Manual
If you return to the XSCF shell from the domain console, or if
you terminate the XSCF shell, both without logging out of the
domain, you will be automatically logged out from the domain.
At the same time, a termination signal might be sent to any
program that is running in the background on the domain
console.
forcerebootxs forcerebootxscf(8) command will be added. For detail, see
cf(8) command “forcerebootxscf(8) Command” on page 96.
Chapter 3 Information About Software
91
TABLE 3-15 Software Documentation Updates (Continued)
Title
Page Number
Update
SPARC Enterprise
password(8)
The following limitations are added:
M3000/M4000/M5000/
M8000/M9000 Servers
XSCF Reference Manual
command
When invoked without a user operand, password operates on
the current user account.
When invoked without options, password prompts you to
change the account password. The following restrictions apply:
- Password must be six or more characters.
- Password must be composed of five or more unique characters.
- Passwords must not be composed of only space characters.
- Passwords cannot be five or more consecutive characters such
as "012345" or "zyxwvu".
- Passwords cannot be specified with symmetric strings such as
"qazwswzaq" or "qazwsswzaq".
- Passwords cannot match any National Insurance number,
represented as "XXzzzzzzX" (X: capital letter, z: numeric
character), that is used in the United Kingdom.
- Passwords cannot match a word or term such as a proper name
that can be found in dictionaries.
- Passwords cannot contain the user name at the beginning or
ending.
- Passwords must not match the currently configured password.
- Passwords cannot be specified with only slight change, for
example, changing lower case letters to upper case, or upper
case letters to lower case.
sethostname(8) The host name can include "." (period). The description,
command
"Alphanumeric character and "-" can be used." in hostname
operand will be changed as follows:
Alphanumeric character, "-", and "." can be used.
sethttps(8)
You can now specify the -boption with the -cgenserverkey
command
and -cselfsignoptions.
The synopsis is as follows:
:sethttps [ [-q] -{y|n}] -cgenserverkey[-bbits]
:sethttps [ [-q] -{y|n}] -cselfsigncountry state | province
locality organization organizationalunit common e-mail [-bbits]
The option is as follows:
-bbits
It specifies the size (bit length) of the web server private key to
be created. You can specify either 1024 or 2048 for bits. If no
value is specified, a 2048-bit web server private key is created.
92
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 3-15 Software Documentation Updates (Continued)
Title
Page Number
Update
setnameserver For the domainname operand, description is updated as follows:
(8) command
Specifies the domain name of the DNS search path to be
registered or deleted. You can specify up to five domain names,
each separated by a space. domainname is specified in up to 254
characters with label elements delimited by a "." (period). If a
domain name exceeding 254 characters is specified, an error
occurs. The label element of the domain name can contain letters
(a to z, A to Z), numbers (0 to 9), and the special characters "-"
(hyphens) and "." (period). The domain name must begin with a
letter and end with either a letter or number. A "." (period) can
be used as delimiter.
SPARC Enterprise
setntp(8),
For the address operand, a part of description is updated as
follows:
M3000/M4000/M5000/
M8000/M9000 Servers
XSCF Reference Manual
showntp(8)
command
A host name is specified in up to 64 characters with label
elements delimited by a "." (period). The label element of the
host name can contain letters (a to z, A to Z), numbers (0 to 9),
and the special characters "-" (hyphens) and "." (period). The host
name must begin with a letter and end with either a letter or
number. A "." (period) can be used as delimiter. Host name, if
specified, must be resolvable.
setpasswordpo The following description for the initial setting of -mminlen will
licy(8)
be added:
command
The initial setting is 9.
setupfru(8)
The following description also applies to M4000/M5000 servers:
command
Although a CMU with two CPUs can be configured into Quad-
XSB mode on an M8000/M9000 server, the server generates a
"configuration error" message for those XSBs that do not have a
CPU and memory.
showhardconf In DESCRIPTION, "External I/O Expansion Unit information"
and "PCI card information" are added with the following note:
(8) command
(displayed only while the domain power is on)
Chapter 3 Information About Software
93
Updates of the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers
XSCF User’s Guide
The following information supersedes the information in the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
Configuring XSCF Routing
In a redundant XSCF unit configuration, the following are examples of data when
routing is done in each subnet.
<Example 1> This example shows one default gateway.
XSCF Unit 0
XSCF Unit 1
xscf#0-lan#0 [192.168.11.10]
+------------------------------+
XSCF-LAN#0
xscf#1-lan#0 [192.168.11.20]
XSCF-LAN#0
XSCF Unit 0
XSCF Unit 1
xscf#0-lan#1 [10.12.108.10]
+------------------------------+
XSCF-LAN#1
xscf#1-lan#1 [10.12.108.20]
XSCF-LAN#1
Destination
Gateway
Netmask
Interface
[192.168.11.0] -
[192.168.11.0] -
[10.12.108.0] -
[255.255.255.0] xscf#0-lan#0
[255.255.255.0] xscf#1-lan#0
[255.255.255.0] xscf#0-lan#1
[default]
[10.12.108.1]
[0.0.0.0]
xscf#0-lan#1
[10.12.108.0] -
[255.255.255.0] xscf#1-lan#1
[0.0.0.0] xscf#1-lan#1
[default]
[10.12.108.1]
Note – The method of determining the routing for an XSCF interface depends on
the network environment at the installation site. The network environment for
system operation must have a suitable configuration.
