HP Hewlett Packard 100BASE TX User Manual

Usin g Du a l Por t 100Ba se-TX a n d Du a l  
Por t Wid e Ultr a 2 SCSI  
Ver sion B.11.00.01  
Ma n u fa ctu r in g Pa r t Nu m ber :  
E1099  
U. S. A.  
© Copyright 1999 Hewlett-Packard Company.  
Con ten ts  
Manual Speed and Duplex Mode Configuration . . . . . . . . . . . . . . . . . . .11  
Autonegotiation and Autosensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .14  
What Manuals are Available. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15  
Software Availability in Native Languages . . . . . . . . . . . . . . . . . . . . . . .16  
2. Con figu r in g Netw or k Con n ectivity Usin g SAM  
Step 1: Configuring Network Connectivity . . . . . . . . . . . . . . . . . . . . . . .19  
Step 2: Deleting a Default Gateway . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21  
3. SCSI Pa r a m eter a n d Ter m in a tion In for m a tion  
Configuring SCSI Parameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24  
4. Tr ou blesh ootin g SCSI  
Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28  
Error Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48  
Logging Messages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49  
Manual Installation and Configuration . . . . . . . . . . . . . . . . . . . . . . . . . .51  
Contacting Your HP Representative. . . . . . . . . . . . . . . . . . . . . . . . . . . . .52  
iii  
Con ten ts  
Flowchart 5: Configuration Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76  
Flowchart 8: Transport Level Loopback Test (using ARPA). . . . . . . . 84  
Flowchart 9: Bridge/Gateway Loopback Test . . . . . . . . . . . . . . . . . . . 86  
LAN Interface Status Display . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90  
RFC 1213 MIB II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92  
RFC 1284 Ethernet-Like Interface Statistics . . . . . . . . . . . . . . . . . . . . . 95  
Create a Record or Map of Your Internetwork . . . . . . . . . . . . . . . . . . . . 97  
Basic Troubleshooting Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100  
Connector Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101  
PCI 10/100Base-TX Card Twisted-Pair Connector . . . . . . . . . . . . . . 101  
Connector Pin Usage for 10-Mbit/s Twisted-Pair Connector . . . . . . 101  
Connector Pin Usage for 100-Mbit/s Twisted-Pair Connector . . . . . 102  
iv  
Con ten ts  
Cable Lengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104  
Dual Port 100Base-TX and Dual Port Wide  
Ultra 2 SCSI Specifications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105  
C. Ha r d w a r e Regu la tor y Sta tem en ts  
FCC Statement (For U.S.A.) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .108  
Canada . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .108  
EMI Statement (European Community) . . . . . . . . . . . . . . . . . . . . . . . .109  
Glossa r y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111  
v
Con ten ts  
vi  
P r efa ce  
The information in this manual is intended for network managers who  
install and administer 10/100Base-TX networks and Wide Ultra2 SCSI  
devices. It is assumed the reader is experienced with the basics of local  
and wide area networking and SCSI technology.  
This manual describes how to install, configure, and troubleshoot the 10/  
100Base-TX and SCSI software and hardware product on HP 9000 PCI  
systems.  
Chapter 1 , Installing and Configuring PCI 10/100 Base-TX,” describes  
how to install and configure 10/100Base-TX software and hardware. Also  
it includes regulatory statements for the USA, Japan, and the European  
community.  
Chapter 2 , Configuring Network Connectivity Using SAM,” describes  
the steps to configure remote connectivity automatically using the  
System Administration Manager (SAM).  
steps to configure remote connectivity automatically using the System  
Administration Manager (SAM).  
the SCSI ports on the card.  
useful tools for installing, configuring, and maintaining HP  
10/100Base-TX software.  
Chapter 6 , Troubleshooting 10/100Base-TX/9000,' provides flowcharts  
to help diagnose 10/100Base-TX.  
Appendix A , 10/100Base-TX Interface Card Statistics,” defines the  
terms listed in the lanadmin(1M) command display.  
Appendix B , Hardware Reference Information,” provides information  
about card specifications and cabling requirements.  
7
8
1
In sta llin g a n d Con figu r in g P CI  
10/100 Ba se-TX  
This document contains information on the PCI 10/100Base-TX/9000  
amd Wide Ultra2 SCSI cards for use with the 11.0 operating systems.  
9
 
Installing and Configuring PCI 10/100 Base-TX  
Compatibility and Installation Requirements  
Com p a tibility a n d In sta lla tion Requ ir em en ts  
Following are the limits of the current PCI 10/100Base-TX/9000:  
The PCI 10/100Base-TX/9000 ports support autonegotiation and  
autosensing. You should not normally need to manually configure the  
not support autonegotiation but is set to full-duplex mode, there may  
be a mismatch between the card and the switch, because the card  
defaults to half-duplex for switches that do not support  
autonegotiation. You can determine what the card is set to by using  
lanadmin -xand reset it if necessary using lanadmin -X. See  
Manual Speed and Duplex Mode Configuration” for details.  
If you manually set the speed and duplex mode of the Base-TX card,  
autonegotiation will be turned off.  
Both full- and half-duplex modes are supported. If your hub or switch  
does not support autonegotiation, ensure that your hub or switch is  
set to the desired duplex mode.  
The PCI 0/100Base-TX/9000 LAN software is for use with only the  
following protocols: TCP/IP, ARPA, and NFS.  
When using the ioscan -f command to verify installation, the last digit  
of the H/W Path (hardware path) will show the port number of the  
card. The other fields of the output will show the driver as btlan , and  
the Hardware Type Description will be PCI.  
10  
Chapter1  
 
Installing and Configuring PCI 10/100 Base-TX  
Manual Speed and Duplex Mode Configuration  
Ma n u a l Sp eed a n d Du p lex Mod e  
Con figu r a tion  
Because the PCI 10/100Base-TX/9000 LAN ports support  
autonegotiation, you should not normally need to manually set the  
duplex mode. Sometimes you may need to manually set the duplex mode  
of the cardfor example, if the switch is operating at full-duplex but  
does not autonegotiate. Because the card defaults to half-duplex when  
autonegotiation is turned off, this could cause a mismatch between the  
card and switch (at either 10 or100 Mbits/s). To fix this, use the  
lanadmin -Xcommand as described later in this section.  
The CSMA/CD media access method used in IEEE 802.3u-1995 is  
inherently a half-duplex mechanism. That is, at any one time, there can  
be only one sender of data on the link segment. It is not possible for  
devices on either end of the link segment to transmit simultaneously.  
Since Category 5 UTP contains multiple pairs of wires, it is possible to  
have devices on both ends of a link segment sending data to each other  
simultaneously. This is known as full-duplex operation. While the details  
of full-duplex operation are not currently defined by IEEE 802.3u-1995  
(full-duplex mode essentially involves turning off” the CSMA/CD access  
method which is the foundation of IEEE 802.3), the autonegotiation  
mechanism defined in IEEE 802.3u-1995 allows devices to advertise and  
configure themselves to operate in a full-duplex mode which is  
essentially vendor-specific. Devices that do not support autonegotiation  
can sometimes be manually configured to operate in full-duplex mode.  
Full-duplex mode is most commonly found in, and indeed only makes  
sense for, switches rather than hubs. It may be found in either 10 Mbit/s  
or 100 Mbit/s switch devices. Full-duplex mode may provide a  
throughput advantage under some circumstances, but the degree of the  
advantage is application-dependent.  
Ensure that the speed, duplex mode, and autonegotiation of the  
associated switch are configured the same as on the PCI 10/100Base-TX  
card. If the switch supports autonegotiation on the ports connected to the  
cards, this should be enabled as explained in Autonegotiation and  
Autosensing.  
Chapter 1  
11  
 
Installing and Configuring PCI 10/100 Base-TX  
Manual Speed and Duplex Mode Configuration  
To manually set the duplex mode of the PCI ports, first ensure that your  
computer has the latest applicable patches installed as listed in the  
Required and Optional Patches section of this release note.  
To list the current speed and duplex mode of the PCI  
10/100Base-TX/9000 ports, use the -xoption (NOTE: lowercase x) of the  
lanadmin command. Determine the speed and duplex mode of your hub  
or switch before performing manual configuration as follows:  
lanadmin -x ppa (HP-UX 11.0)  
To manually set the duplex mode of the interface, install one of the  
patches above and then use the -Xoption of lanadminas follows:  
lanadmin -X mode ppa(on HP-UX 11.0)  
where:  
modecan be any one of the following strings (and the fdor hdare  
case-insensitive):  
10fd =10 full-duplex  
10hd=10 half-duplex  
100fd =100 full-duplex  
100hd=100 half-duplex  
and  
lanadmin -X auto_on ppa(turns autonegotiation on for HP-UX 11.0)  
The ppais the physical point of attachment on HP-UX 11.0. You can get  
the ppafrom the output of the lanscancommand.  
Example:  
If the ppaof the 100Base-TX interface is 5, the command to set the card  
to 10Mbits/s and full-duplex mode would be:  
lanadmin -X 10fd 5  
After issuing the lanadmin -X, you must wait at least 11 seconds before  
attempting to use the specified network interface.  
If you want the Duplex Mode setting to be effective in all subsequent  
reboots, you must enter the information in the following file:  
/etc/rc.config.d/hpbtlanconf  
Manually configuring the speed or duplex setting of a switch port on  
12  
Chapter1  
Installing and Configuring PCI 10/100 Base-TX  
Manual Speed and Duplex Mode Configuration  
some switches may disable that switch port from doing autonegotiation.  
Verify that both the card and the switch port are operating in the same  
speed and duplex mode as desired.  
If you use manual configuration to change the card to a different speed  
and duplex mode, you may need to turn autonegotiation on first before  
the manual setting takes place.  
NOTE  
Mismatches between the speed, autonegotiation, or duplex mode of the  
card and switch will cause incorrect operation.  
Chapter 1  
13  
Installing and Configuring PCI 10/100 Base-TX  
Autonegotiation and Autosensing  
Au ton egotia tion a n d Au tosen sin g  
Autonegotiation is a mechanism defined in the IEEE 802.3u specification  
whereby devices sharing a link segment can exchange information and  
automatically configure themselves to operate at the highest capability  
mode shared between them.  
Autonegotiation is like a rotary switch that automatically switches to the  
correct technology such as 10Base-T or 100Base-TX or between half- and  
full-duplex modes. Once the highest performance common mode is  
determined, auto-negotiation passes control of the link to the  
appropriate technology, sets the appropriate duplex mode, and then  
becomes transparent until the link is broken.  
Following is the IEEE 802.3u-defined hierarchy for resolving multiple  
common abilities for a 10/100Base-TX card. The PCI  
10/100Base-TX/9000 ports provide the means for interfacing various  
types of HP 9000 workstations and servers to either a 10Base-T or  
100Base-TX network. 100Base-TX is a subset of 100Base-T networking  
defined by the IEEE 802.3u-1995 standard. 100Base-TX provides 100  
Mbit/s data transmission over category 5 unshielded twisted-par (UTP)  
cable. Two pairs of wires in the cable are used—one wire pair is for  
receiving data, and one wire pair is for transmitting data. The same card  
port that supports 100Base-TX operation can also support 10Base-T  
operation.  
100Base-TX full-duplex  
100Base-TX half-duplex  
10Base-T full-duplex  
10Base-T half-duplex  
For example, if both devices on the link support 10Base-T (half-duplex)  
and 100Base-TX (half-duplex), autonegotiation at both ends will connect  
the 100Base-TX (half-duplex) instead of the 10Base-T (half-duplex).  
Most Fast Ethernet devices on the market today such as hubs and  
switches do not support autonegotiation. Either the speed and duplex  
mode of the device are fixed (as is usually the case with hubs), or they are  
often manually configured at the desired speed and duplex (as is often  
the case for switches). However, switches that support autonegotiation  
are starting to be offered.  
14  
Chapter1  
 
Installing and Configuring PCI 10/100 Base-TX  
What Manuals are Available  
If a PCI 10/100Base-TX/9000 port is connected to a device, such as a  
switch, that is autonegotiating, the PCI card will autonegotiate with the  
device to mutually determine the highest possible speed and duplex  
settings between them.  
NOTE  
If a PCI 10/ 100Base-TX/ 9000 port is connected to a device that does not  
support autonegotiation or a device that has autonegotiation disabled, the  
PCI port will autosense the speed of the link and set itself accordingly.  
Th e d u p lex m od e of th e ca r d w ill be set to h a lf-d u p lex in th is  
ca se. If you want the card to operate in full-duplex mode, you have to set  
it using the method described in Manual Speed and Duplex Mode  
Configuration.  
The PCI 10/100Base-TX port will sense when the connection between  
itself and a hub or switch on the other end of a link has been broken. If a  
connection is made to another (or the same) device, the autonegotiation  
and autosensing process will be done again automatically.  
Autonegotiation and autosensing are also done whenever the interface is  
reset.  
Wh a t Ma n u a ls a r e Ava ila ble  
The following documents summarize installation, configuration,  
verification and troubleshooting of the Dual Port 100Base-TX and Dual  
Port Wide Ultra2 SCSI card:  
Dual Port 100Bas-TX and Dual Port Wide Ultra2 SCSI Quick  
Installation  
Chapter 1  
15  
 
Installing and Configuring PCI 10/100 Base-TX  
Software Availability in Native Languages  
Softw a r e Ava ila bility in Na tive La n gu a ges  
The commands used with this product are the ones supported by the  
Native Language Support Catalog of HP-UX.  
16  
Chapter1  
 
2
Con figu r in g Netw or k  
Con n ectivity Usin g SAM  
This chapter describes how to configure remote connectivity using SAM.  
It contains the following sections:  
17  
 
