|
|
The following is a strategy to use to identify and solve link
problems. You should use this strategy if you are not sure what
is causing the problem because many times errors in the upper level
software are due to hardware problems. You can also use this strategy
if you have identified a hardware-related possible cause and need
to isolate the actual cause.
LAN, Token Ring, FDDI, 100VG-AnyLAN, 100Base-T Link Problems
For problems that involve LAN, Token Ring, FDDI, 100VG-AnyLAN,
or 100Base-T link, use the following strategy where applicable:
Issue the LINKCONTROL linkname; STATUS=DIAGSTATS
command. Inspect the output and attempt to identify the problem.
Refer to the Online Diagnostic Subsystem Manual, Volume I, for
a detailed analysis of the fields displayed. Retain a copy of the
output from this command for your Hewlett-Packard
representative.
Run PING to confirm whether or not the remote node is
reachable. See Chapter 6 "Using NETTOOL" for
instructions on running PING.
If PING fails, use the LAN node diagnostic that is
appropriate for the type of card on your system:
- Card
Online Diagnostic
- LAN
LANDAD
LAN3PBB
CONSOLAN
- Token Ring
LAN5PBB
- FDDI
FDDIPBA
- 100VG-AnyLAN
VGPBA
- 100Base-T
VGPBA
Refer to your hardware documentation for information on these
diagnostics. These diagnostics are online tools that verify the
hardware components by running the self-test, then a series of tests
of the cables and connectors.
If a failure has taken place, give the files
NMLGxx.PUB.SYS and NETDMPnn.PUB.SYS
to your Hewlett-Packard representative for additional analysis.
If the problem is easily reproducible, and link level tracing was
inactive when the problem took place, turn on tracing using the
LINKCONTROL command. When the problem has been reproduced,
turn off trace and give this trace file to your Hewlett-Packard
representative for additional analysis. If a hardware failure takes
place while trace is active, give the files
NMLGxx.PUB.SYS and NETDMPnn.PUB.SYS
to your HP representative as well.
The log message contains an error code, such as an NMERR. Information
on the cause and recovery of these errors can be found in the NS
3000/iX Error Messages Reference Manual. Keep a copy of the log
file and the output. If you need to submit an CR, send the log file
and output to your Hewlett-Packard representative.
If link level logging is not enabled, enable it through NMMGR so
that the information will be available if this problem can be
repeated.
NS Point-to-Point 3000/iX Link Problems
The NS Point-to-Point 3000/iX link (router link) is connected
with a programmable serial interface (PSI) card. For problems that
involve the PSI, use the following strategy where applicable:
Issue the LINKCONTROL linkname; STATUS=DIAGSTATS
command. Inspect the output and attempt to identify the problem.
Refer to Appendix A "LINKCONTROL Command" for
a detailed analysis of the fields displayed. Retain a copy of the
output from this command for your Hewlett-Packard
representative.
Run PING to confirm whether or not the remote node is
reachable. See Chapter 6 "Using NETTOOL" for
instructions on running PING.
If PING fails, use PSIDAD. PSIDAD is an
on-line diagnostic tool. It verifies the PSI components by running
the PSI self-test, then extends the testing as far into the
communications network as possible, depending on which equipment is
connected to the PSI. Refer to the On-Line Diagnostic Subsystem
Manual, Volume I, for instructions.
If a PSI failure has taken place, give the files
NMLGxx.PUB.SYS and NETDMPnn.PUB.SYS
to your Hewlett-Packard representative for additional
analysis.
If the problem is easily reproducible, and link level tracing was
inactive when the problem took place, turn on tracing using the
LINKCONTROL command. When the problem has been reproduced,
turn off tracing. Save both the raw trace file and the formatted
output for your Hewlett-Packard representative for analysis. It is
important to save any PSI dump file (NETDMPnn.PUB.SYS)
that is created while link level tracing was enabled. Send both the
PSI dump file and the link trace file to your Hewlett-Packard
representative for additional analysis.
Check the MPE/iX log file for I/O error logging. Format the log
file. Keep a copy of the file and the output for your Hewlett-Packard
representative to study.
 |
NOTE: If you lose connections on an NS 3000/iX Point-to-Point
link as a result of successive power failures, you can recover the
connections by issuing the following commands:
NETCONTROL NET=niName;DELLINK=linkName
NETCONTROL NET=niName;ADDLINK=linkName
|
DTC/X.25 iX Network Link Problems
The DTC/X.25 iX Network Link operates using a DTC/X.25 Network Access
card on the DTC. For problems that involve the DTC, perform the
following steps when applicable:
Issue the LINKCONTROL linkname;
STATUS=DIAGSTATS command on the LAN link, where
linkname is the name of the DTS link. Inspect the
output and attempt to identify the problem. Retain a copy of the
output from this command for your Hewlett-Packard
representative.
Use the OpenView DTC Manager to verify the status of the DTC/X.25
Network Access card if you are using PC-based network management.
Use TermDSM to verify the status of the DTC/X.25 Network
Access card if you are using host-based network management.
|