Note – You cannot set the routing to the takeover IP address.
94
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
<Example 2> This example shows two default gateways.
XSCF Unit 0
XSCF Unit 1
xscf#0-lan#0 [192.168.11.10]
+------------------------------+
XSCF-LAN#0
xscf#1-lan#0 [192.168.11.20]
XSCF-LAN#0
XSCF Unit 0
XSCF Unit 1
xscf#0-lan#1 [10.12.108.10]
+------------------------------+
XSCF-LAN#1
xscf#1-lan#1 [10.12.108.20]
XSCF-LAN#1
Destination
Gateway
Netmask
Interface
[192.168.11.0] -
[default]
[255.255.255.0] xscf#0-lan#0
[192.168.11.1] [0.0.0.0] xscf#0-lan#0
[192.168.11.0] -
[default]
[255.255.255.0] xscf#1-lan#0
[192.168.11.1] [0.0.0.0]
xscf#1-lan#0
[10.12.108.0] -
[255.255.255.0] xscf#0-lan#1
[default]
[10.12.108.1]
[0.0.0.0]
xscf#0-lan#1
[10.12.108.0] -
[255.255.255.0] xscf#1-lan#1
[0.0.0.0] xscf#1-lan#1
[default]
[10.12.108.1]
Note – When using two default gateways, you must configure the network to
recognize both gateways as default gateways. However, either one of two default
gateways is selected automatically after this configuration is done.
❑ Command operation
1. Use the showroute(8) command to display the routing environment.
XSCF> showroute -a
Destination
Gateway
Netmask
Flags Interface
192.168.11.0 *
255.255.255.0 U
255.255.255.0 U
xscf#0-lan#0
xscf#0-lan#1
xscf#0-lan#1
10.12.108.0
default
*
10.12.108.1
0.0.0.0
UG
Destination
Gateway
Netmask
Interface
192.168.11.0 *
255.255.255.0 xscf#1-lan#0
255.255.255.0 xscf#1-lan#1
10.12.108.0
default
*
10.12.108.1
0.0.0.0
xscf#1-lan#1
Chapter 3 Information About Software
95
2. Use the setroute(8) command to specify the routing environment for a network
interface.
<Example 1> Add routing with Destination 192.168.11.0 and Netmask
255.255.255.0 to XSCF-LAN#0 in the XSCFU#0.
XSCF> setroute -c add -n 192.168.11.0 -m 255.255.255.0 xscf#0-lan#0
<Example 2> Add routing with the default Gateway 10.12.108.1 to
XSCF-LAN#1 in the XSCFU#0.
XSCF> setroute -c add -n 0.0.0.0 -g 10.12.108.1 xscf#0-lan#1
<Example 3> Delete routing with destination 192.168.11.0 and
netmask 255.255.255.0 to XSCF-LAN#0 in the XSCFU#0.
XSCF> setroute -c del -n 192.168.11.0 -m 255.255.255.0 xscf#0-lan#0
Updates of the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers
XSCF Reference Manual
The following information supersedes the information in the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual.
forcerebootxscf(8) Command
NAME
forcerebootxscf - resets XSCF forcibly
SYNOPSIS
forcerebootxscf
forcerebootxscf -h
DESCRIPTION
The forcerebootxscf(8) command resets the XSCF forcibly.
The command registers an error log and resets XSCF forcibly when the
rebootxscf(8) command cannot respond for a firmware error after
logging in XSCF.
When the forcerebootxscf(8) command is executed in the system
with the dual XSCF configuration, only the XSCF, on which the
command has been executed, will be reset. When the active XSCF is reset,
the other XSCF becomes active.
Privileges
You must have platadmor fieldengprivilege to run this command.
Refer to setprivileges(8) for more information.
OPTIONS
The following option is supported:
-h Displays usage statement.
96
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
EXTENDED
DESCRIPTION
• When the command is executed, a prompt to confirm execution of the
command is displayed. Enter "y" to execute the command or "n" to
cancel the command.
• When the command is executed, an error log indicating a firmware
problem is registered.
Be sure to execute this command only when the rebootxscf(8)
command does not respond and the system cannot be recovered from
the detected firmware problem.
• The forcerebootxscf(8) command terminates all connection to
XSCF, such as telnet or ssh.
• When you cancel the XSCF reset using the setdate(8) which
commands reset XSCF automatically, even if you perform this
command, the information that is set is not applied in XSCF.
• By using the forcerebootxscf(8) command, you can reset XSCF
while the domain is in operation. However, if you set up XSCF to serve
as the upper NTP server of the domain, time lag may arise between
XSCF and the domain. In such a case, adjust the domain time so that
XSCF and the domain will be synchronized.