Configuring Network Connectivity Using SAM  
Step 1: Configuring Network Connectivity  
Step 2: Deleting a Default Gateway (Optional)  
18  
Chapter2  
Configuring Network Connectivity Using SAM  
Step 1: Configuring Network Connectivity  
Step 1: Con figu r in g Netw or k Con n ectivity  
Your system may not be able to communicate with other systems, for  
example, PCs, workstations, servers, etc., until you configure  
system-to-system connections by adding an entry in hosts for the remote  
system. You can use SAM to do this automatically by completing the  
following steps:  
1. At the HP-UX prompt, type: sam  
2. Double click Networking and Communications at the SAM main  
window.  
3. Double click Internet Addresses to enable your system to  
communicate with other systems using the TCP/IP protocol.  
SAM displays the remote system names and Internet addresses that  
are already configured.  
4. Choose Addfrom the Actions” menu to open the Add Internet  
Address window to add the internet address and system name of a  
remote system.  
Use the SAM online help system for information about adding remote  
system connections.  
a. Enter the Internet address for the remote system.  
Upon exiting the Internet Addressfield, SAM checks to make  
sure you have entered a valid IP/Internet address. SAM also  
determines if a gateway is required for the connection (see step  
4c).  
b. Enter the remote system name.  
Upon exiting the Remote System Namefield, SAM checks to make  
sure that connectivity has not already been configured for this  
system. If it has, SAM displays an error message.  
c. Optionally, choose Add Aliasesto open the Add Aliases window if  
you want to configure aliases for a remote system.  
You can modify or remove alias names for a remote system on this  
menu  
Activate the OK button to perform the task and return to the Add  
Chapter 2  
19  
 
Configuring Network Connectivity Using SAM  
Step 1: Configuring Network Connectivity  
Internet Addresses window.  
Proceed to step 5 if a gateway is not required for this remote  
connection.  
SAM displays fields for entering gateway information if a gateway  
is required for this remote system connection. Use the SAM online  
help system for information about gateways.  
5. Activate the OK button to enable your system to communicate with  
this system and return to the System-to-System Connectivity object  
list.  
SAM updates the object list to include the remote system you  
configured.  
NOTE  
You can modify or remove remote systems and modify default gateways  
by highlighting the Remote System Name from the object list and  
choosing Modify, Remove, or Modify Default Gateway from the Actions”  
menu.  
6. Choose Exit from the File” menu.  
7. At the Networking Communications window, choose Exit SAM from  
the File” menu to leave SAM.  
8. Verify remote system configuration.  
a. View the list of remote systems you can communicate with using a  
symbolic name by typing the following command at the HP-UX  
prompt:  
more /etc/hosts  
b. View the configured destinations reached through gateways and  
the gateways used to reach those destinations by typing the  
following command at the HP-UX prompt:  
netstat -r  
To verify that you can communicate with a remote system via the  
10/100Base-TX product, return to chapter 1, Step 8: Verify the  
Installation.”  
20  
Chapter2  
Configuring Network Connectivity Using SAM  
Step 2: Deleting a Default Gateway  
Step 2: Deletin g a Defa u lt Ga tew a y  
To delete a default gateway that you have added with SAM, do the  
following:  
1. Enter the following command at the HP-UX prompt:  
route delete default gateway_hostname  
where gateway_hostname is the hostname of the default gateway you  
want to delete.  
2. Edit the / etc/ rc.cong.d/ netconf file to remove the corresponding  
internet routing configuration parameter values for the gateway. For  
example:  
ROUTE_DESTINATION [0] = ROUTE_GATEWAY [0] =  
ROUTE_COUNT [0] =  
Chapter 2  
21  
 
Configuring Network Connectivity Using SAM  
Step 2: Deleting a Default Gateway  
22  
Chapter2  
3
SCSI Pa r a m eter a n d  
Ter m in a tion In for m a tion  
23  
 
SCSI Parameter and Termination Information  
Configuring SCSI Parameters  
Con figu r in g SCSI Pa r a m eter s  
In order to change the SCSI ID, the path for the card of interest must  
first be determined. Once the path is known, the SCSI ID may be  
changed. The example below lists the commands used to determine the  
path for the card of interest, change the SCSI ID of the A5838A, and also  
some of the other parameters of the card.  
The machine should be booted to the BCH prompt before beginning this  
procedure.  
The path for the A5838A SCSI card is determined as follows:  
- at the Main Menu, type "in" (for information menu)  
- at the Information Menu, type "io" (to display I/O interface information)  
The information displayed should be similar to the following:  
PCI DEVICE INFORMATION  
Path  
Vendor Device Bus Slot  
Id Id  
---- ---- --- ---  
Description  
(dec)  
#
#
-----------  
-----  
.
.
SCSI bus cntlr  
0/2/0/0  
0x1000 0xb 16  
6
.
.
.
The path for the card in this example is '0/2/0/0'.  
Once the path for the card is known:  
- type "main" (to return to main menu)  
- type "ser" (to display service menu)  
- type "scsi" (to display current status of SCSI devices)  
The information displayed should be similar to the following:  
24  
Chapter3  
 
SCSI Parameter and Termination Information  
SCSI Termination Information  
Path (dec) Initiator ID SCSI Rate Auto Term  
------------ -------------- ---------- ---------------  
.
.
0/2/0/0  
7
Fast  
ON  
.
.
.
The SCSI bus speed (rate) can be changed as follows:  
- if at the Main Menu, type "ser" (to display service menu)  
- type "scsi rate <path> <speed>"  
(where <path> is the path obtained as shown above, and <speed> is 'fast'  
or 'ultra')  
The SCSI ID of the controller card can be changed as follows:  
- if at the Main Menu, type "ser" (to display service menu)  
- type "scsi init <path> <new ID>"  
(where <path> is the path obtained as shown above, and  
<new ID> is a single digit, 0 - 7)  
SCSI Ter m in a tion In for m a tion  
The A5838A card is shipped from the factory with autotermination  
enabled. If the A5838A is placed at the end of a SCSI bus,  
autotermination is enabled, so there is no need for external terminators.  
If the A5838A card is installed in the middle of a SCSI bus, you must  
change its SCSI ID and disable autotermination for the port by placing a  
jumper across two-pins on the card. Refer to the Quick Installation  
Guide for details.  
Chapter 3  
25  
 
SCSI Parameter and Termination Information  
SCSI Termination Information  
26  
Chapter3  
4
Tr ou blesh ootin g SCSI  
27  
 
Troubleshooting SCSI  
Troubleshooting  
Tr ou blesh ootin g  
The A5838A host bus adapter is a single field-replaceable unit (FRU) and does not  
contain any field-serviceable parts. Troubleshooting procedures described in this  
section are limited to verifying that the controller is operational and a valid  
connection is established.  
General Procedure  
1. Check the connection.  
Make sure that the correct cable is used, connected, and operating properly and  
that there are no bent pins in any of the connectors.  
2. Check SCSI bus compatibility.  
Ensure transfer rate compatibility with the attached devices using the SCSI boot  
menu command.  
3. Check the controller.  
Inspect the controller to make sure it is seated properly in the PCI bus slot. If  
necessary, power down the system, reseat the controller, and restart the system.  
4. Run diagnostics.  
If a visual inspection of the controller and cable does not reveal any problems,  
or if an action taken as a result of the inspection does not produce a working  
controller, you may want to run diagnostics to determine whether the controller  
can communicate and respond to PCI bus instructions. Diagnostics are  
described in the next section.  
28  
Chapter4  
 
Troubleshooting SCSI  
Troubleshooting  
If diagnostics determine that the controller is defective, you must replace it. Contact  
your local Hewlett-Packard customer representative or call the HP Response Center.  
SCSI Bus Compatibility  
Use the SCSI command to ensure SCSI controller and SCSI device compatibility by  
displaying and selecting SCSI bus parameters.  
The SCSI command is available from the boot menu displayed after the test station  
has booted, provided autoboot is disabled.  
Command  
-------  
Description  
-----------  
AUto [BOot|SEArch ON|OFF]  
BOot [PRI|ALT|<path> <args>]  
BootTimer [time]  
CLEARPIM  
Display or set the specified flag  
Boot from a specified path  
Display or set boot delay time  
Clear PIM storage  
CPUconfig [<proc>] [ON|OFF]  
DEfault  
DIsplay  
Configure/Deconfigure Processor  
Set the sytem to defined values  
Display this menu  
ForthMode  
IO  
LS [<path>|flash]  
OS [hpux|sppux]  
PASSword  
Switch to the Forth OBP interface  
List the I/O devices in the system  
List the boot or flash volume  
Display/Select Operating System  
Set the Forth password  
PAth [PRI|ALT|CON] [<path>]  
PDT [CLEAR|DEBUG]  
Display or modify a path  
Display/clear Non-Volatile PDT state  
PIM_info [cpu#] [HPMC|TOC|LPMC] Display PIM of current or any CPU  
RESET [hard|debug]  
RESTrict [ON|OFF]  
Force a reset of the system  
Display/Select restricted access  
toForth  
SCSI [INIT|RATE] [bus slot val] List/Set SCSI controller parms  
SEArch [<path>]  
SECure [ON|OFF]  
TIme [cn:yr:mo:dy:hr:mn[:ss]]  
VErsion  
Search for boot devices  
Display or set secure boot mode  
Display or set the real-time clock  
Display the firmware versions  
Command:  
Figure 1  
Boot Menu  
Chapter 4  
29  
Troubleshooting SCSI  
Troubleshooting  
SCSI Command  
Use the SCSI command to ensure SCSI controller and SCSI device compatibility by  
displaying and selecting SCSI bus parameters.  
The SCSI command is available from the boot menu displayed after the test station  
has booted, provided autoboot is disabled.  
The SCSI parameters can be displayed and modified using the SCSI command. The  
syntax for this command is:  
SCSI rate bus# slot# rate  
SCSI init bus# slot# id#  
bus#  
slot#  
rate  
The bus number  
The adapter’s slot number  
The adapter’s transfer rate  
0: no limit  
10: Fast SCSI  
20: Ultra SCSI  
id#  
The SCSI ID number of the adapter  
Display and Set SCSI Transfer Rates  
1. Display the SCSI transfer rate for an adapter using the SCSI command.  
The following example lists the SCSI transfer rate for an adapter on bus 5 slot  
2:  
command: SCSI rate 5 2  
PCI device /5.2 = no limit  
Enter the command without specifying a bus or slot number to list transfer rates  
for all bus and slot numbers:  
command: SCSI rate  
2. Set the SCSI transfer rate for an adapter using the SCSI command.  
The following example sets the adapter on bus 5 slot 2 to Fast SCSI and then  
displays the results:  
command: SCSI rate 5 2 10  
30  
Chapter4  
Troubleshooting SCSI  
Troubleshooting  
command: SCSI rate 5 2  
PCI device /5.2 = fast scsi  
Display and Set SCSI IDs  
1. Display the initiator (SCSI) IDs for all controllers, buses, and slots using the  
SCSI command.  
The following example shows the SCSI ID for the controller on bus 5 slot 2.  
SCSI init 5 2  
PCI device /5.2 = 7  
Enter the command without specifying a bus or slot number to list initiator IDs  
for all bus and slot numbers:  
SCSI init  
2. Set the initiator (SCSI) ID of an adapter using the SCSI command.  
The following example sets bus 5 slot 2 to initiator ID 6 and displays the  
results:  
SCSI init 5 2 6  
SCSI init 5 2  
PCI device /5.2 = 6  
Chapter 4  
31  
Troubleshooting SCSI  
Troubleshooting  
Contacting Your HP Representative  
If the equipment is covered by an HP service contract, document the problem as a  
service request and forward it to your HP representative. Include the following  
information where applicable:  
Describe the problem, including the events and symptoms leading up to the  
problem. Attempt to describe the source of the problem.  
Include HP-UX commands, communication subsystem commands,  
functionality of user programs, result codes and messages, and data that can  
reproduce the problem.  
Obtain the version, update, and fix information for all software. To check the  
version of the kernel, enter this command:  
uname -r  
To check patches, enter:  
what /stand/vmunix | grep scsi  
This allows HP to determine if the problem is already known and the correct  
software is installed at your site.  
Illustrate as clearly as possible the context of any messages. Record all error  
messages and numbers that appear at the user terminal and the system console.  
Prepare the formatted output and a copy of the log file for the HP representative  
to analyze.  
Prepare a listing of the HP-UX I/O configuration being used for the HP  
representative to analyze.  
Try to determine the general area within the software where the problem may  
exist. Refer to the appropriate reference manual and follow the guidelines on  
gathering information for that product.  
Document your interim (workaround) solution. The cause of the problem can  
sometimes be found by comparing the circumstances in which the problem  
occurs with the circumstances in which the problem does not occur.  
32  
Chapter4  
Troubleshooting SCSI  
Troubleshooting  
In the event of a system failure, obtain a full memory dump. If the directory  
/var/adm/crash exists, the HP-UX utility /sbin/savecore automatically executes  
during reboot to save the memory dump. HP recommends that you create the  
/tmp/syscore directory after successfully installing this product. Send the output  
of the system failure memory dump to the HP representative.  
Chapter 4  
33  
Troubleshooting SCSI  
Troubleshooting  
If the equipment is not covered by an HP service contract, there may be a charge for  
time and materials.  
SCSI Sense Codes  
The following example shows a typical SCSI error message.  
[+6708 72410001 002a9858 0:7] scsi disk: CHECK CONDITION on  
disk 0:6:5:0  
Read of logical block 509856, count 128  
disk sd45a, block 254920, 65536 bytes  
Valid = 1, Error code = 0x70  
Segment number = 0x00, Filemark = 0, EOM = 0, ILI = 0  
Sense key = 0x1, "RECOVERED ERROR"  
Information = 0x00 0x07 0xc7 0xe4  
[+6709 72410001 002a9a10 0:7] scsi disk:  
length = 0x0a  
Additional sense  
Command-specific information = 0x00 0x00 0x00 0x00  
Additional sense = 0x18, Qualifier = 0x01  
Field replaceable unit code = 0xea  
SKSV = 1, C/D = 0, BPV = 0, Bit pointer = 0  
The status (CHECK CONDITION) and sense key (RECOVERED ERROR) are  
interpreted. The Additional sense and Qualifier codes require interpretation. Use  
both codes to locate the interpretation. In the example, the Additional sense (0x18)  
and Qualifier (0x01) codes are interpreted as “recovered data with error correction  
and retries applied.” and list all possible status and sense key codes. interprets the  
Additional sense and Qualifier codes contained in SCSI error messages reported by  
the console.  
34  
Chapter4  
Troubleshooting SCSI  
Troubleshooting  
SCSI Status Codes  
SCSI Status  
Code  
Name  
0x00  
0x02  
0x04  
0x08  
0x10  
0x14  
0x18  
0x22  
0x28  
good  
check condition  
condition met  
busy  
intermediate  
intermediate - condition met  
reservation conflict  
command terminated  
queue full  
SCSI Sense Keys  
Sense Key  
Name  
0
no sense  
0x1  
0x2  
0x3  
0x4  
0x5  
0x6  
0x7  
0x8  
0x9  
0xa  
0xb  
0xc  
recovered error  
not ready  
medium error  
hardware error  
illegal request  
unit attention  
data protect  
blank check  
vendor-specific  
copy aborted  
aborted command  
equal  
Chapter 4  
35  
   
