Crash in 6500

rc5295509 · ‎09-12-2012

Hi

I have a rare case. switch 6500 make crash and go to rommon. when review the crash appear that the switch charge the configuration from nvram, but, at the ending there are a few line command, that make the switch go to crash.. then I have to booting from rommon and start again. I use the version 12.2.33.sxi9 and 12.2.sxi3 and the configregister is 0x2102

I think that the switch get the startup-config file from other file-system.

Appear this line at the end

*Sep 1 03:42:42.352 Inviern: %SYS-5-CONFIG_I: Configured from memory by console

access-list 199 permit icmp host 10.10.10.10 host 20.20.20.20

crypto map NiStTeSt1 10 ipsec-manual

match address 199

set peer 20.20.20.20

exit

no access-list 199

no crypto map NiStTeSt1

*Sep 1 03:42:46.952 Inviern: %SYS-5-RESTART: System restarted --

if somebody want to see the crash file.. here is it

Some idea ?

skarthic · ‎09-12-2012

Hi,

Indeed this seems to be strange.

How frequent does the switch copy the configuration from the memory. Normally the config would end up with the NTP configs. Can you check what is the config below. Any idea why this config is added?

CMD: 'access-list 199 permit icmp host 10.10.10.10 host 20.20.20.20' 03:42:42 Inviern Sat Sep 1 2012

CMD: 'crypto map NiStTeSt1 10 ipsec-manual' 03:42:42 Inviern Sat Sep 1 2012

CMD: 'match address 199

' 03:42:42 Inviern Sat Sep 1 2012

CMD: 'set peer 20.20.20.20

' 03:42:42 Inviern Sat Sep 1 2012

CMD: 'exit' 03:42:42 Inviern Sat Sep 1 2012

CMD: 'no access-list 199' 03:42:43 Inviern Sat Sep 1 2012

CMD: 'no crypto map NiStTeSt1' 03:42:43 Inviern Sat Sep 1 2012

Also attach the SP crash-info file.

rc5295509 · ‎09-12-2012

Hi,,

Well I found this messages in other devices, is part of algoritm of testing of Crypto. and the problem seem to be, a little space of flash. tomorrow will probe this solution.

Zeeeee · ‎05-12-2017

Hi,

I am aware that this thread started 5yrs ago but may I know what is the solution or workaround to it?

Thank you.

pille1234 · ‎09-12-2012

Hi there,

I once wasted allot of time with similar configuration changes on a cisco 36xx router. While it looks like some hAcKiNG teenager is trying to get access to your router it is nothing else but the crypto engine going through startup selftest. I didn't know that Cat 6500 was doing this as well. I consider that to be a rather amateurish implementation of POST by Cisco.

Anyway, the only useful information I could find in the crash info was

*Sep 1 03:44:06.664 Inviern: %CPU_MONITOR-3-PEER_EXCEPTION: CPU_MONITOR peer has failed due to exception , reset by [1/0]

%Software-forced reload

This looks like a software bug. Cisco shares its whole wisdom with us here:

http://www.cisco.com/en/US/docs/ios/12_2sx/system/messages/sm2sx02.html#wp1013983

Error Message    %CPU_MONITOR-3-PEER_EXCEPTION: CPU_MONITOR peer has failed due to 
exception , resetting [[dec]/[dec]]

Explanation The CPU monitor messages failed on the peer processor. The system is no longer operational.

Recommended Action No action is required.

See? No action is required

What I find confusing is that your switch fails to reboot and stops in ROMMON mode. Does the switch reboot normally after a simple reboot command? Is your boot var setting correct?

Regards,

Pille