• When the rebootxscf(8) command could not be executed and the
forcerebootxscf(8) command is executed to reset XSCF, after
setting the XSCF by executing the applynetwork(8),
setaltitude(8), setdualpowerfeed(8), sethttps(8), setntp(8),
setssh(8), or settelnet(8) command, there may be a case where this
setting cannot be reflected properly. So be sure to execute the
applynetwork(8), setaltitude(8), setdualpowerfeed(8),
sethttps(8), setntp(8), setssh(8), or settelnet(8)command, then
reflect the settings by executing the rebootxscf(8) command again.
EXAMPLES
EXAMPLE 1 Resets the XSCF forcibly.
XSCF> forcerebootxscf
Please take snapshot after reboot, when the XSCF is
ready. [y|n] :y
The XSCF will be forcibly reset. Continue? [y|n]:y
EXIT STATUS
SEE ALSO
The following exit values are returned:
0
Successful completion.
>0 An error occurred.
rebootxscf(8)
Chapter 3 Information About Software
97
Miscellaneous Information
Identifying Degraded Memory in a System
1. Log in to XSCF.
2. Type the following command:
XSCF> showstatus
The following example identifies that DIMM number 00A on CMU#3 has
degraded memory.
XSCF> showstatus
CMU#3 Status:Normal;
*
MEM#00A Status:Degraded;
Identifying Different Memory Sizes in a System
Board
To identify if the domain contains system boards with different memory sizes, you
can use either of the following commands to display the list of memory sizes:
■ showdevices(8) command on the XSCF
■ prtdiag(1M) command on the domain
▼ Using the showdevicesCommand
1. Log in to XSCF.
98
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
2. Type the following command:
XSCF> showdevices -d domain_id
The following example displays 00-0 has 64GB of memory while the other system
boards have 16GB.
XSCF> showdevices -d 1
Memory:
-------
board
perm
base
domain target deleted remaining
mem MB XSB mem MB mem MB
0 0x0000004000000000 260288
7384 0x0000034000000000 260288
DID XSB mem MB mem MB address
01 00-0 63680
01 03-0 16384
01 03-1 16384
01 03-2 16384
01 03-3 16384
0 0x0000030000000000 260288
0 0x000002c000000000 260288
0 0x0000028000000000 260288
▼ Using the prtdiagCommand to Identify Memory Size
●
On the domain, execute the prtdiagcommand.
# prtdiag
The following example displays different memory sizes.
# prtdiag
============================ Memory Configuration ============================
Memory Available
LSB Group Size
Memory
Status
DIMM
Size
# of Mirror Interleave
DIMMs Mode Factor
--- ------ ------------------ ------- ------ ----- ------- ----------
00
00
03
03
04
04
05
05
06
A
B
A
B
A
B
A
B
A
32768MB
32768MB
8192MB
8192MB
8192MB
8192MB
8192MB
8192MB
8192MB
okay
okay
okay
okay
okay
okay
okay
okay
okay
2048MB
2048MB
2048MB
2048MB
2048MB
2048MB
2048MB
2048MB
2048MB
16 no
16 no
4 no
4 no
4 no
4 no
4 no
4 no
4 no
8-way
8-way
2-way
2-way
2-way
2-way
2-way
2-way
2-way
Chapter 3 Information About Software
99
Identifying Permanent Memory in a Target Board
1. Log in to XSCF.
2. Execute the following command:
XSCF> showdevices -d domain_id
The following example shows a display of the showdevices -dcommand
where 0 is the domain_id.
XSCF> showdevices -d 0
...
Memory:
-------
board
perm
base
domain target deleted remaining
mem MB XSB mem MB mem MB
0 0x0000000000000000 24576
DID XSB mem MB mem MB address
00 00-0
00 00-2
00 00-3
8192
8192
8192
1674 0x000003c000000000 24576
0 0x0000034000000000 24576
...
The entry for column 4 perm mem MB indicates the presence of permanent
memory if the value is non-zero.
The example shows permanent memory on 00-2, with 1674 MB.
If the board includes permanent memory, when you execute the deleteboard(8)
command or the moveboard(8) command, the following notice appears:
System may be temporarily suspended, proceed? [y|n]:
100
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
CHAPTER
4
Information About Hardware of
This section describes the special instructions and the issues about the M3000 server
hardware.
■
■
■
“Notes and Limitations” on page 101
“Hardware RAID Function” on page 105
“Hardware Documentation Updates” on page 105
Notes and Limitations
Notes on Hardware RAID
Notes on Using the Hardware RAID
■
Make backup copies of important data and programs periodically. In some
failures, rebuild of hardware RAID and the data restore from the backup media
might become necessary.
■
■
To ensure data integrity in time of power failure, it is recommended to use an
uninterruptible power supply (UPS).
If you require higher availability such as the controller or the data path
redundancy, use the dedicated RAID system.
101
Notes on Building or Deleting the Hardware RAID
■
The data in the disk drive is not secured during the build or the deletion of a
hardware RAID. When you build a new hardware RAID while the system is
running, or delete the hardware RAID, ensure to make backup copies of the data.
After building a hardware RAID, you need to install data or restore data from the
backup media.