Troubleshooting SCSI  
Troubleshooting  
SCSI Sense Keys (Continued)  
Sense Key  
0xd  
Name  
volume overflow  
miscompare  
reserved  
0xe  
0xf  
SCSI Additional Sense and Qualifier Codes  
Additional  
Sense Code  
Qualifier  
Code  
Description  
no additional sense information  
0x00  
0x00  
0x01  
0x02  
0x03  
0x04  
0x05  
0x06  
0x11  
0x12  
0x13  
0x14  
0x15  
0x00  
0x00  
0x00  
0x01  
0x02  
0x00  
0x01  
0x02  
0x03  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x03  
0x03  
0x03  
0x04  
0x04  
0x04  
0x04  
filemark detected  
end-of-partition/medium detected  
setmark detected  
beginning of partition/medium detected  
end-of-data detected  
i/o process terminated  
audio play operation in progress  
audio play operation paused  
audio play operation successfully completed  
audio play operation stopped due to error  
no current audio status to return  
no index/sector signal  
no seek complete  
peripheral device write fault  
no write current  
excessive write errors  
logical unit not ready, cause not reportable  
logical unit in process of becoming ready  
logical unit not ready, initializing command required  
logical unit not ready, manual intervention required  
36  
Chapter4  
 
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
0x04  
0x04  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x00  
0x01  
0x02  
0x03  
0x00  
0x00  
0x01  
0x02  
0x00  
0x00  
0x01  
0x02  
0x03  
0x04  
0x05  
0x06  
0x07  
0x08  
0x09  
0x0a  
0x0b  
0x0c  
logical unit not ready, format in progress  
logical unit does not respond to selection  
reference position found  
0x05  
0x06  
0x07  
0x08  
0x08  
0x08  
0x09  
0x09  
0x09  
0x09  
0x0a  
0x0c  
0x0c  
0x0c  
0x10  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
0x11  
multiple peripheral devices selected  
logical unit communication failure  
logical unit communication time-out  
logical unit communication parity error  
track following error  
tracking servo failure  
focus servo failure  
spindle servo failure  
error log overflow  
write error  
write error recovered with auto reallocation  
write error - auto reallocation failed  
id crc or ecc error  
unrecovered read error  
read retries exhausted  
error too long to correct  
multiple read errors  
unrecovered read error - auto reallocate failed  
l-ec uncorrectable error  
circ unrecovered error  
data resynchronization error  
incomplete block read  
no gap found  
miscorrected error  
unrecovered read error - recommend reassignment  
unrecovered read error - recommend rewrite the data  
Chapter 4  
37  
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
0x12  
0x00  
0x00  
0x00  
0x01  
0x02  
0x03  
0x04  
0x00  
0x01  
0x02  
0x00  
0x00  
0x01  
0x02  
0x03  
0x04  
0x05  
0x06  
0x07  
0x08  
0x00  
0x01  
0x02  
0x03  
0x04  
0x05  
0x06  
0x00  
0x01  
address mark not found for id field  
address mark not found for data field  
recorded entity not found  
0x13  
0x14  
0x14  
0x14  
0x14  
0x14  
0x15  
0x15  
0x15  
0x16  
0x17  
0x17  
0x17  
0x17  
0x17  
0x17  
0x17  
0x17  
0x17  
0x18  
0x18  
0x18  
0x18  
0x18  
0x18  
0x18  
0x19  
0x19  
record not found  
filemark or setmark not found  
end-of-data not found  
block sequence error  
random positioning error  
mechanical positioning error  
positioning error detected by read of medium  
data synchronization mark error  
recovered data with no error correction applied  
recovered data with retries  
recovered data with positive head offset  
recovered data with negative head offset  
recovered data with retries and/or circ applied  
recovered data using previous sector id  
recovered data without ecc - data auto-reallocated  
recovered data without ecc - recommend reassignment  
recovered data without ecc - recommend rewrite  
recovered data with error correction applied  
recovered data with error correction and retries applied  
recovered data - data auto-reallocated  
recovered data with circ  
recovered data with lec  
recovered data - recommend reassignment  
recovered data - recommend rewrite  
defect list error  
defect list not available  
38  
Chapter4  
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
defect list error in primary list  
0x19  
0x02  
0x03  
0x00  
0x00  
0x00  
0x01  
0x02  
0x00  
0x00  
0x00  
0x00  
0x01  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x03  
0x00  
0x00  
0x01  
0x00  
0x00  
0x01  
0x02  
0x00  
0x00  
0x01  
0x19  
0x1a  
0x1b  
0x1c  
0x1c  
0x1c  
0x1d  
0x1e  
0x20  
0x21  
0x21  
0x22  
0x24  
0x25  
0x26  
0x26  
0x26  
0x26  
0x27  
0x28  
0x28  
0x29  
0x2a  
0x2a  
0x2a  
0x2b  
0x2c  
0x2c  
defect list error in grown list  
parameter list length error  
synchronous data transfer error  
defect list not found  
primary defect list not found  
grown defect list not found  
miscompare during verify operation  
recovered id with ecc  
invalid command operation code  
logical block address out of range  
invalid element address  
illegal function  
invalid field in cdb  
logical unit not supported  
invalid field in parameter list  
parameter not supported  
parameter value invalid  
threshold parameters not supported  
write protected  
not ready to ready transition (medium may have changed)  
import or export element accessed  
power on, reset, or bus device reset occurred  
parameters changed  
mode parameters changed  
log parameters changed  
copy cannot execute since host cannot disconnect  
command sequence error  
too many windows specified  
Chapter 4  
39  
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
0x2f  
0x00  
0x00  
0x01  
0x02  
0x03  
0x00  
0x00  
0x01  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x03  
0x04  
0x05  
0x06  
0x07  
0x08  
0x09  
0x0a  
0x0b  
0x0c  
0x0d  
0x0e  
0x00  
commands cleared by another initiator  
incompatible medium installed  
cannot read medium - unknown format  
cannot read medium - incompatible format  
cleaning cartridge installed  
medium format corrupted  
0x30  
0x30  
0x30  
0x30  
0x31  
0x32  
0x32  
0x33  
0x36  
0x37  
0x39  
0x3a  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3b  
0x3d  
no defect spare location available  
defect list update failure  
tape length error  
ribbon, ink, or tower failure  
rounded parameter  
saving parameters not supported  
medium not present  
sequential positioning error  
tape position error at beginning-of-medium  
tape position error at end-of-medium  
tape or electronic vertical forms unit not ready  
slew failure  
paper jam  
failed to sense top-of-form  
failed to sense bottom-of-form  
reposition error  
read past end of medium  
read past beginning of medium  
position past end of medium  
position past beginning of medium  
medium destination element full  
medium source element empty  
invalid bits in identify message  
40  
Chapter4  
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
0x3e  
0x00  
0x00  
0x01  
0x02  
0x03  
0x00  
nn  
logical unit has not self-configured yet  
target operating conditions have changed  
microcode has been changed  
changed operating definition  
inquiry data has changed  
ram failure  
0x3f  
0x3f  
0x3f  
0x3f  
0x40  
0x40  
0x41  
0x42  
0x43  
0x44  
0x45  
0x46  
0x47  
0x48  
0x49  
0x4a  
0x4b  
0x4c  
0x4e  
0x50  
0x50  
0x50  
0x51  
0x52  
0x53  
0x53  
0x53  
0x54  
diagnostic failure on component nn  
data path failure  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x00  
0x00  
0x00  
0x01  
0x02  
0x00  
power-on or self-test failure  
message error  
internal target failure  
select or reselect failure  
unsuccessful soft reset  
scsi parity error  
initiator detected error message received  
invalid message error  
command phase error  
data phase error  
logical unit failed self-configuration  
overlapped commands attempted  
write append error  
write append position error  
position error related to timing  
erase failure  
cartridge fault  
media load or eject failed  
unload tape failure  
medium removal prevented  
scsi to host system interface failure  
Chapter 4  
41  
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
0x55  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x030  
0x00  
0x01  
0x02  
0x03  
0x00  
0x01  
0x02  
0x00  
0x00  
0x00  
0x00  
0x00  
0x01  
0x02  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
system resource failure  
reserved  
0x56  
0x57  
0x58  
0x59  
0x5A  
0x5A  
0x5A  
0x5A  
0x5B  
0x5B  
0x5B  
0x5B  
0x5C  
0x5C  
0x5C  
0x5D  
0x5E  
0x5F  
0x60  
0x61  
0x61  
0x61  
0x62  
0x63  
0x64  
0x65  
0x66  
0x67  
unable to recover table of contents  
generation does not exist  
updated block read  
operator request or state change input (unspecified)  
operator medium removal request  
operator selected write protect  
operator selected write permit  
log exception  
threshold condition met  
log counter at maximum  
log list codes exhausted  
rpl status change  
spindles synchronized  
spindles not synchronized  
reserved  
reserved  
reserved  
lamp failure  
video acquisition error  
unable to acquire video  
out of focus  
scan head positioning error  
end of user area encountered on this track  
illegal mode for this track  
reserved  
reserved  
reserved  
42  
Chapter4  
Troubleshooting SCSI  
Troubleshooting  
SCSI Additional Sense and Qualifier Codes (Continued)  
Additional  
Sense Code  
Qualifier  
Code  
Description  
0x68  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
0x00  
reserved  
reserved  
reserved  
reserved  
reserved  
reserved  
reserved  
reserved  
0x69  
0x6A  
0x6B  
0x6C  
0x6D  
0x6E  
0x6F  
Chapter 4  
43  
Troubleshooting SCSI  
Troubleshooting  
44  
Chapter4  
5
10/100Ba se-TX Resou r ces  
In addition to this manual, use the following resources to maintain and  
administer PCI 10/100Base-TX/9000.  
45  
 
10/100Base-TX Resources  
HP-UX Manual Reference Pages  
HP -UX Ma n u a l Refer en ce Pa ges  
While installing, configuring, or troubleshooting 10/100Base-TX, you  
may need to refer to any of the following online manual reference pages  
(man pages) for useful HP-UX operating system or 10/100Base-TX  
commands. To display a man page, type the following at the system  
prompt: man <command name>. For example, man arp.  
arp(1M) displays and modifies the Internet-to-station address  
mapping tables used by the Address Resolution Protocol.  
hosts(4) is a database that contains a single line entry for each host  
name entry.  
ifconfig(1M) assigns an address to a network interface, and configures  
and displays network parameters.  
ioscan(1M) scans system hardware, usable I/O system devices, or  
kernel I/O system data structures as appropriate, and lists the  
results.  
lanadmin(1M) resets or reports the status of the LAN card.  
lanconfig(1M) sets/resets the packet encapsulation method for a  
network interface.  
lanscan(1M) displays information about LAN cards that are  
successfully bound to the system.  
linkloop(1M) verifies network connectivity through the Data Link  
Layer (OSI Layer 2).  
netfmt(1M) formats common tracing and logging binary files.  
netstat(1) provides network statistics and information about network  
connections.  
nettl(1M) logs network events and traces packets as they enter and  
exit the 10/100Base-TX driver.  
ping(1M) verifies network connectivity through the Network Layer  
(OSI Layer 3) and reports the round-trip time of communications  
between the local and remote hosts.  
route(1M) adds and deletes entries to the network routing table.  
sam(1M) configures networking software.  
46  
Chapter5  
 
10/100Base-TX Resources  
HP-UX Manual Reference Pages  
swinstall(1M) loads software filesets onto 10.x systems.  
swverify(1M) verifies software installation.  
Chapter 5  
47  
10/100Base-TX Resources  
Error Messages  
Er r or Messa ges  
PCI 10/100Base-TX comes with an online message catalog that is used to  
report networking problems. You must use the nettl logging and tracing  
utility to display the probable cause and action for a message.  
48  
Chapter5  
 
10/100Base-TX Resources  
Logging Messages  
Loggin g Messa ges  
HP 100Base-TX/9000 uses the nettl(1M) logging and tracing facility  
supplied with HP-UX. You may access the logging and tracing utility  
using either the graphical user interface (GUI) version or the command  
line interface.  
Features of the GUI version, which are now a part of your HP 9000  
system, include:  
An interface which guides you through logging and tracing tasks.  
An interface which allows you to create and format reports.  
The capability to collect logging and tracing subsystem-specific  
information.  
Report screens which are updated instantaneously with current  
logging and tracing information by the subsystem.  
Context-sensitive on-line help.  
To access the GUI version of the logging and tracing utility, run the  
command:  
nettladm  
See the nettladm(1M) man page for information on using the GUI  
version, or the nettl(1M) manual (man) page for information on using the  
command line interface.  
Listed below are some example commands using the command line  
interface.  
To examine the log file with cause and action descriptions.  
netfmt -v -f /var/adm/nettl.LOG00| more  
The -voption enables the reporting of available cause and action  
descriptions for each log message. A sample 10/100Base-TX log  
message using the -voption is shown below.  
******100 Mb/s LAN/9000 Networking**********************  
Fri May 16 PDT 1997 15:08:07.091398 DISASTER  
Subsys:LAN100 Loc:00000  
Chapter 5  
49  
 
10/100Base-TX Resources  
Logging Messages  
<6011> HP PCI 10/100Base-T driver detected bad cable  
connection between the adapter in slot 2 and the hub  
or switch.  
To examine just the log messages in the log file.  
netfmt -f /var/adm/nettl.LOG00  
To check network logging and tracing status.  
nettl -status  
To start 10/100Base-TX tracing to the file / tmp/ tracefile.TRC0.  
nettl(1m) adds the .TRC0 postfix for you.  
nettl -traceon all -entity PCI100bt -file  
/tmp/tracefile  
To stop 100Base-TX tracing.  
nettl  
-traceoff  
-entity PCI100bt  
To format the 10/100Base-TX trace file into the file / tmp/ traceout.  
netfmt -f /tmp/tracefile.TRC0 > /tmp/traceout  
Refer to the netfmt(1M) man page for further information about this card  
and how to create a filter for trace formatting.  
50  
Chapter5  
10/100Base-TX Resources  
Manual Installation and Configuration  
Ma n u a l In sta lla tion a n d Con figu r a tion  
If you want to manually install and configure your 10/100Base-TX/9000  
product, refer to the detailed instructions in the Installing and  
Administering LAN/ 9000 Software manual.  
You may need some of the following 10/100Base-TX/9000-specific  
information when you follow those steps:  
HP 9000 PCI driver keyword: btlan  
The driver for the PCI 10/100Base-TX card has a floating major  
number (that is, a major number assigned dynamically by the  
operating system).  
Chapter 5  
51  
 
