4B00F002 error code

[teqmem]

Hello,

I have this AIX server model "7017-S7A" running AIX 4.3.10. It crashed this morning with the error code "4B00F002" on the LED display.

I found an article on IBM's site stating that this is a TCP/IP problem and suggested a fix. However, the fix is for SNA.rte which is not loaded on this server.

The error log didn't any error when the server went down.

Any idead on this problem?

The IBM article is:

http://www-1.ibm.com/support/docview.wss?uid=apr39cd8b01b4e473c3e852567c9005835b9

Thanks.

 

[DukeSSD]

The service guide shows what to do for this error:
Error Code Description Action/
Possible Failing FRU
4B00F002 AIX terminated with no fault information
These are AIX 888 crashes. Please
follow the procedure under
"Action/Possible Failing FRU" for how
to identify the type of crash and gather
the necessary information.
Review Operator Panel for fault information.
1. Do not attempt to reboot or power
cycle the system by any means, as
critical failure data will be lost.
2. While this SRC is still displayed on
the operator panel, and with a TTY
attached to a native serial port
(preferably S1), hit any key on the
keyboard (this will put you at the
main Service Processor Menu).
3. Read the Service Processor Error
Logs (Option 3, Option 3).
4. The most recent entry (at the
bottom) in the error log will show
the 888 crash code string (you may
have to scroll a few pages).
5. Make note of this code string and
then consult the Diagnostic Information
for Multiple Bus Systems
manual for the particular code and
suggestions on how to proceed.
6. Exit to the main Service Processor
Menu and POR the system by
hitting the operator panel power
button twice (in most cases,
machine will reboot successfully).
7. If machine re-boots successfully,
view the AIX errpt for more information
about the crash, contacting
the appropriate support group
(either RS/6000 H/W or AIX S/W
support, based on the crash code
as listed in the Diagnostic Information
for Multiple Bus Sytems).
NOTE: in most cases, AIX will
have produced a system dump that
needs to be sent into AIX Software
Support for analysis, should the
crash code be listed as
S/W-related in the Diagnostic Information
for Multiple Bus Sytems.
8. If machine fails to reboot, follow
P/D procedures based upon the
particular failure now occuring.

 

[DukeSSD]

Did it crash while you were logging into the CDE as root?
if not then the link you posted does not apply and it crashed for some other reason.
If your error log does not show any problems you should shut down and use an ASCII terminal on the serial port to look at the service processor logs, record the information and ask IBM to analyse it.
What sort of AIX level is "4.3.10", if it is 4.3.3 ML10 then upgrade to 4.3.3 ML11 + fixes, if not then upgrade anyway.
4.3.3 is off support so 5.1 is the minimum if you need a dump analysed by IBM AIX support.

 

[teqmem]

Thanks DukeSSD.

I mistyped and really meant AIX 4330-10 (4.3.3.0 ML10). The crash happened at night so I agree the link I mentioned does not apply. This server needs to stay at AIX 4.3.3 as there are web services (Apache/Chilisoft) that can only run in 4.3.3.

Anyway, I rebooted it as soon as I found the crash and it came up without problems. The errpt didn't log anything when it crashed.

So at this point I really don't know what happened. It would be nice to know so I can prevent future happenings.