■
With no load, the hardware RAID build or the synchronization associated with
maintenance takes about 180 minutes for a 300 GB disk drive. It takes about 360
minutes for a 600 GB disk drive.
■
■
If there is a system restart during the hardware RAID build or synchronization,
the build/synchronization must be started again from the beginning.
When the hardware RAID applied, the volume size is smaller than the size of
original disk.
Notes on Operating the Hardware RAID
In some cases, RAID controller cannot conclude the disk drive is completely in
failure, and system slowdown might occur. To do maintenance in this status, follow
the procedures below.
1. Delete the hardware RAID.
2. Replace all the member disks related to the hardware RAID.
3. Rebuild the hardware RAID.
4. Restore the data from the backup media.
Notes on raidctl(1M) Command
The following options of the raidctl(1M) command can not be used because the
hardware specifications do not support them.
■ raidctl-C-r1E(To create a RAID 1E)
■ raidctl-c-r1E(To create a RAID 1E)
■ raidctl-a(To assign hot-spare disks)
■ raidctl-p(To set the cache)
■ raidctl-C-z(To specify the volume capacity)
102
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Notes on DVD Drive and Discs
See the "Notes on DVD Drives and Discs in SPARC Enterprise" on the website below
before using the CD/DVD discs in the standard DVD drive mounted in this server.
Notes on the Use of USB Memory
To execute the dumpconfig(8), restoreconfig(8) or snapshot(8) command, if
you specify USB memory as the destination to store data, prepare the USB memory
as a medium in advance.
The data stored will include the information about the system. To use USB memory,
you need to pay attention to the management of the USB memory in which the data
stored, from the data security viewpoint.
We do not provide guarantees to every USB memory from any manufacturers that
currently on the market against its connectivity to XSCF and proper operation.
Depending on the USB memory in use, defects such as the XSCF firmware error or
reset may occur. In case such defects occurred, stop the use of USB memory
immediately.
directly to the USB port. If connected via USB hub or USB extension cables, it may
cause errors.
Power Control and Operator Panel Mode Switch
Note – Before using the RCI function, see “Notes on RCI Function” on page 19.
When you use the remote power control utilizing the RCI function or the automatic
power control system (APCS), you can disable this remote power control or the
APCS by setting the mode switch to Service mode on the operator panel.
Disabling these features ensures that you do not unintentionally switch the system
power on or off during maintenance. Note that system power off with the APCS
cannot be disabled with the mode switch. Therefore, be sure to turn off automatic
power control via APCS before starting maintenance.
If you switch the mode while using the RCI or the automatic power control, the
system power is controlled as follows.
Chapter 4 Information About Hardware of M3000 Server
103
TABLE 4-1
Function
RCI and APCS Control with Mode Switch
Mode Switch
Locked
Service
RCI
Remote power-on/power-off
operations are enabled.
Remote power-on/power-off
operations are disabled.
APCS
Automatic power-on/power-off
operations are enabled.
Automatic power-on is disabled, but
power-off remains enabled.
To use the RCI function, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers RCI Build Procedure and the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI User’s Guide.
To use the APCS, see the Enhanced Support Facility User’s Guide for Machine
Administration Automatic Power Control Function (Supplement Edition).
Limitation for 1027A-Z/X1027A-Z
For 1027A-Z/X1027A-Z, PCIe Dual 10 Gigabit Ethernet Fiber XFP cards, this
maximum limit apply:
■
No more than one card in an M3000 server.
Limitation for 4447A-Z/X4447A-Z
For PCIe Quad-port Gigabit Ethernet UTP Low Profile Adapter (4447A-Z/
X4447A-Z), this maximum limit apply:
■
No more than four cards in an M3000 server.
Patches for Emulex PCI Express (PCIe) Adapters
The following Emulex PCIe adapters require drivers supplied in patch 120222-26:
■
Sun StorageTek Enterprise Class 4Gb Dual-Port Fibre Channel PCIe HBA
(XSEFC402AF)
■
Sun StorageTek Enterprise Class 4Gb Single-Port Fibre Channel PCIe HBA
(XSEFC401AF)
104
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Hardware RAID Function
RAID technology enables the construction of a logical volume made up of several
physical disks to provide data redundancy, improved performance, or both.
The on-board SAS controller of the M3000 server which consists of the SPARC64
VII+ processors supports RAID 1 (mirroring) volumes using the Oracle Solaris OS
raidctlutility.
Requirements for Hardware RAID Support
The table below shows the requirements for the hardware RAID support.
TABLE 4-2
Item
Requirements for Hardware RAID Support
Description
Required Oracle Solaris OS
patches
At this time, no patches are required.
3.2 or later
Required ESF (Enhanced
Support Facility) patches
XSCF firmware
RAID level
XCP 1101 or later
RAID 1 (IM: Integrated Mirror) only.
Two member disks can be configured per RAID volume.
Disk drive
The member disks in a single RAID volume shall be made
up of disk drives of the same capacity and of the same
rotational speed.
Solid State Drive (SSD) is not supported.
Number of RAID volumes
Up to two RAID volumes on a server.
For notes on hardware RAID, see “Notes on Hardware RAID” on page 101.