10/100Base-TX Resources  
Contacting Your HP Representative  
Con ta ctin g You r HP Rep r esen ta tive  
If you have no service contract with HP, you may follow the procedure  
described below, but you will be billed accordingly for time and materials.  
If you have a service contract with HP, document the problem as a  
Service Request (SR) and forward it to your HP representative. Include  
the following information where applicable:  
A characterization of the problem. Describe the events leading up to  
and including the problem. Attempt to describe the source and  
symptoms of the problem.  
Your characterization should include: HP-UX commands;  
communication subsystem commands; job streams; result codes and  
messages; and data that can reproduce the problem. You should also  
provide a network map with the host name, IP/Internet address, and  
station address of each system connected with the HP system.  
Illustrate as clearly as possible the context of any message(s).  
Prepare copies of information displayed at the system console and  
user terminal.  
Obtain the version, update, and fix information for all software. To  
check the 10/100Base-TX version number, execute what vmunix and  
look for the keyword, vtlan1.  
To check the version of your kernel, execute uname -r.  
This allows HP to determine if the problem is already known and if  
the correct software is installed at your site.  
Prepare copies of the /etc/h osts, and /etc/r c.con fig.d /n etcon f files.  
Execute the dmesg command and record messages about the status of  
the 10/100Base-TX card.  
Execute the lanscan -v command and record the output.  
Execute the display command of the lanadmin diagnostic on the  
10/100Base-TX interface and record the output.  
Record the troubleshooting flowchart number and step number where  
you are unable to resolve the problem.  
Record all error messages and numbers that appear at the user  
terminal and the system console.  
52  
Chapter5  
 
10/100Base-TX Resources  
Contacting Your HP Representative  
Save all network log files. Make sure that ERROR and DISASTER log  
classes are enabled when log files are collected.  
Prepare the formatted output and a copy of the log file for your HP  
representative to further analyze.  
Prepare a listing of the HP-UX I/O configuration you are using for  
your HP representative to further analyze. Use the ioscan(1M)  
command to help collect this information  
Try to determine the general area within the software where you  
think the problem exists. Refer to the appropriate reference manual  
and follow the guidelines on gathering information for that product.  
Document your interim, or workaround,” solution. The cause of the  
problem can sometimes be found by comparing the circumstances in  
which it occurs with the circumstances in which it does not occur.  
Create copies of any Internet or 10/100Base-TX/9000 link trace files  
that were active when the problem occurred for your HP  
representative to further analyze.  
In th e even t of a system fa ilu r e, a fu ll m em or y d u m p m u st be  
ta k en . Use the HP-UX utility savecore(1M) to save a core dump.  
Send the output to your HP representative.  
Chapter 5  
53  
10/100Base-TX Resources  
Contacting Your HP Representative  
54  
Chapter5  
6
Tr ou blesh ootin g  
10/100Ba se-TX/9000  
This chapter provides guidelines for troubleshooting 10/100Base-TX. It  
contains the following sections:  
55  
 
Troubleshooting 10/100Base-TX/9000  
Troubleshooting Overview.  
Diagnostic Flowcharts.  
56  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Performance Troubleshooting  
Per for m a n ce Tr ou blesh ootin g  
This section is intended to provide system administrators or  
advanced users with detailed information on how to troubleshoot  
performance related problems with the PCI 100BT product.  
Below, a few key terms are defined to help in understanding the  
troubleshooting information.  
Key Terms:  
Tr a n sm it Th r esh old :  
The transmit threshold value determines how many bytes must be in the  
PCI 100BT transmit FIFO before transmission of the bits onto the  
ethernet cable will begin.  
Tr a n sm it Un d er r u n :  
A transmit underrun error occurs when the PCI 100BT transmitter  
encounters an empty transmit FIFO during the transmission of bits onto  
the ethernet cable.  
Mem or y Su bsystem La ten cy:  
The memory subsystem latency is defined to be the amount of time it  
takes to move data from system memory to an I/O device. This time  
includes the arbitration delay for the I/O device and for each bus bridge  
between the system memory controller and the I/O device.  
Ar bitr a tion Dela y:  
The time it takes an I/O device, or bus bridge to acquire the I/O bus for  
data transfer.  
Tr a n sm it F IF O:  
The transmit FIFO is a buffer on the PCI 100BT card used to hold data  
transferred from system memory to the PCI 100BT card.  
The PCI 100BT product is currently optimized to achieve the best single  
card performance. In order to achieve this performance the PCI 100BT  
product has set the Transmit Threshold to an aggressive value. The  
Transmit Threshold is set so that transmission will begin after 512 bytes  
Chapter 6  
57  
 