Hardware Documentation Updates
TABLE 4-3 lists known documentation updates.
Chapter 4 Information About Hardware of M3000 Server
105
TABLE 4-3
Title
Hardware Documentation Updates
Page Number
Update
SPARC Enterprise M3000
Server Service Manual
1-3
The following caution will be added in "1.2.1 Electrical Safety
Precautions":
Caution - Before doing the maintenance, unplug the power
cords. This product uses double pole/neutral fusing which could
create an electric shock hazard.
6-8
B-2
FIGURE 6-5 "Removing the Motherboard Unit" in Section 6.2,
"Removing the Motherboard Unit will be changed. For detail, see
“Removing the Motherboard Unit” on page 107.
The following footnote will be added to Table B-1 "FRU List" in
Section B.1, "Server Overview":
Do not remove or swap non-FRU components such as MEMR
and DDCs as this interferes with the repair depot’s ability to
diagnose and repair field returns.
106
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Updates of SPARC Enterprise M3000 Server Service
Manual
The following information supersedes the information in the SPARC Enterprise
M3000 Server Service Manual.
Removing the Motherboard Unit
2. Hold the two opaque handles on both sides of the motherboard unit and lift
the motherboard unit while sliding it out.
FIGURE 6-5 Removing the Motherboard Unit
Handles
Chapter 4 Information About Hardware of M3000 Server
107
108
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
CHAPTER
5
Information About Hardware of
This section describes the special instructions and the issues about the
M4000/M5000 servers hardware.
■
■
■
“Notes and Limitations” on page 109
“Hardware Issues and Workarounds” on page 112
“Hardware Documentation Updates” on page 112
Notes and Limitations
Notes on Server Installation
The plug (NEMA L6-20) of the power cables supplied with the M4000/M5000
servers has the following dimensions.
Confirm in advance that the plug does not become an obstruction when plugged
into the input power source at the installation site.
■
■
■
Outside diameter of the connector part of the plug: 50 mm
Maximum outside diameter of the plug cover: 57 mm
Plug cover length from the connector end: 69.9 mm
109
Notes on DVD Drive and Discs
See the "Notes on DVD Drives and Discs in SPARC Enterprise" on the website below
before using the CD/DVD discs in the standard DVD drive mounted in this server.
Notes on the Use of USB Memory
To execute the dumpconfig(8), restoreconfig(8) or snapshot(8) command, if
you specify USB memory as the destination to store data, prepare the USB memory
as a medium in advance.
The data stored will include the information about the system. To use USB memory,
you need to pay attention to the management of the USB memory in which the data
stored, from the data security viewpoint.
We do not provide guarantees to every USB memory from any manufacturers that
currently on the market against its connectivity to XSCF and proper operation.
Depending on the USB memory in use, defects such as the XSCF firmware error or
reset may occur. In case such defects occurred, stop the use of USB memory
immediately.
directly to the USB port. If connected via USB hub or USB extension cables, it may
cause errors.
Power Control and Operator Panel Mode Switch
Note – Before using the RCI function, see “Notes on RCI Function” on page 19.
When you use the remote power control utilizing the RCI function or the automatic
power control system (APCS), you can disable this remote power control or the
APCS by setting the mode switch to Service mode on the operator panel.
Disabling these features ensures that you do not unintentionally switch the system
power on or off during maintenance. Note that system power off with the APCS
cannot be disabled with the mode switch. Therefore, be sure to turn off automatic
power control via APCS before starting maintenance.
If you switch the mode while using the RCI or the automatic power control, the
system power is controlled as follows.
110
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 5-1
Function
RCI and APCS Control with Mode Switch
Mode Switch
Locked
Service
RCI
Remote power-on/power-off
operations are enabled.
Remote power-on/power-off
operations are disabled.
APCS
Automatic power-on/power-off
operations are enabled.
Automatic power-on is disabled, but
power-off remains enabled.
To use the RCI function, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers RCI Build Procedure and the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI User’s Guide.
To use the APCS, see the Enhanced Support Facility User's Guide for Machine
Administration Automatic Power Control Function (Supplement Edition).
Notes on Air Filter
Fujitsu does not handle the air filter which is shown in the SPARC Enterprise
M4000/M5000 Servers Service Manual as a product.
Limitation for 1027A-Z/X1027A-Z
For PCIe Dual 10 Gigabit Ethernet Fiber XFP Low Profile Adapter (1027A-Z/
X1027A-Z), these maximum limits apply:
■
■
■
No more than two cards per a domain in an M4000 server.
No more than four cards per a domain in an M5000 server.
Do not use these cards in an External I/O Expansion Unit.
Limitation for 4447A-Z/X4447A-Z
For PCIe Quad-port Gigabit Ethernet UTP Low Profile Adapter (4447A-Z/
X4447A-Z), these maximum limits apply:
■
No more than four cards per a domain in an M4000 server.
No more than eight cards per a domain in an M5000 server.
■
Chapter 5 Information About Hardware of M4000/M5000 Servers
111
■
No more than four cards in an External I/O Expansion Unit (two per PCIe I/O
boat).