Troubleshooting 10/100Base-TX/9000  
Performance Troubleshooting  
are in the transmit FIFO.  
While the current Transmit Threshold value allows the PCI 100BT  
product to achieve it's best performance, it also increases the probability  
of Transmit Underrun errors. A large number of Transmit Underrun  
errors (more than 1 out of every 1000 packets) can cause a noticeable  
drop in networking performance.Transmit Underrun errors may occur  
when there is sufficient bus contention from competing I/O devices.  
These errors can be monitored in two ways:  
1. Examine the output from the netstat -I interfacecommand. If  
the number of output errors is high (more than 1 out of every X  
packets) then the system is most likely suffering from transmit  
underruns on the specified network interface and corrective action  
must be taken to resolve the problem.  
2. Turn on nettl errors and warnings for the network interface being  
monitored. The following command will turn on disasters, errors and  
warnings for the network interface with Instance number 1. (NOTE:  
It is highly recommended to always keep disasters and errors  
enabled).  
n ettl -log 0xe -e P CI100bt -C 1  
The nettl log file (by default is called /var/adm/nettl.LOG00) should then  
be monitored for the following message:  
HPPCI 10/100Base-T driver encountered a Transmit Underflow  
If a significant number of these messages occur, and the timestamps for  
each of the messages are within 30 seconds of each other, then the  
specified networking interface will suffer a noticeable performance drop.  
Corrective action must be taken to resolve this problem.  
Cor r ective Action  
The PCI 100BT product supports 3 levels of Transmit Threshold. These  
3 levels are modified via the -S option of the lanadmin command as  
follows:  
lanadmin -S TransmitThreshold nmid  
where:  
a TransmitThreshold of 1024 is somewhat aggressive  
a TransmitThreshold of 512 is most aggressive  
58  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Performance Troubleshooting  
a TransmitThreshold of 1500 is conservative  
In all of the cases above after setting the Transmit Threshold mode as  
specified, the lanadmin command will echo the current speed of the  
interface as follows; this output may be ignored (output shown for 100  
Mbits/s operation):  
old speed= 100000000  
new speed= 100000000  
After issuing the lanadmin -Syou must wait at least 5 seconds before  
attempting to use the specified network interface.  
If the desired Transmit Threshold setting needs to be effective in all  
subsequent reboots, you must create an SD script and include it in the  
/sbin /in it.d directory so that it gets executed on each reboot.  
Chapter 6  
59  
Troubleshooting 10/100Base-TX/9000  
Troubleshooting Overview  
Tr ou blesh ootin g Over view  
10/100Base-TX problems can be caused by problems in a variety of  
hardware and software components. The problem impacting your system  
may originate in another part of the 10/100Base-TX network.  
As with any troubleshooting, a systematic approach is helpful. The  
following two tables and the following flowcharts provide a logical  
sequence of steps to follow when troubleshooting 10/100Base-TX/9000.  
Using the diagnostic flowcharts provided in this chapter, identify  
whether the problem is with 10/100Base-TX/9000 or any of the  
connections to the hub or switch, or whether it is in some other part of  
the network, verify your assumptions and, if it is limited to  
10/100Base-TX/9000 software or hardware, correct the problem.  
NOTE  
To quickly isolate and diagnose 10/100Base-TX/9000 problems, follow the  
steps in the troubleshooting flowcharts, beginning with Flowchart 1, and  
stay with the flowcharts until the problems are resolved. Continue  
sequentially through flowcharts 2, 3, 4, 5, 6, and 7, referring back to  
flowchart 1 (ping) until you have corrected the problems.  
If you cannot solve the problem on your own, contact your HP  
representative. Use the guidelines at the end of chapter 3 to help you  
effectively communicate what is wrong. The 10/100Base-TX product uses  
diagnostic tools compatible with the HP LAN/9000 Link product.  
60  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
Dia gn ostic F low ch a r ts  
Below is a summary of the types of network tests in the diagnostic  
flowcharts. Follow the flowcharts in sequence beginning with flowchart 1.  
Continue sequentially through flowcharts 2, 3, 4, 5, 6, 7, 8, and 9,  
referring back to flowchart 1 (ping), as indicated at the end of each  
flowchart, until you have corrected the problem.  
Ta ble 6-1  
F low ch a r t Descr ip tion s  
Flowchart  
Description  
1
2
Network Level Loopback Test  
10/100Base-TX Connections/LED Test  
Configuration Test  
3, 4, and 5  
6
7
8
9
Network Level Loopback Test  
Link Level Loopback Test  
Transport Level Loopback Test (using ARPA)  
Bridge/Gateway Loopback Test  
Netw or k Level Loop ba ck Test: Checks roundtrip communication  
between Network Layers on the source and target host using the  
ping(1M) command.  
10/100Ba se-TX Con n ection s/LED Test: Checks that all the hardware  
connections between your system and the 10/100Base-TX network are  
connected and operational.  
Con figu r a tion Test: Verifies the configuration of the network interface  
on a host using the lanscan(1M), netfmt -vf, lanadmin(1M), and  
ifconfig(1M) commands.  
Netw or k Level Loop ba ck Test (con t): Checks arp entries using the  
arp(1M) command.  
Lin k Level Loop ba ck Test: Checks roundtrip communication between  
Link Levels on the source and target host using the linkloop(1M)  
diagnostic.  
Chapter 6  
61  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
Tr a n sp or t Level Loop ba ck Test: Checks roundtrip communication  
between Transport Layers on the source and target host using ARPA  
services telnet and ftp commands.  
Br id ge/Ga tew a y Loop ba ck Test: Checks general network connections  
through a gateway.  
62  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 1: Netw or k Level Loop ba ck Test  
Figu r e 6-1  
F low ch a r t 1: Netw or k Level Loop ba ck Test  
1
A
Execute: ping  
to remote host  
B
ping  
successful  
yes  
Stop  
?
no  
C
D
E
yes  
yes  
Network  
unreachable  
?
3
no  
Command  
hangs  
2,3,4,5  
6 & 7  
?
no  
F
Unknown  
yes  
yes  
Correct BIND, YP or  
/etc/hosts configuration  
host  
?
1
no  
H
G
No route  
to host  
?
Add route  
table entry  
no  
Call HP  
Chapter 6  
63  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 1 P r oced u r es  
A.  
Execu te: p in g to r em ote h ost. Using ping(1M), send  
a message to the remote host to which you are having  
problems connecting. For example:  
ping spiff  
B.  
p in g su ccessfu l? A message is printed to stdout for  
each ping packet returned by the remote host. If  
packets are being returned, your system has network  
level connectivity to the remote host. Note what  
percentage of the total packets are lost, if any. Losing  
ten percent or more may indicate the network or  
remote host is extremely busy. You may also find it  
useful to note the round-trip transmission times.  
Periodically high transmission times may indicate that  
the network or remote host is extremely busy.  
Consistently high transmission times may indicate the  
local host is extremely busy. If a message is not  
returned after executing ping, ping is not successful.  
Do Cn tr l C to stop the ping output.  
C.  
D.  
Netw or k u n r ea ch a ble? If yes, go to flowchart 3 to  
display connection status using the lanscan(1M)  
command.  
Com m a n d h a n gs. If a message is not returned after  
executing ping, go to flowcharts 2 through 7, referring  
back to flowchart 1 (ping) until you have corrected the  
problem.  
E.  
F.  
Un k n ow n h ost? If you receive this message, go to step  
F.  
Cor r ect BIND, YP or h osts con figu r a tion . Add the  
missing host name and start again with flowchart 1.  
G.  
No r ou te to h ost? If Error= Sendto: No route to  
host, go to Step H. Otherwise, call your HP  
representative for help.  
H.  
Ad d r ou te ta ble en tr y. Using route, add a route table  
entry for that host. Refer to the route(1M) online man  
page for more details. Start again with flowchart 1.  
64  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 2: 10/100Ba se-TX Con n ection s/LED Test  
Figu r e 6-2  
F low ch a r t 2: 10/100Ba se-TX Con n ection s/LED Test  
2
C
A
Check card installation.  
Reset card. Call HP if  
problem persists.  
B
yes  
Check:  
Test Error Message  
Power outlet  
on Screen (dmesg ouput)  
?
no  
1
D
Check status of  
Link LED  
F
Check connection to hub  
or switch. Ensure switch  
is autonegotiating.  
E
yes  
Link LED = OFF  
Ensure hub/switch is  
10Base-T or 100Base-TX.  
Reset card.  
?
no  
1
H
G
no  
Set attached hub or switch  
to correct speed, duplex mode,  
and autonegotiation.  
Reset card.  
Do link speed and duplex  
mode match switch  
?
yes  
1
3
Chapter 6  
65  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 2 P r oced u r es  
A.  
Ch eck Pow er ou tlet. Ensure the power cord is  
plugged in to a live outlet.  
B.  
Test Er r or Messa ge on Scr een ? At th e HP -UX  
p r om p t, typ e th e d m esg com m a n d , a n d look for  
a n er r or m essa ge. Does th e d m esg ou tp u t sh ow  
a n er r or m essa ge fr om btla n ? If not, go to step D.  
Note: even if the Test LED is OFF, a card problem is  
still possible if either of the following two messages  
appear:  
btlan: Error: Motherboard failed to complete  
reset.  
btlan: Error: Motherboard failed selftest;error  
code= 0x?  
C.  
Ch eck ca r d in sta lla tion . If d m esg r ep or ted a n  
er r or m essa ge fr om btla n , r eset ca r d a ccor d in g  
to Step s D th r ou gh G in F low ch a r t 4. If p r oblem  
p er sists, ca ll HP. Go back to flowchart 1.  
D.  
E.  
Check status of Link LED.  
Lin k LED = OF F ? If it is off, proceed to step F.  
If Lin k LED = ON, proceed to step G.  
F.  
If Link LED = OFF, check connection to hub or switch.  
Ensure switch is notautonegotiating. Ensure hub or  
switch is 10Base-T or 100Base-TX. Reset card  
according to Steps D through G in Flowchart 4. Go  
back to flowchart 1.  
G.  
H.  
Do lin k sp eed a n d d u p lex m od e m a tch sw itch ? If  
they do, proceed to flowchart 3.  
If Lin k sp eed a n d d u p lex m od e d o n ot m a tch  
w h a t you exp ect, set a tta ch ed h u b or sw itch to  
th e cor r ect lin k sp eed a n d d u p lex m od e, a n d  
en a ble a u ton egotia tion . Reset ca r d a ccor d in g to  
Step s D th r ou gh G in F low ch a r t 4. Go back to  
flowchart 1.  
66  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 3: Con figu r a tion Test  
Figu r e 6-3  
F low ch a r t 3: Con figu r a tion Test  
3
A
Execute: lanscan  
Is  
B
yes  
your interface  
displayed  
?
C
yes  
Hardware  
up  
no  
5
D
?
no  
Run ioscan  
-f  
Install driver.  
Verify or edit  
/stand/system  
F
E
no  
Is driver in  
kernel  
?
to add driver keyword  
btlan  
Regen kernel.  
yes  
G
H
Reboot the  
system  
Check  
hardware  
I
no  
Problem  
fixed  
?
1
yes  
Stop  
Chapter 6  
67  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 3 P r oced u r es  
NOTE  
Check that your 10/100Base-TX connectors to the card and hub (or wall  
plug) are fully connected before beginning this flowchart.  
A.  
B.  
Execu te: la n sca n . Enter the lanscan command to  
display information about LAN cards that are  
successfully bound to the system. See the lanscan  
online manpage for more detailed information.  
Is you r in ter fa ce d isp la yed ? lanscan shows  
information about every LAN card in the system  
backplane. The Hardware Path of one of the entries  
should correspond to the PCI 10/100Base-TX card slot  
multiplied times 4. For example, a hardware path of 32  
corresponds to an PCI 10/100Base-TX card in slot 8.  
C.  
D.  
Hardware up.The hardware state is operational if up is  
displayed for the 10/100Base-TX card under the  
Hardware State heading. If it is, continue to flowchart  
5. If not, go to D.  
Ru n iosca n . ioscan will scan the system hardware  
and list the results. If you execute ioscan -f, output  
68  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
similar to the following will be displayed:  
Chapter 6  
69  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
Class  
I H/W Path  
Driver  
S/W State H/W Type Description  
========================================================================  
bc  
0
root  
ccio  
ccio  
c720  
CLAIMED BUS_NEXUS  
bc  
1 8  
2 10  
0 10/0  
CLAIMED BUS_NEXUS I/O Adapter  
CLAIMED BUS_NEXUS I/O Adapter  
bc  
ext_bus  
CLAIMED INTERFACE GSC built-in  
Fast/Wide SCSI Interface  
bc  
3 10/4  
bc  
CLAIMED BUS_NEXUS Bus Converter  
CLAIMED INTERFACE MUX  
tty  
0 10/4/0  
2 10/4/4  
1 10/4/4.1  
0 10/8  
mux2  
lanmux  
lan  
lanmux0  
lan3  
CLAIMED INTERFACE HP J2146A - 80  
CLAIMED INTERFACE  
ba  
GSCtoPCI  
CLAIMED BUS_NEXUS PCI Bus Bridge  
lan  
2 10/8/1/0  
3 10/8/2/0  
1 10/12  
btlan CLAIMED PCI (10110009)  
lan  
btlan CLAIMED PCI (10110009)  
ba  
bus_adapter CLAIMED BUS_NEXUS Core I/O Adapt  
ext_bus  
2 10/12/0  
CentIf  
CLAIMED INTERFACE Built-in Paral  
Interface  
ext_bus  
target  
1 10/12/5  
c720  
tgt  
CLAIMED INTERFACE Built-in SCSI  
CLAIMED DEVICE  
3 10/12/5.2  
disk  
2 10/12/5.2.0 sdisk  
CLAIMED DEVICE  
TOSHIBA CD-ROM  
XM-5401TA  
target  
ct1  
3 10/12/5.7  
tgt  
CLAIMED DEVICE  
CLAIMED DEVICE  
1 10/12/5.7.0 sct1  
Initiator  
lan  
0 10/12/6  
0 10/12/7  
lan2  
ps2  
CLAIMED INTERFACE Built-in LAN  
ps2  
CLAIMED INTERFACE Built-in Keyboar  
processor 0 32  
processor CLAIMED PROCESSOR Processor  
70  
Chapter6  
processor 1 34  
processor CLAIMED PROCESSOR Processor  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
E.  
Is d r iver in k er n el? If the driver has not been  
generated into the kernel, ioscan output will be:  
ioscan -f  
Class  
I H/W Path Driver  
S/W State H/W Type  
Description  
===============================================================  
====  
unknown  
-1 10/4/4  
UNKNOWN UNCLAIMED INTERFACE  
The class and driver fields alone will indicate  
unknown” status if the kernel has not been generated.  
If the driver has not been generated, continue to step  
H. If the driver is in the kernel, go to step G.  
F.  
Ver ify or ed it /sta n d /system a n d r egen k er n el.  
Verify/edit / stand/ system contains the btlan keyword.  
If not, see Creating a New Kernel” in chapter 3 of the  
Installing and Administering LAN/ 9000 Software  
manual for instructions on how to edit /sta n d /system  
to create a new kernel.  
G.  
Ch eck h a r d w a r e. Verify that the network card is  
seated correctly and that it is operational.  
H.  
I.  
Reboot th e system .  
P r oblem fixed ? If you have found the 10/100Base-TX  
card problem, stop. If not, start again with flowchart 1.  
Chapter 6  
71  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 4: Con figu r a tion Test  
Figu r e 6-4  
F low ch a r t 4: Con figu r a tion Test  
4
A
Execute: netfmt  
B
Check causes and  
actions on display  
in the formatted log  
output  
C
yes  
Problem  
solved  
1
?
no  
D
E
Execute: lanadmin  
Select LAN from  
Menu  
F
Select NMID command  
and enter 10/100  
Base-TX nmid  
G
Reset card  
H
Reset  
successful  
?
yes  
1
no  
4A  
72  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 4 P r oced u r es  
A.  
Execu te: n etfm t. Use the netfmt command to view log  
data (error and disaster messages). An example  
command is shown below.  
netfmt -v -f /var/adm/nettl.LOG00 | more  
B.  
Ch eck ca u ses a n d a ction s on d isp la y in th e  
for m a tted log ou tp u t. Use the time stamp to find the  
proper logs. Ensure that you are looking at the  
10/100Base-TX information.  
C.  
D.  
P r oblem solved . If yes, go to flowchart 1. If not,  
continue with step D.  
Execu te la n a d m in . Run lanadmin(1M). For a  
complete description of this command, refer to the  
lanadmin(1M) on-line manual page.  
E.  
F.  
Select LAN fr om Men u . Select la n from the menu to  
enter LAN Interface Diagnostic.  
Select th e NMID com m a n d a n d en ter th e  
10/100Ba se-TX NMID. You can use the lanscan  
command to find the current NMID for 10/100Base-TX.  
The NMID you enter becomes the current device to be  
tested.  
G.  
H.  
Reset th e ca r d a ccor d in g to Step s D th r ou gh G in  
F low ch a r t 4. Using the reset command in lanadmin  
re-executes the LAN card self-test.  
Reset su ccessfu l? The reset is successful if no errors  
are displayed as a result of the reset command. If the  
self-test was successful, the problem may be that you  
are not connected to the 10/100Base-TX network.  
Correct the problem and verify the resolution by  
continuing with flowchart 1. Otherwise, go to flowchart  
4A.  
Chapter 6  
73  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 4A: Con figu r a tion Test  
Figu r e 6-5  
F low ch a r t 4A: Con figu r a tion Test  
4A  
A
Execute: netfmt  
B
Check causes and  
actions on display  
in the formatted log  
output  
C
yes  
Problem  
solved  
1
?
no  
Call HP  
F low ch a r t 4A P r oced u r es  
A.  
Execu te: n etfm t. Use the netfmt command to view log  
data (error and disaster messages). An example netfmt  
command is shown below:  
netfmt -v -f /var/adm/nettl.LOG00 | more  
Extend the search to LOG01 as information may have  
rolled (overflowed) into this file from LOG00.  
B.  
Ch eck ca u ses a n d a ction s on d isp la y in th e  
for m a tted log ou tp u t. Use the time stamp to find the  
proper logs. Ensure that you are looking at the  
74  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
10/100Base-TX information.  
C.  
P r oblem solved . If yes, go to flowchart 1. If not,  
contact your HP representative.  
Chapter 6  
75  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 5: Con figu r a tion Test  
Figu r e 6-6  
F low ch a r t 5: Con figu r a tion Test  
5
A
Execute  
ifconfig <interface>  
...<IP address> up  
B
C
Execute:  
>
ifconfig <interface  
E
D
F
Are  
flags correct  
?
ifconfig  
successful  
no  
no  
Correct ifconfig  
flag settings  
5
?
yes  
yes  
ifconfig  
entry in  
H
I
Any error  
messages  
returned  
yes  
no  
Call HP  
/etc/rc.config.d/netconf  
?
?
yes  
no  
G
Correct problem  
according to the  
message received  
Add ifconfig  
command to  
/etc/rc.config.d/netconf  
5
1
F low ch a r t 5 P r oced u r es  
A.  
Execu te: ifcon fig <in ter fa ce> <IP a d d r ess> u p .  
76  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
Execute ifconfig on the interface you want to configure  
in order to ensure that the interface is enabled. For  
example, to configure the 10/100Base-TX interface  
lan1, enter:  
ifconfig lan1 192.6.1.17 up  
For more examples of the ifconfig command, refer to  
the ifconfig(1M) online man page.  
B.  
C.  
Execu te: ifcon fig <in ter fa ce>. Execute ifconfig  
without the up parameter again on the interface you  
want to test to check the flag setting for the up  
parameter. For example, to check the 10/100Base-TX  
interface lan1, enter:  
ifconfig lan1  
ifcon fig su ccessfu l? ifconfig is successful if the  
output shows the correct Internet address and the  
flags: <UP,BROADCAST, NOTRAILERS, RUNNING>.  
Note: Make sure the UP flag is displayed.  
D.  
E.  
Are flags correct? If flags are not correct, use the  
ifconfig command to correct them. If they are correct,  
go to step F.  
Cor r ect ifcon fig fla g settin gs. If ifconfig returns an  
incorrect flag setting, re-execute the command with the  
proper setting. For more information, refer to the  
ifconfig(1M) online man page. Start again with  
flowchart 5, as necessary.  
F.  
An y er r or m essa ge r etu r n ed ? If ifconfig is not  
successful, and an error message appears, go to Step G.  
If no error messages appear, contact your HP  
representative.  
G.  
H.  
Cor r ect p r oblem a ccor d in g to th e m essa ge  
r eceived . If you received an error message, make the  
appropriate corrections stated in the message and then  
begin this procedure again.  
ifcon fig en tr y in /etc/r c.con fig.d /n etcon f? Check  
that there is an entry in the / etc/ rc.cong.d/ netconf  
file for your 10/100Base-TX card.  
Chapter 6  
77  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
I.  
Ad d ifcon fig com m a n d to /etc/r c.con fig.d /n etcon f  
file. Add the ifconfig command to  
/ etc/ rc.cong.d/ netconf, and reboot. For more  
information, refer to the ifconfig(1M) online man page.  
Go to flowchart 1 to verify that the problem has been  
solved.  
78  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 6: Netw or k Level Loop ba ck Test  
Figu r e 6-7  
F low ch a r t 6: Netw or k Level Loop ba ck Test  
6
A
B
Host entry  
in ARP  
cache  
yes  
no  
Remote  
host up  
?
1
?
no  
yes  
C
E
Bring up  
remote host  
1
no  
D
F
Entry  
complete  
Use arp to  
complete entry  
?
yes  
ping local host  
1
Chapter 6  
79  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 6 P r oced u r es  
A.  
Host en tr y in ARP ca ch e? Using arp, check that an  
entry exists for the remote host in your system's ARP  
cache. For example:  
arp spiff  
B.  
Rem ote h ost u p ? If there is no ARP cache entry for  
the remote host, first check that the remote host is up.  
If not, the remote host has not broadcast an ARP  
message, and that probably is why there is no entry in  
the ARP cache.  
C.  
D.  
E.  
Br in g-u p r em ote h ost. Have the node manager of the  
remote host bring that system up and start again with  
flowchart 1.  
En tr y com p lete? Perhaps there is an ARP cache  
entry, but it is wrong or not complete. If the entry is  
complete, go to step F.  
Use a r p to com p lete en tr y. Using arp, enter the  
correct Station Address. For more information, refer to  
the arp(1M) online man page. Start again with  
flowchart 1.  
F.  
p in g loca l h ost. Using ping, do an internal loopback  
on your own system. In other words, ping your own  
system.  
If the internal loopback is successful, your system is  
operating properly to the Network Layer (OSI Layer 3).  
In addition, you know an ARP cache entry for the  
remote host exists on your system. Start again with  
Flowchart 1.  
80  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 7: Lin k Level Loop ba ck Test  
Figu r e 6-8  
F low ch a r t 7: Lin k Level Loop ba ck Test  
7
A
B
yes  
Execute: linkloop  
to remote host  
linkloop  
successful  
?
1
no  
C
D
E
Loopback FAILED:  
Not an individual  
address  
Loopback FAILED:  
Address has bad  
format  
Loopback  
FAILED  
F
G
Choose a different  
remote host;  
re-execute  
Correct the link  
address parameter  
linkloop  
7
H
linkloop  
6
successful  
no  
?
yes  
I
Check remote host’s  
connectivity to  
100Base-TX  
1
Chapter 6  
81  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 7 P r oced u r es  
A.  
Execu te: lin k loop to r em ote h ost. Enter the NMID  
of your 10/100Base-TX card and link level address  
(station address) of the remote host in hexadecimal  
form (preceded by 0x). Execute lanscan (1M) on the  
local system to find the NMID and obtain the link level  
address (station address) of the remote host. For more  
information on linkloop, refer to the linkloop(1M)  
online man page.  
B.  
lin k loop su ccessfu l? If the test was successful, go to  
flowchart 1 to verify that the problem is solved.  
Network connectivity is o.k. through the Link Layer  
(OSI Layer 2). If not successful, note which error was  
returned and continue with this flowchart.  
C.  
D.  
Loop ba ck fa iled : Ad d r ess h a s ba d for m a t. The  
link level address is not correct. Go to F.  
Loop ba ck fa iled : Not a n in d ivid u a l a d d r ess. The  
link level address is not correct. The first hexadecimal  
digit has its high order bit set (if the value is equal to or  
greater than 8, it is set). This means it is a multicast or  
broadcast address, which is not allowed. The address  
must be unique to one remote host. Go to F.  
E.  
F.  
Loop ba ck fa iled . The remote host did not respond.  
Go to G.  
Cor r ect th e lin k a d d r ess p a r a m eter. Change the  
link level address to an allowed value and start again  
with flowchart 7.  
G.  
H.  
Ch oose a d iffer en t r em ote h ost; r e-execu te  
lin k loop . Restart flowchart 7 using a different remote  
host.  
lin k loop su ccessfu l? If the test was successful, go to  
step I. Network connectivity is o.k. through the Link  
Layer (OSI Layer 2). If not successful, the problem may  
be with the remote system. Go to flowchart 6.  
I.  
Ch eck r em ote h ost's con n ectivity to  
10/100Ba se-TX. Contact the node manager of the  
remote host. Check that the host is configured correctly  
and that its network interface is up. If necessary, use  
82  
Chapter6  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
flowchart 1 to verify configuration of the remote host.  
Chapter 6  
83  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 8: Tr a n sp or t Level Loop ba ck Test (u sin g  
ARPA)  
Figu r e 6-9  
F low ch a r t 8: Tr a n sp or t Level Loop ba ck Test (u sin g ARPA)  
8
A
Execute: telnet to  
remote host  
B
yes  
Successful  
Stop  
?
no  
C
Execute: ftp to  
remote host  
D
E
yes  
Successful  
?
Call HP  
no  
TCP  
not configured  
on local or  
remote  
G
yes  
Configure  
TCP  
8
host  
?
no  
F
yes  
Network  
congested  
?
Call HP  
no  
Call HP  
84  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 8 P r oced u r es  
A.  
Execu te: teln et to r em ote h ost. Try to establish a  
telnet connection to the remote host.  
B.  
Su ccessfu l? If your telnet attempt was successful,  
stop. The connection is o.k. through the Transport  
Layer (OSI Layer 4).  
C.  
Execu te: ftp to r em ote h ost. Unlike telnet, ftp does  
not go through a pseudoterminal driver (pty) on your  
system. This step tests to see if the pty is why telnet  
failed.  
D.  
E.  
Su ccessfu l? If ftp is successful, you likely have a  
problem with a pty on your system. Contact your HP  
representative.  
TCP n ot con figu r ed on loca l n or r em ote h ost?  
Neither telnet or ftp will work if TCP is not configured  
on either side of the connection. Check the  
/ etc/ protocols file on both hosts to be sure TCP is  
installed and configured.  
F.  
Netw or k con gested ? If TCP is installed on both  
hosts, do a file transfer to another remote host on the  
network. Use netstat(1) to check for lost packets.  
If network congestion is not the cause, more detailed  
diagnostics are required. Again, contact your HP  
representative.  
G.  
Con figu r e TCP. If necessary, install TCP on either or  
both hosts. Start again with this flowchart.  
Chapter 6  
85  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 9: Br id ge/Ga tew a y Loop ba ck Test  
Figu r e 6-10  
F low ch a r t 9: Br id ge a n d Ga tew a y Loop ba ck Test  
9
A
C
Execute: ping from  
known good host  
through gateway to  
known good host  
B
yes  
Check route table  
on problem host and  
all hosts between  
Successful  
?
no  
D
E
Examine  
gateway  
Correct route  
tables  
1
G
F
Non-HP 9000 or  
other vendors.  
Refer to networking  
documentation  
If HP 9000  
execute: ifconfig  
on gateway host  
H
yes  
Network  
interface up  
3
?
no  
I
Configure interface  
up  
1
86  
Chapter6  
 