Patches for Emulex PCI Express (PCIe) Adapters
The following Emulex PCIe adapters require drivers supplied in patch 120222-26:
■
Sun StorageTek Enterprise Class 4Gb Dual-Port Fibre Channel PCIe HBA
(XSEFC402AF)
■
Sun StorageTek Enterprise Class 4Gb Single-Port Fibre Channel PCIe HBA
(XSEFC401AF)
Hardware Issues and Workarounds
Sun Crypto Accelerator 6000
If you are not using the correct version of the Sun Crypto Accelerator (SCA) 6000
driver, hot-plug operations on SCA 6000 can cause the M4000/M5000 servers to
panic or hang. Version 1.1 of the SCA6000 driver and firmware supports hot-plug
operations after the required bootstrap firmware upgrade has been performed.
Version 1.0 of the SCA6000 driver does not support hot-plug and should not be used.
Hardware Documentation Updates
This section contains late-breaking hardware information that became known after
the documentation set was published and corrections in the M4000/M5000 servers
hardware documentation.
TABLE 5-2 lists known documentation updates.
112
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 5-2
Title
Hardware Documentation Updates
Page Number
Update
SPARC Enterprise
M4000/M5000 Servers
Installation Guide
2-8
"B-type plug" in the Note in "2.2.2 Cable Connections" should be
substituted with "the plug with lock function."
SPARC Enterprise
M4000/M5000 Servers
Service Manual
B-2
The description for CPU Module in Table B-1 will add the
following:
At least one CPU Module (CPUM) is required on each eXtended
System Board (XSB). The XSB includes the CPU, memory, and
optional I/O. The XSB can be set to one of two logical
configurations: uni-XSB or quad-XSB mode. In the uni-XSB
mode, the M4000 server has one configurable XSB and the M5000
server has two separate configurable XSBs. For more information
on the rules for system configuration, refer to the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers
Administration Guide.
Chapter 5 Information About Hardware of M4000/M5000 Servers
113
114
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
CHAPTER
6
Information About Hardware of
This section describes the special instructions and the issues about the
M8000/M9000 servers hardware.
■
■
■
■
■
“Notes and Limitations” on page 115
“Hardware Issues and Workarounds” on page 119
“Hardware RAID Function” on page 120
“Hardware Documentation Updates” on page 121
“Miscellaneous Information” on page 127
Notes and Limitations
Notes on Using a RAID-capable IOUA
When using a RAID-capable IOUA on the M8000/M9000 servers, update XSCF
firmware to XCP 1110 or later.
115
Notes on Hardware RAID
Notes on Using the Hardware RAID
■
Make backup copies of important data and programs periodically. In some
failures, rebuild of hardware RAID and the data restore from the backup media
might become necessary.
■
■
To ensure data integrity in time of power failure, it is recommended to use an
uninterruptible power supply (UPS).
If you require higher availability such as the controller or the data path
redundancy, use the dedicated RAID system.
Notes on Building or Deleting the Hardware RAID
■
The data in the disk drive is not secured during the build or the deletion of a
hardware RAID. When you build a new hardware RAID while the system is
running, or delete the hardware RAID, ensure to make backup copies of the data.
After building a hardware RAID, you need to install data or restore data from the
backup media.
■
With no load, the hardware RAID build or the synchronization associated with
maintenance takes about 180 minutes for a 300 GB disk drive. It takes about 360
minutes for a 600 GB disk drive.
■
■
If there is a system restart during the hardware RAID build or synchronization,
the build/synchronization must be started again from the beginning.
When the hardware RAID applied, the volume size is smaller than the size of
original disk.
Notes on Operating the Hardware RAID
In some cases, RAID controller cannot conclude the disk drive is completely in
failure, and system slowdown might occur. To do maintenance in this status, follow
the procedures below.
1. Delete the hardware RAID.
2. Replace all the member disks related to the hardware RAID.
3. Rebuild the hardware RAID.
4. Restore the data from the backup media.
116
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Notes on raidctl(1M) Command
The following options of the raidctl(1M) command can not be used because the
hardware specifications do not support them.
■ raidctl -C -r 1E(To create a RAID 1E)
■ raidctl -c -r 1E(To create a RAID 1E)
■ raidctl -a(To assign hot-spare disks)
■ raidctl -p(To set the cache)
■ raidctl -C -z(To specify the volume capacity)
Notes on DVD Drive and Discs
See the "Notes on DVD Drives and Discs in SPARC Enterprise" on the website below
before using the CD/DVD discs in the standard DVD drive mounted in this server.
Notes on the Use of USB Memory
To execute the dumpconfig(8), restoreconfig(8) or snapshot(8) command, if
you specify USB memory as the destination to store data, prepare the USB memory
as a medium in advance.
The data stored will include the information about the system. To use USB memory,
you need to pay attention to the management of the USB memory in which the data
stored, from the data security viewpoint.
We do not provide guarantees to every USB memory from any manufacturers that
currently on the market against its connectivity to XSCF and proper operation.
Depending on the USB memory in use, defects such as the XSCF firmware error or
reset may occur. In case such defects occurred, stop the use of USB memory
immediately.
To connect the USB memory to the USB port for XSCF, connect the USB memory
directly to the USB port. If connected via USB hub or USB extension cables, it may
cause errors.
Chapter 6 Information About Hardware of M8000/M9000 Servers
117
Power Control and Operator Panel Mode Switch
Note – Before using the RCI function, see “Notes on RCI Function” on page 19.
When you use the remote power control utilizing the RCI function or the automatic
power control system (APCS), you can disable this remote power control or the
APCS by setting the mode switch to Service mode on the operator panel.
Disabling these features ensures that you do not unintentionally switch the system
power on or off during maintenance. Note that system power off with the APCS
cannot be disabled with the mode switch. Therefore, be sure to turn off automatic
power control via APCS before starting maintenance.
If you switch the mode while using the RCI or the automatic power control, the
system power is controlled as follows.
TABLE 6-1
Function
RCI and APCS Control with Mode Switch
Mode Switch
Locked
Service
RCI
Remote power-on/power-off
operations are enabled.
Remote power-on/power-off
operations are disabled.
APCS
Automatic power-on/power-off
operations are enabled.
Automatic power-on is disabled, but
power-off remains enabled.
To use the RCI function, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers RCI Build Procedure and the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI User’s Guide.
To use the APCS, see the Enhanced Support Facility User's Guide for Machine
Administration Automatic Power Control Function (Supplement Edition).
Limitation for 1027A-Z/X1027A-Z
For PCIe Dual 10 Gigabit Ethernet Fiber XFP Low Profile Adapter (1027A-
Z/X1027A-Z), these maximum limits apply:
■
No more than six cards per a domain.
■
Do not use these cards in an External I/O Expansion Unit.
118
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
Limitation for 4447A-Z/X4447A-Z
For PCIe Quad-port Gigabit Ethernet UTP Low Profile Adapter (4447A-Z/X4447A-
Z), these maximum limits apply:
■
No more than eight cards per a domain.
■
No more than four cards in an External I/O Expansion Unit (two per PCIe I/O
boat).
Patches for Emulex PCI Express (PCIe) Adapters
The following Emulex PCIe adapters require drivers supplied in patch 120222-26:
■
Sun StorageTek Enterprise Class 4Gb Dual-Port Fibre Channel PCIe HBA
(XSEFC402AF)
■
Sun StorageTek Enterprise Class 4Gb Single-Port Fibre Channel PCIe HBA
(XSEFC401AF)
Hardware Issues and Workarounds
DVD Drives and cfgadm
The Solaris cfgadm(1M) command does not always unconfigure a DVD drive from a
domain on the M8000/M9000 servers.
Disable the Volume Management Daemon (vold) before unconfiguring a DVD drive
with the cfgadm(1M) command. To disable vold, stop the daemon by issuing the
command /etc/init.d/volmgt stop. After the device has been removed or
inserted, restart the daemon by issuing the command /etc/init.d/volmgt
start.
Chapter 6 Information About Hardware of M8000/M9000 Servers
119
Sun Crypto Accelerator 6000
If you are not using the correct version of the Sun Crypto Accelerator (SCA) 6000
driver, hot-plug operations on SCA 6000 can cause the M8000/M9000 servers to
panic or hang. Version 1.1 of the SCA6000 driver and firmware supports hot-plug
operations after the required bootstrap firmware upgrade has been performed.
Version 1.0 of the SCA6000 driver does not support hot-plug and should not be used.
Hardware RAID Function
RAID technology enables the construction of a logical volume made up of several
physical disks to provide data redundancy, improved performance, or both.
The M8000/M9000 servers mounted with a RAID-capable IOUA support RAID 1
(mirroring) volumes using the Oracle Solaris OS raidctl utility.
Note – When the onboard device card is RAID-capable, the showhardconf(8)
command displays Type 2 in the output.
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T826 ; Type:2;
+ FRU-Part-Number:CA21138-B84X 010AE/371-5000-05 ;
Requirements for Hardware RAID Support
The table below shows the requirements for the hardware RAID support on the
M8000/M9000 servers.
TABLE 6-2
Item
Requirements for Hardware RAID Support
Description
*
Required Oracle Solaris
OS/packages/patches
Oracle Solaris 10 8/07
SUNWigb
-
138174-04
138174-04
Oracle Solaris 10 5/08 through Oracle
Solaris 10 5/09
Oracle Solaris 10 10/09 or later
Oracle Solaris 11 11/11
-
-
-
-
120
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
TABLE 6-2
Item
Requirements for Hardware RAID Support (Continued)
Description
ESF (Enhanced Support
3.0 or later
Oracle Solaris 10 914527-17
Facility)/Supported Oracle
Solaris OS/Required patches
4.0
Oracle Solaris 11 T006481SP-01
XSCF firmware
RAID level
XCP 1110 or later
• RAID 1 (IM: Integrated Mirror) only
• Two member disks can be configured per RAID volume
Disk drive
of the same capacity and of the same rotational speed
• Solid State Drive (SSD) is not supported
Number of RAID volumes
Up to one RAID volume per a RAID-capable IOUA
* To obtain SUNWigb package, contact your sales representative.