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
F low ch a r t 9 P r oced u r es  
A.  
B.  
C.  
Execu te: p in g fr om k n ow n good h ost th r ou gh  
ga tew a y to k n ow n good r em ote h ost. This will test  
gateway connectivity to the remote network.  
Su ccessfu l? If the executing ping returned  
successfully, the problem may exist in the routing table  
for the problem host. Go to C.  
Ch eck r ou te ta ble on p r oblem h ost a n d a ll h osts  
in betw een . Execute netstat -r to examine a route  
table.  
D.  
E.  
Exa m in e ga tew a y. If the gateway is an HP 9000, go  
to G. If it is not, go to F.  
Cor r ect r ou te ta bles. Ensure that the proper  
IP/Internet addresses are assigned in the Destination  
and Gateway fields. If you are using subnetting, make  
sure that the destination is what you expect: a network  
or a host. Go to flowchart 1 to verify that the problem is  
solved.  
F.  
Non -HP 9000 or oth er ven d or s. Refer to  
n etw or k in g d ocu m en ta tion . Refer to the  
documentation that came with the gateway for  
additional diagnostics.  
G.  
H.  
If HP 9000, execu te ifcon fig on ga tew a y h ost.  
Execute ifconfig for all network interfaces on the  
gateway.  
Netw or k in ter fa ce u p ? If the output from ifconfig  
does not include the UP parameter, the network  
interface is down. Execute netstat -i to check the status  
of the network interfaces. An asterisk (*) indicates that  
the interface is down. If the network interface is down,  
go to I.  
If the network interfaces are UP, start again with  
flowchart 3. Using flowchart 3, test all network  
interfaces on the gateway.  
I.  
Con figu r e in ter fa ce u p . Execute ifconfig on each  
interface to bring it up. Start again with flowchart 1.  
Using flowchart 1, test all network interfaces on the  
Chapter 6  
87  
Troubleshooting 10/100Base-TX/9000  
Diagnostic Flowcharts  
gateway.  
88  
Chapter6  
A
10/100Ba se-TX In ter fa ce Ca r d  
Sta tistics  
89  
 
10/100Base-TX Interface Card Statistics  
LAN Interface Status Display  
LAN In ter fa ce Sta tu s Disp la y  
This appendix contains descriptions of the RFC 1213 MIB II statistics  
fields for LAN interface cards which are displayed on the screen with the  
display command in lanadmin LAN Interface Test Mode. A description  
of each field follows the display.  
LAN INTERFACE STATUS DISPLAY  
Tue, Nov 30,1999 11:45:17  
Network Management ID  
Description  
= 5  
= btlan Hewlett-Packard  
10/100Base-TX Full-Duplex  
Hw Rev 0  
Type (value)  
MTU Size  
= ethernet-csmacd(6)  
= 1500  
Speed  
= 100000000  
Station Address  
= 0x80009d40d69  
Administration Status (value) = up(1)  
Operation Status (value)  
Last Change  
= down(2)  
= 0  
Inbound Octets  
= 0  
Inbound Unicast Packets  
Inbound Non-Unicast Packets  
Inbound Discards  
= 0  
= 0  
= 0  
Inbound Errors  
= 0  
Inbound Unknown Protocols  
Outbound Octets  
= 0  
= 0  
Outbound Unicast Packets  
Outbound Non-Unicast Packets  
Outbound Discards  
= 0  
= 0  
= 0  
Outbound Errors  
= 0  
Outbound Queue Length  
Specific  
= 0  
= 655367  
Ethernet-like Statistics Group  
Index  
Alignment Errors  
FCS Errors  
Single Collision Frames  
Multiple Collision Frames  
= 3  
= 0  
= 0  
= 0  
= 0  
90  
AppendixA  
 
10/100Base-TX Interface Card Statistics  
LAN Interface Status Display  
Deferred Transmissions  
Late Collisions  
Excessive Collisions  
Internal MAC Transmit Errors  
Carrier Sense Errors  
Frames Too Long  
= 0  
= 0  
= 0  
= 0  
= 0  
= 0  
= 0  
Internal MAC Receive Errors  
Appendix A  
91  
10/100Base-TX Interface Card Statistics  
RFC 1213 MIB II  
RF C 1213 MIB II  
For more detailed information about the fields described below, refer to  
RFC 1213.  
Field  
Description  
Network Management ID A unique ID assigned by the system for the  
network management of each network interface.  
Description  
A textual string containing information about the  
interface.  
Type (value)  
The type of interface, distinguished according to the  
physical/link protocols, immediately below the network  
layer in the protocol stack.  
10/100Base-TX can have one of the following values:  
ethernet-csmacd(6), or iso88023-csmacd(7).  
The following values are for other networking products.  
MTU Size  
The size of the largest datagram which can be  
sent/received on the interface specified in octets. This  
value is 1500.  
Speed in bits per second The speed of the 10/100Base-TX card, 10 Mbit/s  
or 100 Mbit/s.  
Station Address The interface address at the protocol layer immediately  
below the network layer in the protocol stack. For  
interfaces which do not have such an address, such as  
serial line, this object contains an octet string of zero  
length.  
Administration Status The desired state of the interface. This  
parameter is set to up(1)and is not configurable. It  
will have one of the following values:  
up(1)  
Ready to pass packets  
Not operative  
down(2)  
testing(3)  
In test mode  
92  
AppendixA  
 
10/100Base-TX Interface Card Statistics  
RFC 1213 MIB II  
Operation Status The current operational state of the interface. This  
value is the same as the hardware status displayed by  
lanscan(1M). It will have one of the following values.  
up(1)  
Ready to pass packets  
Not operative (card is down)  
In test mode  
down(2)  
testing(3)  
Last Change  
The value of SysUpTime at the time the interface  
entered its current operational state. If the current  
state was entered prior to the last reinitialization of the  
local network management subsystem, then this object  
contains a zero value.  
Inbound Octets The total number of octets received on the interface,  
including framing characters.  
Inbound Unicast Packets The number of subnetwork-unicast packets  
delivered to a high-layer protocol.  
Inbound Non-Unicast Packets The number of non-unicast  
(subnetwork-broadcast or subnetwork-multicast)  
packets delivered to a higher-layer protocol.  
Inbound Discards The number of inbound packets that were discarded  
even though no errors had been detected, to prevent  
their being delivered to a higher-layer protocol. One  
possible reason for discarding such a packet could be to  
free up buffer space.  
Inbound Errors The number of inbound packets that contained errors  
preventing them from being deliverable to a  
higher-layer protocol.  
Inbound Unknown Protocols The number of packets received via the  
interface which were discarded because of an unknown  
or unsupported protocol.  
Outbound Octets The total number of octets transmitted out of the  
interface, including framing characters.  
Outbound Unicast Packets The total number of packets that  
higher-level protocols requested be transmitted to a  
subnetwork-unicast address, including those that were  
discarded or not sent.  
Appendix A  
93  
10/100Base-TX Interface Card Statistics  
RFC 1213 MIB II  
Outbound Non-Unicast Packets The total number of packets that  
higher-level protocols requested be transmitted to a  
non-unicast (a subnetwork-broadcast or  
subnetwork-multicast) address, including those that  
were discarded or not sent.  
Outbound Discards The number of outbound packets that were  
discarded even though no errors had been detected to  
prevent their being transmitted. One possible reason  
for discarding such a packet could be to free up buffer  
space.  
Outbound Errors The number of outbound packets that could not be  
transmitted because of errors.  
Outbound Queue Length The length of the output packet queue (in  
packets).  
94  
AppendixA  
10/100Base-TX Interface Card Statistics  
RFC 1284 Ethernet-Like Interface Statistics  
RF C 1284 Eth er n et-Lik e In ter fa ce Sta tistics  
Field  
Description  
Index  
A value that uniquely identifies an interface to an  
802.3 medium.  
Alignment Errors A count of frames received on a particular interface  
that are not an integral number of octets in length and  
do not pass the FCS check.  
FCS Errors  
A count of frames received on a particular interface  
that are not an integral number of octets in length and  
do not pass the FCS check.  
Single Collision Frames A count of successfully transmitted frames on a  
particular interface for which transmission is inhibited  
by exactly one collision.  
Multiple Collision Frames A count of successfully transmitted frames on  
a particular interface for which transmission is  
inhibited by more than one collision.  
Deferred Transmissions A count of frames for which the first  
transmission attempt on a particular interface is  
delayed because the medium is busy. The count  
represented by an instance of this object does not  
include frames involved in collisions.  
Late Collisions The number of times that a collision is detected on a  
particular interface later than 512 bit-times into the  
transmission of a packet.  
Excessive Collisions A couple of frames for which transmission on a  
particular interface fails due to excessive collisions in  
10-Base-T mode. For 100Base-TX mode, excessive  
collisions indicate the number of packets dropped.  
Internal MAC Transmit Errors A count of frames for which  
transmission on a particular interface fails due to an  
internal MAC sublayer transmit error.  
Carrier Sense Errors The number of times that the carrier sense  
Appendix A  
95  
 
10/100Base-TX Interface Card Statistics  
RFC 1284 Ethernet-Like Interface Statistics  
condition was lost or never asserted when attempting  
to transmit a frame on a particular interface.  
Frames Too Long A count of frames received on a particular interface  
that exceed the maximum permitted framer size.  
Internal MAC Receive Errors A count of frames for which reception on a  
particular interface fails due to an internal MAC  
sublayer receive error.  
96  
AppendixA  
10/100Base-TX Interface Card Statistics  
Create a Record or Map of Your Internetwork  
Cr ea te a Recor d or Ma p of You r In ter n etw or k  
Be sure to create or update a record of your network and internetwork  
before attempting 10/100Base-TX installation. You may wish to create a  
map showing how pieces of your internetwork are related. Your records  
should include:  
Approximate dimensions of the building or room containing the  
10/100Base-TX network.  
Location of, routers, bridges, and gateways  
Location of nodes and node connections.  
Location of network segments and subnets within each segment  
Hostname of each node.  
Internet Address and Alias of each node (in the case of gateways, each  
10/100Base-TX card has its own Internet Address and Alias).  
Hardware Path of each card in the system including 10/100Base-TX  
cards. You can use this information as part of a disaster recovery  
plan.  
Version number of the operating system installed on each node.  
Appendix A  
97  
 
10/100Base-TX Interface Card Statistics  
Create a Record or Map of Your Internetwork  
98  
AppendixA  
B
Ha r d w a r e Refer en ce In for m a tion  
This appendix contains information about the card LEDs, cabling  
specifications and card specifications.  
99  
 
Hardware Reference Information  
Basic Troubleshooting Tips  
Ba sic Tr ou blesh ootin g Tip s  
Listed below are some tips on troubleshooting common hardware problems.  
Refer to this information when you are trying to identify 10/100Base-TX  
hardware problems.  
Check the network cables. Make sure the network cable connections are  
secure and that the cables are not damaged. If you find any connections  
that are loose, or cables that are damaged, fix the problem and then see if  
your computer can communicate on the network  
Check the Link LED (10/00Base-TX) on the card bulkhead. If the LED is  
OFF, or all LEDs on the card bulkhead are ON, then, at the HP-UX  
command line, type: dmesgand view the output on your screen to see if any  
error messages exist.  
The possible causes of a fault condition could be:  
Defective cable  
Cable not connected to active hub or switch  
Defective card  
100  
AppendixB  
 
Hardware Reference Information  
Connector Information  
Con n ector In for m a tion  
This section includes pin usage information for the RJ -45 twisted pair  
connector. Connectors on LAN adapters adhere to appropriate standards  
agreed upon by various standards bodies and are widely available.  
Incorrectly wired or installed cabling is the most common cause of  
communications problems for local area networks. HP recommends that you  
work with a qualified cable installer for assistance in your cabling  
requirements.  
CAUTION  
The unshielded twisted-pair cables you use with the PCI 10/100Base-TX card  
must comply with the IEEE 802.3u 100Base-TX standards in order to meet  
emissions requirements. These standards support cabling up to 100 meters  
only.  
P CI 10/100Ba se-TX Ca r d Tw isted -Pa ir Con n ector  
The same connector on the card is used for either 10Base-T or 100Base-TX  
operation. The operating mode is determined by the setting of the hub or  
switch to which the card is connected.  
Con n ector P in Usa ge for 10-Mbit/s Tw isted -Pa ir  
Con n ector  
Use unshielded twisted-pair cables that comply with the IEEE 802.3 Type  
10Base-T standard.  
Ta ble B-1  
IEEE 802.3 Typ e 10Ba se-T Sta n d a r d  
Pins  
Signal  
1
2
(transmit +)  
(transmit -)  
3
6
(receive +)  
(receive -)  
Appendix B  
101  
     
Hardware Reference Information  
Connector Information  
Figu r e B-1  
P in La you t of RJ -45 Con n ector on P CI Ca r d  
8
1
Available HP Cables:  
HP 92268A - 4 meter with attached 8-pin connectors.  
HP92268B - 8-meter with attached 8-pin connectors.  
HP 92268C - 16-meter with attached 8-pin connectors.  
HP 92268D - 32-meter with attached 8-pin connectors.  
HP 92268N- 300-meter (no connectors supplied).  
Con n ector P in Usa ge for 100-Mbit/s Tw isted -Pa ir  
Con n ector  
Use Category 5 unshielded twisted-pair cables that comply with the IEEE  
802.3u 100Base-TX standard.  
Ta ble B-2  
IEEE 802.3u 100Ba se-TX Sta n d a r d  
Signal TX End  
Pins  
Node  
1
2
TX:1+  
TX:1-  
3
6
RX:1+  
RX:1-  
102  
AppendixB  
 
Hardware Reference Information  
Connector Information  
Ta ble B-2  
IEEE 802.3u 100Ba se-TX Sta n d a r d  
Signal TX End  
Pins  
Node  
4
5
Unused  
Unused  
7
8
Unused  
Unused  
Available HP Cables:  
HP 92268A - 4 meter with attached 8-pin connectors.  
HP92268B - 8-meter with attached 8-pin connectors.  
HP 92268C - 16-meter with attached 8-pin connectors.  
HP 92268D - 32-meter with attached 8-pin connectors.  
HP 92268N- 300-meter (no connectors supplied).  
Ca ble Len gth s:  
The maximum length of the cable from the hub to each node for 100Base-TX is  
100 meters. The cable must be category 5 UTP for 100Base-TX operation. For  
additional information on cable lengths and number of nodes supported by  
10/100Base-TX, refer to the IEEE 802.3u Specification and Fast Ethernet,  
Dawn of a New Network by Howard W. Johnson (published 1996 by Prentice  
Hall PTR, Upper Saddle River, New Jersey 07458. Phone 800-382-3419. The  
ISBN number is 0-13-352643-7).  
Appendix B  
103  
Hardware Reference Information  
Cabling Information  
Ca blin g In for m a tion  
Connectors on LAN adapters adhere to appropriate standards agreed upon by  
various standards bodies and are widely available.  
Incorrectly wired or installed cabling is the most common cause of  
communications problems for local area networks. HP recommends that you  
work with a qualified cable installer for assistance in your cabling  
requirements.  
Ca ble Len gth s  
Following are the maximum cable lengths from the switch to each node.  
P CI 10/100Ba se-T  
Ta ble B-3  
Op er a tin g Dista n ces for Va r iou s Ca ble Typ es — 10/100Ba se-TX  
Cable Description  
Operating Distance  
10/100Base-TX/9000  
Cat 5 or Cat 5E UTP  
100 meters  
104  
AppendixB  
   
Hardware Reference Information  
Dual Port 100Base-TX and Dual Port Wide Ultra 2 SCSI Specifications  
Du a l Por t 100Ba se-TX a n d Du a l Por t Wid e  
Ultr a 2 SCSI Sp ecifica tion s  
A5838A  
Sp ecifica tion s  
P h ysica l  
Dimensions:  
7.9 in by 4.2 in  
+15 watts max  
Electr ica l  
Power requirement (:  
En vir on m en ta l  
Temperature  
Degrees F = (1.8 x Degrees C) + 32  
o
o
Operating Temperature:  
Storage Temperature:  
+5 C to 40 C  
o
o
-40 C to 70 C  
o
o
Recommended Operating  
Temperature:  
+20 C to 30 C  
Hu m id ity  
Operating Relative humidity  
15 to 80% non-condensing  
5 to 90% non-condensing  
o
range@ 22 C  
Non-operating/storage  
Relative humidity:  
Altitu d e  
Operating:  
10,000 ft (3.1KM)  
15,000 ft (4.6KM  
Non-operating:  
Electr om a gn etic Com p a tibility  
FCC Class A  
USA  
CISPR-22/EN55022 Class A  
EN50082-1  
International and Europe  
Europe  
For compliance to European directives and related specifications, see the  
Declaration of Conformity statement in Appendix C.  
Appendix B  
105  
 
Hardware Reference Information  
Dual Port 100Base-TX and Dual Port Wide Ultra 2 SCSI Specifications  
Ca ble In ter fa ces  
The 10/100Base-TX ports are compatible with IEEE 802.3u standard and  
use RJ -45 connectors.  
Com m u n ica tion s Sta n d a r d s  
The physical layer of IEEE 802.3ustandard supports Cat 5 or Cat 5E UTP  
cables.  
106  
AppendixB  
C
Ha r d w a r e Regu la tor y Sta tem en ts  
This section contains hardware regulatory statements for the Dual Port  
100Base-TX and Dual Port Ultra2 SCSI product used in the United States,  
Canada, and the European community. Refer to your Dual Port 100Base-TX  
and Dual Port Ultra2 SCSI Quick Installation card for product installation  
instructions.  
107  
 
Hardware Regulatory Statements  
FCC Statement (For U.S.A.)  
F CC Sta tem en t (For U.S.A.)  
Fed er a l Com m u n ica tion s Com m ission Ra d io Fr equ en cy In ter fer en ce  
Sta tem en t  
WARNING  
Th is d evice com p lies w ith Pa r t 15 of th e F CC r u les. Op er a tion is  
su bject to th e follow in g tw o con d ition s:  
(1) Th is d evice m a y n ot ca u se h a r m fu l in ter fer en ce a n d  
(2) th is d evice m u st a ccep t a n y in ter fer en ce r eceived , in clu d in g  
in ter fer en ce th a t m igh t ca u se u n d esir ed op er a tion .  
Th is equ ip m en t h a s been tested a n d fou n d to com p ly w ith th e lim its  
for a Cla ss A d igita l d evice, p u r su a n t to Pa r t 15 of th e F CC r u les.  
Th ese lim its a r e d esign ed to p r ovid e r ea son a ble p r otection a ga in st  
h a r m fu l in ter fer en ce w h en th e equ ip m en t is op er a ted in a  
com m er cia l en vir on m en t. Th is equ ip m en t gen er a tes, u ses a n d ca n  
r a d ia te r a d io fr equ en cy en er gy, a n d , if n ot in sta lled a n d u sed in  
a ccor d a n ce w ith th e in str u ction m a n u a l, m a y ca u se h a r m fu l  
in ter fer en ce to r a d io com m u n ica tion s.  
Op er a tion of th is equ ip m en t in a r esid en tia l a r ea is lik ely to ca u se  
in ter fer en ce, in w h ich ca se th e u ser a t h is ow n exp en se w ill be  
r equ ir ed to ta k e w h a tever m ea su r es m a y be r equ ir ed to cor r ect th e  
in ter fer en ce.  
Hew lett-Pa ck a r d s system cer tifica tion tests w er e con d u cted w ith HP -  
su p p or ted p er ip h er a l d evices a n d ca bles, su ch a s th ose r eceived w ith  
you r system . Ch a n ges or m od ifica tion s to th is equ ip m en t n ot  
exp r essly a p p r oved by Hew lett-Pa ck a r d cou ld void th e u ser s  
a u th or ity to op er a te th e equ ip m en t.  
Ca n a d a  
WARNING  
Th is Cla ss A d igita l a p p a r a tu s m eets a ll r equ ir em en ts of th e Ca n a d ia n  
In ter fer en ce-Ca u sin g Equ ip m en t Regu la tion s.  
Cet a p p a r eil n u m ér iqu e d e la cla sse A r esp ecte tou tes les exigen ces d u  
r èglem en t su r le m a tér iel br ou illeu r d u Ca n a d a .  
108  
AppendixC  
   
Hardware Regulatory Statements  
EMI Statement (European Community)  
EMI Sta tem en t (Eu r op ea n Com m u n ity)  
NOTE  
This is a Class A product. In a domestic environment, this product may cause  
radio interference, in which case you may be required to take adequate  
measures.  
Declaration of Conformity: The following Declaration of Conformity has been  
issued per ISO/IEC Guide 22 and EN 45014 and identifies the product, the  
manufacturers name and address, and the applicable specifications that are  
recognized in the European community  
.
Appendix C  
109  
 
Hardware Regulatory Statements  
EMI Statement (European Community)  
110  
AppendixC  
Glossa r y  
10Ba se-T: A 10 Mbit/s communication method specified in the IEEE  
802.3u-1995 standard.  
100Ba se-T: A 100 Mbit/s communication method specified in the IEEE  
802.3u-1995 standard. The official name for Fast Ethernet.  
100Ba se-TX: A specific implementation of 100Base-T designed to  
operate over Category 5 UTP cabling.  
Ad d r ess: A specific location in memory, designated either numerically or  
by a symbolic name.  
Alia s: Name of the interface that corresponds to a given Internet  
address on a system. Refer to the network map in appendix B for  
example usage.  
Asyn ch r on ou s Da ta Tr a n sfer : One of the ways data is transferred  
over the SCSI bus. It is slower than synchronous data transfer.  
Au ton egotia tion : A mechanism defined in IEEE 802.3u-1995 whereby  
devices sharing a link segment can exchange data and automatically  
configure themselves to operate at the highest capability mode shared  
between them.  
Au tosen sin g: The ability of the 10/100Base-TX card to detect a static  
speed of a hub or switch and automatically configure itself to operate  
accordingly. This does not require the two-way information exchange and  
negotiation process of full autonegotiation.  
BIOS (Ba sic In p u t/Ou tp u t System ): Software that provides basic  
read/write capability. Usually kept as firmware (ROM based). The  
system BIOS on the main board of a computer is used to boot and control  
the system. The SCSI BIOS on the host adapter acts as an extension of  
the system BIOS.  
Bit A binary digit. The smallest unit of information a computer uses. The  
value of a bit (0 or 1) represents a two-way choice, such as on or off, true  
or false.  
Bu s A collection of wires in a cable or copper traces on a circuit board  
111  
 