For notes on hardware RAID, see “Notes on Hardware RAID” on page 116.
Hardware Documentation Updates
This section contains late-breaking hardware information that became known after
the documentation set was published and corrections in the M8000/M9000 servers
hardware documentation.
TABLE 6-3 lists known documentation updates.
Chapter 6 Information About Hardware of M8000/M9000 Servers
121
TABLE 6-3
Title
Hardware Documentation Updates
Page Number
Update
SPARC Enterprise
M8000/M9000 Servers
Overview Guide
1-6
The description about Hardware RAID function will add the
following:
Two hard disks which are connected to the IOU onboard device
card (IOUA) can be constructed as a single logical volume. The
mirrored configuration of the constructed logical volume can
secure the data redundancy, as well as achieving the
improvement in system fault tolerance.
Note - Hardware RAID is supported only on the M8000/M9000
servers on which a RAID-capable onboard device card (IOUA) is
mounted. The RAID-capable IOUA requires a minimum XSCF
firmware with applicable Oracle Solaris OS patches. See the
latest Product Notes for this information.
SPARC Enterprise
M8000/M9000 Servers
Site Planning Guide
1-18
A part of values in inches on FIGURE 1-13 to FIGURE 1-16 will
be corrected. For detail, see “M9000 Servers Installation Area” on
page 123.
1-35
7-8
For the order number of Vibration-proof bracket and Vibration-
proof feet, see “Miscellaneous Information” on page 127.
SPARC Enterprise
M8000/M9000 Servers
Service Manual
The mounting location of I/O Unit (IOU) indicated on FIGURE
7-4 will be corrected. For detail, see “I/O Unit (IOU) Mounting
Locations” on page 126."
Updates of SPARC Enterprise M8000/M9000 Servers
Site Planning Guide
The following information supersedes the information in the SPARC Enterprise
M8000/M9000 Servers Site Planning Guide
122
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
FIGURE 1-14 M9000 Server (Base Cabinet) + Power Cabinet Installation Area
Service area
M9000
(Base cabinet)
317
Service area
(12.5)
837 (33.0)
Unit: mm (inch)
1154 (45.3)
Front
124
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
FIGURE 1-15 M9000 Server (Base Cabinet + Expansion Cabinet) Installation Area
Service area
M9000
M9000
(Ex. cabinet)
(Base cabinet)
Service area
837 (33.0)
Unit: mm (inch)
1674 (65.9)
Front
Chapter 6 Information About Hardware of M8000/M9000 Servers
125
FIGURE 1-16 M9000 Server (Base Cabinet + Expansion Cabinet) + Power Cabinet
Installation Are
Service area
M9000
(Base cabinet)
M9000
(Ex. cabinet)
317
(12.5)
Service area
824 (32.4)
Unit: mm (inch)
2282 (89.8)
Front
Updates of SPARC Enterprise M8000/M9000 Servers
Service Manual
The following information supersedes the information in the SPARC Enterprise
M8000/M9000 Servers Service Manual.
I/O Unit (IOU) Mounting Locations
FIGURE 7-4 shows the mounting locations of IOUs in the M9000 server (with the
expansion cabinet).
126
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
FIGURE 7-4 IOU Mounting Locations (Front of M9000 with the Expansion Cabinet)
IOU#0
IOU#2
IOU#4
IOU#6
IOU#8
IOU#10
IOU#12
IOU#14
Miscellaneous Information
Cabinet Stabilization Measures
This section shows useful methods of securing an M8000/M9000 server to prevent it
from shifting due to vibration.
Chapter 6 Information About Hardware of M8000/M9000 Servers
127
Method of Securing the Cabinet to the Floor Surface
Vibration-proof Bracket
■
order number: CA06620-H901
By specifying the above, you can issue an order to get the following four kinds of
parts. (You should order the bolt separately.)
■
■
■
■
Fixing bracket: C230-4130-X514 (x4)
Bushing: C230-4080-X003 (x4)
Metal washer: F6-WB12-121 (x4)
Metal washer: F6-WM12-121 (x4)
■
Bolt: F6-B12-***121 (Bolt diameter: M12)
***= Bolt length (mm)
The bolt length must be specified here.
Note – Consult with the company that constructed the building because the length
depends on the structure of the building.
Vibration-proof Feet
■
order number: CA06620-H902
By specifying the above, you can issue an order to get the following parts. (You
should order the bolt separately.)
■
F6-DA6G (x4)
Change this board leg (height adjustment leg) to the "F6-DA6G" and tighten the bolt
from under the floor.
■
Bolt diameter: M20
Bolt length: 15 mm + floor thickness
Note – Consult with the company that constructed the building because the length
depends on the structure of the building.
128
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
FIGURE 6-1 Example of Securing the Cabinet to the Floor Surface Using the Fixing Bracket (M9000 Server)
Chapter 6 Information About Hardware of M8000/M9000 Servers
129
FIGURE 6-2 Example of Securing the Cabinet Through a Hole in the Floor
130
SPARC Enterprise Mx000 Servers Product Notes for XCP Version 1124 • August 2018
|