used to transmit data, status, and control signals. EISA, PCI, and SCSI  
are examples of buses.  
Bus Mastering A high-performance way to transfer data. The host  
adapter controls the transfer of data directly to and from system memory  
without bothering the computers microprocessor. This is the fastest way  
for multi-tasking operating systems to transfer data.  
Byte A unit of information consisting of eight bits.  
CSMA/CD: Carrier sense multiple access with collision detection. The  
media access method implemented in IEEE 802.3u-1995.  
Ca r d In sta n ce Nu m ber : A number that uniquely identifies a device  
within a class. A class of devices is a logical grouping of similar devices.  
Ch a in A topology in which every device is connected to two others,  
except for two-end devices that are connected to only one other.  
CISPR (Committee, International and Special, for Protection in Radio) An  
international committee on radio frequency interference.  
Configuration Refers to the way a computer is set up; the combined hardware  
components (computer, monitor, key board, and peripheral devices) that make up a  
computer system; or the software settings that allow the hardware components to  
communicate with each other.  
CPU (central processing unit)  
The brain” of the computer that performs the actual computations. The  
term Micro Processor Unit (MPU) is also used.  
Destin a tion Ad d r ess: A field in the message packet format identifying  
the end node(s) to which the packet is being sent.  
Device Dr iver A program that allows a microprocessor (through the operating  
system) to direct the operation of a peripheral device.  
Differential A hardware configuration for connecting SCSI devices. It uses a pair of  
lines for each signal transfer (as opposed to single-ended SCSI which references  
each SCSI signal to a common ground).  
DLP I: Data Link Provider Interface. An industry-standard definition for  
112  
message communications to STREAMS-based network interface drivers.  
DMA (direct memory access) A method of moving data from a storage device  
directly to RAM, without using the CPU’s resources.  
DMA Bus Master A feature that allows a peripheral to control the flow of data to  
and from system memory by blocks, as opposed to PIO (Programmed I/O) where  
the flow is byte by byte.  
EEP ROM (electronically erasable programmable read-only memory) A  
memory chip typically used to store configuration information. See NVRAM.  
EISA (Extended Industry Standard Architecture) An extension of the 16-bit  
ISA bus standard. It allows devices to perform 32-bit data transfers.  
Eth er n et: A 10 Mbit/s LAN, developed by Digital Equipment  
Corporation, Intel, and Xerox Corporation, upon which the IEEE 802.3  
network is based.  
Exter n a l SCSI Device A SCSI device installed outside the computer cabinet.  
External SCSI devices are connected in a chain using shielded cables.  
Fa st Eth er n et: A commonly used name applied to 100Base-T.  
F CC Federal Communications Commission.  
File  
A named collection of information, usually stored on a disk.  
Firmware Software that is permanently stored in ROM. In the case of BIOS, it can  
be accessed during boot time without the aid of an operating or file system.  
F u ll-Du p lex Mod e: A mode of media utilization whereby data can flow  
in both directions simultaneously across the multiple wire pairs of a  
physical link. While full-duplex operation is not defined per se in the  
IEEE 802.3u-1995 specification, the specification does define a  
mechanism for this mode to be autonegotiated between devices on each  
end of a link. Full-duplex mode is typically found on switches.  
HSC: High speed connect bus.  
Ha lf-Du p lex Mod e: The media utilization mode of IEEE 802.3u-1995  
113  
networks whereby data can flow in only one direction at a time across the  
multiple wire pairs of a physical link.  
Ha r d Disk A rigid disk permanently sealed into a drive cartridge. A hard disk can  
store very large amounts of information.  
Ha r d w a r e Pa th : An identifier assigned by the system according to the  
physical location (slot) of the card in the hardware backplane. On Series  
800 systems, the I/O subsystem identifies each LAN card by its hardware  
path.  
Host The computer system in which a SCSI host adapter is installed. It uses the  
SCSI host adapter to transfer information to and from devices attached to the SCSI  
bus.  
Host Adapter A circuit board and/or integrated circuit device that provides a SCSI  
bus connection to the computer system.  
Hostn a m e: Name of system on the network. Refer to the network map  
in appendix B for example usage.  
Hu b: A network interconnection device that allows multiple devices to  
share a single logical link segment. Hubs are generally either 10 Mbit/s  
or 100 Mbit/s devices. Use either a 10Base-T or 100Base-TX hub with the  
10/100Base-TX card.  
IEEE: The Institute of Electrical and Electronics Engineers. A national  
association, whose activities include publishing standards applicable to  
various electronic technologies. The IEEE technical committees are  
numbered and grouped by area. For example, the 800 committees study  
local area network technologies. The 802.3 committee produced the  
standard for a CSMA/CD local area network, which has been adopted by  
ANSI.  
IEEE 802.3u -1995 n etw or k : A 10 or 100 megabit-per-second LAN,  
specified in the IEEE 802.3u-1995 Standard for Local Area Networks. It  
uses the Carrier Sense Multiple Access/Collision Detection (CSMA/CD)  
network access method to give every node equal access to the network.  
In ter n a l SCSI Device A SCSI device installed inside the computer cabinet.  
These devices are connected in a chain using an unshielded ribbon cable.  
114  
In ter n et Ad d r ess: The network address of a computer node. This  
address identifies both which network the host is on and which host it is.  
Refer to the Installing and Administering LAN/ 9000 Software manual  
for detailed information about network addressing.  
IP Ad d r ess: See Internet Address glossary entry.  
IRQ (in ter r u p t r equ est ch a n n el) A path through which a device can  
get the immediate attention of the computers CPU. The PCI bus assigns  
an IRQ path for each SCSI host adapter.  
ISA (Industry Standard Architecture) A type of computer bus used in most PCs.  
It allows devices to send and receive data 16 bits at a time.  
KByte (k ilobyte) A measure of computer storage equal to 1024 bytes.  
LAN: See Local Area Network.  
Loca l Ar ea Netw or k (LAN): A data communications system that  
allows a number of independent devices to communicate with each other.  
Loca l Bu s A way to connect peripherals directly to the computer processor’s data  
path. It bypasses the slower ISA and EISA buses. PCI is a local bus standard.  
Loca l Netw or k : The network to which a node is directly attached.  
Logica l Un it A subdivision, either logical or physical, of a SCSI device. Most  
devices have only one logical unit, but up to sixteen are allowed for a 16-bit SCSI  
bus and eight for an 8-bit SCSI bus.  
LUN (logical unit number) An encoded three-bit number for the logical unit.  
LVD (low-voltage differential) A robust design methodology that improves power  
consumption, data integrity, cable lengths, and support for multiple devices while  
providing a migration path for increased I/O performance.  
Ma in boa r d A large circuit board that holds RAM, ROM, the microprocessor,  
custom integrated circuits, and other components that make a computer work. It also  
has expansion slots for host adapters and other plug-in boards.  
Main Memory The part of a computer’s memory that is directly accessible by the  
CPU (usually synonymous with RAM).  
115  
Ma jor Nu m ber : Unique value that identifies an individual hardware  
device.The number for the 10/100Base-TX card floats.  
Ma xim u m Tr a n sm ission Un it (MTU). Largest amount of data that  
can be transmitted through that interface. This value does not include  
the LLC or MAC headers.  
MByte (m ega byte) A measure of computer storage equal to 1024 kilobytes.  
Moth er boa r d See mainboard. In some countries, the term motherboard is not  
appropriate.  
Multi-tasking The initiation and control of more than one sequence of operations.  
This allows programs to operate in parallel.  
Multi-threading The simultaneous accessing of data by more than one SCSI  
device. This increases the aggregate data throughput.  
Netw or k In ter fa ce: A communication path through which messages  
can be sent and received. A hardware network interface has a hardware  
device associated with it, such as a LAN or FDDI card. A software  
network interface does not include a hardware device, for example the  
loopback interface. For every IP address instance, there must be one  
network interface configured.  
Netw or k Ma n a gem en t Id en tifier (NMID): On HP-UX 10.x, it was a  
unique ID assigned by the system for the network management of each  
network interface. Replaced on HP-UX 11.x by the PPA or physical point  
of attachment  
Nod e: Any point in a network where services are provided or  
communications channels are interconnected. A node could be a  
workstation or a server processor.  
NVRAM (Non -Vola tile Ra n d om Access Mem or y) An EEPROM  
(Electronically Erasable Read Only Memory chip) used to store configuration  
information.  
Operating System A program that organizes the internal activities of the computer  
and its peripheral devices. An operating system performs basic tasks such as moving  
data to and from devices, and managing information in memory. It also provides the  
user interface.  
116  
Pa ck et: A sequence of binary digits that is transmitted as a unit in a  
computer network. A packet usually contains control information plus  
data.  
Pa r ity Ch eck in g A way to verify the accuracy of data transmitted over the SCSI  
bus. One bit in the transfer is used to make the sum of all the 1 bits either odd or  
even (for odd or even parity). If the sum is not correct, an error message appears.  
SCSI uses odd parity.  
P CI: Peripheral component interconnect. A local bus specification that  
allows connection of integrated peripheral controller components,  
peripheral add-in boards, and processor/memory systems. It bypasses  
the slower ISA and EISA busses.  
Per ip h er a l Devices A hardware device (such as a video monitor, disk drive,  
printer, or CD-ROM) used with a computer and under the computer’s control. SCSI  
peripherals are controlled through a SCSI host adapter.  
Pin-1 Orientation The alignment of pin 1 on a SCSI cable connector and the pin 1  
position on the SCSI connector into which it is inserted. External SCSI cables are  
keyed to ensure proper alignment, but internal SCSI ribbon cables may not be.  
PIO (programmed input/output) A way the CPU can transfer data to and from  
memory via the computer’s I/O ports. PIO can be faster than DMA, but requires  
CPU time.  
Por t Ad d r ess Also Port Number. The address through which commands are sent  
to a host adapter board. This address is assigned by the PCI bus.  
Port Number See Port Address.  
P PA: Physical point of attachment.  
P r otocol: A specification for coding messages exchanged between two  
communications processes.  
Qu eu e Ta gs A way to keep track of multiple commands while allowing increased  
throughput on the SCSI bus.  
RAM (Random Access Memory) Generally, the computer’s primary working  
memory in which program instructions and data are stored and are accessible to the  
CPU. Information can be written to and read from RAM. The contents of RAM are  
117  
lost when the computer is turned off.  
RISC Core Symbios SCSI chips contain a RISC (Reduced Instruction Set  
Computer) processor, programmed through microcode scripts.  
RJ -45: The name for the connector type used with UTP cabling.  
ROM (Rea d -On ly Mem or y) Memory from which information can be read but  
not changed. The contents of ROM are not erased when the computer is turned off.  
SCAM (SCSI Configured AutoMatically) A method to automatically allocate  
SCSI IDs via software when SCAM compliant SCSI devices are attached.  
SCSI (small computer system interface) A specification for a high-performance  
peripheral bus and command set. The original standard is now referred to as SCSI-1.  
SCSI-2 The current SCSI specification that adds features to the original SCSI-1  
standard.  
SCSI-3 The next SCSI specification, that adds features to the SCSI-2 standard.  
SCSI Bus A host adapter and one or more SCSI peripherals connected by cables in  
a linear chain configuration. The host adapter may exist anywhere on the chain,  
allowing connection of both internal and external SCSI devices. A system may have  
more than one SCSI bus by using multiple host adapters.  
SCSI Device Any device conforming to the SCSI standard that attaches to the SCSI  
bus by means of a SCSI cable. This includes SCSI host adapters and SCSI  
peripherals.  
SCSI ID A unique identification for each SCSI device on the SCSI bus. Each SCSI  
bus has fifteen available SCSI IDs numbered 0 through 15 for Wide SCSI or 0-7 for  
8-bit SCSI. The host adapter is assigned ID 7, which gives it priority to control the  
bus.  
SDMS (SCSI Device Ma n a gem en t System ) A Symbios software product  
that manages SCSI system I/O.  
Single-Ended SCSI A hardware specification for connecting SCSI devices. It  
references each SCSI signal to a common ground, as opposed to differential SCSI  
and low-voltage differential SCSI, which use a separate return for each signal.  
Su bn etw or k : Small discrete physical networks connected via gateways  
118  
which share the same network address space. Refer to the Installing and  
Administering LAN/ 9000 Software manual for detailed information  
about subnetworks and subnet addressing.  
Su bn et m a sk : A 32-bit mask which, when AND'd with an internet  
address, determines a subnetwork address. When the internet address is  
AND'd with the subnet mask, the ones in the host portion of the subnet  
mask will overwrite” the corresponding bits of the host portion of the  
internet address, resulting in the subnet address. Refer to the Installing  
and Administering LAN/ 9000 Software manual for detailed information  
about subnet masks.  
Sw itch : A network interconnection device that allows multiple  
connected senders and receivers to communicate simultaneously in  
contrast to a hub (repeater) where only one device can send at a time.  
Some switches have fixed port speeds (10 Mbit/s or 100 Mbit/s) while  
others allow port speeds to be configured or autonegotiated.  
Syn ch r on ou s Da ta Tr a n sfer One of the ways data is transferred over  
the SCSI bus. Transfers are clocked with fixed-frequency pulses.  
System BIOS Controls the low level POST (Power On Self Test) and  
basic operation of the CPU and computer system.  
Ter m in a tion The electrical connection required at each end of the SCSI  
bus, composed of a set of resistors.  
Top ology: The physical and logical geometry governing placement of  
nodes in a computer network. Also, the layout of the transmission  
medium for a network.  
Ultr a SCSI A standard for SCSI data transfers. It allows a transfer rate of up to 20  
MBytes/sec over an 8-bit SCSI bus, and up to 40 MBytes/sec over a 16-bit SCSI  
bus. STA (SCSI Trade Association) supports using the term “Ultra SCSI” over the  
older term “Fast-20.”  
Ultra2 SCSI A standard for SCSI data transfers. It allows a transfer rate of up to 40  
MBytes/sec over an 8-bit SCSI bus, and up to 80 MBytes/sec over a 16-bit SCSI  
bus. STA (SCSI Trade Association) supports using the term “Ultra2 SCSI” over the  
older term “Fast-40.”  
UTP (Un sh ield ed Tw isted Pa ir ) Ca blin g: A data cable type  
119  
consisting of pairs of wires twisted together without an electrically  
shielding jacket.  
VCCI Voluntary Control Council for Interference.  
VHDCI Very High Density Cable Interconnect.  
Wide SCSI A SCSI-2 feature allowing 16 or 32-bit transfers on the SCSI bus. This  
dramatically increases the transfer rate over the standard 8-bit SCSI bus.  
Wide Ultra SCSI The SCSI Trade Association term for SCSI bus width 16 bits,  
SCSI bus speed maximum data rate 40 MBytes/sec.  
Wide Ultra2 SCSI The SCSI Trade Association term for SCSI bus width 16 bits,  
SCSI bus speed maximum data rate 80 MBytes/sec.  
120  

HP Hewlett Packard HP Ultra Slim LED LCD Monitor W2072B User Manual
HP Hewlett Packard DVD100I User Manual
Hotpoint 6166 User Manual
Gateway FPD2185W User Manual
Dell UltraSharp 1704FPV User Manual
Blodgett SHO E User Manual
Audiovox D1888 User Manual
AT T E5927B User Manual
Atlas ATM1900T User Manual
Asus Computer Monitor VH198 User Manual