Interesting, the switch did not crash but a debugfile was generated. And something I can read from that file:
++++++++++++++++++++++++++++++++++++++++ Processor Memory: total 388120864, free 276632816, used 111488048 IO Memory: total 67108864, free 22640, used 67086224 (..) CPU utilization for five seconds: 100%/28%; one minute: 99%; five minutes: 99% ++++++++++++++++++++++++++++++++++++++++ Memory leak ? It's the first time I see an equipment recovering from a situation like this without human intervention. It would be nice if the impact was not an STP loop... Regards, Antonio Soares, CCIE #18473 (R&S/SP) [email protected] -----Original Message----- From: [email protected] [mailto:[email protected]] On Behalf Of Antonio Soares Sent: sexta-feira, 5 de Novembro de 2010 17:15 To: 'Jared Mauch' Cc: 'cisco-nsp' Subject: Re: [c-nsp] Spanning-Tree Loop (12.2.18SXF7) The 6500's uptime is 1 year and 16 weeks. The funny thing is that the 6500 recovered by itself, no reload or sup switchover occurred. And I don't see anything wrong with the memory stats. Really strange. Sometimes we prefer a crash :) This STP loop was active a few hours... I can't upgrade without knowing what is the bug. I will keep searching. Thanks. Regards, Antonio Soares, CCIE #18473 (R&S/SP) [email protected] -----Original Message----- From: Jared Mauch [mailto:[email protected]] Sent: sexta-feira, 5 de Novembro de 2010 17:00 To: Antonio Soares Cc: 'cisco-nsp' Subject: Re: [c-nsp] Spanning-Tree Loop (12.2.18SXF7) There could have been any sort of a memory leak since SXF7 and now that caused you to see this. How long was your device up? I hate to say this, but I would go to recent software (eg: SXF15a/SXF16). - Jared On Nov 5, 2010, at 12:25 PM, Antonio Soares wrote: > Hello group, > > I'm troubleshooting a STP loop that seemed to be triggered by these errors: > > ++++++++++++++++++++++++++++++++ > %PM_SCP-SP-3-LCP_FW_ABLC: Late collision message from module 1, port:029 > %IPC-SP-5-WATERMARK: 15612 messages pending in xmt for the port > CHKPT:STANDBY SP(2080000.B) seat 2080000 > %IPC-SP-5-WATERMARK: 20068 messages pending in xmt for the port > CHKPT:STANDBY SP(2080000.B) seat 2080000 > %IPC-SP-5-WATERMARK: 22904 messages pending in xmt for the port > CHKPT:STANDBY SP(2080000.B) seat 2080000 > %PM_SCP-SP-3-LCP_FW_ABLC: Late collision message from module 2, port:039 > %IPC-SP-5-WATERMARK: 25710 messages pending in xmt for the port > CHKPT:STANDBY SP(2080000.B) seat 2080000 > %IPC-SP-5-WATERMARK: 28510 messages pending in xmt for the port > CHKPT:STANDBY SP(2080000.B) seat 2080000 > %PM_SCP-SP-3-LCP_FW_ABLC: Late collision message from module 2, port:039 > %IPC-SP-5-WATERMARK: 30344 messages pending in xmt for the port > CHKPT:STANDBY SP(2080000.B) seat 2080000 > > %SYS-SP-2-MALLOCFAIL: Memory allocation of 1768 bytes failed from > 0x4020DCC8, alignment 32 > Pool: I/O Free: 148112 Cause: Memory fragmentation > Alternate Pool: None Free: 0 Cause: No Alternate pool > > -Process= "Spanning Tree", ipl= 2, pid= 126 > -Traceback= 40280990 402826F8 4020DCD0 4020E064 402108DC 4020D0B8 40AAB1BC > 40AAB64C 404A3970 404C4F64 404C51A8 404C5460 404AAA58 404AAE40 40A9EE00 > 40A9D71C > ++++++++++++++++++++++++++++++++ > > These were seen on one of core's 6500. The 6500s are running 12.2.18SXF7. > > Fortunately the problem went away by itself. The last thing we want is a STP > loop in a network with several 6500s and 4500s, right ? :) > > Anyone has seen something like this ? I'm now hitting Bug Toolkit to see if > this was reported before. > > > Thanks. > > Regards, > > Antonio Soares, CCIE #18473 (R&S/SP) > [email protected] > > > _______________________________________________ > cisco-nsp mailing list [email protected] > https://puck.nether.net/mailman/listinfo/cisco-nsp > archive at http://puck.nether.net/pipermail/cisco-nsp/ _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/ _______________________________________________ cisco-nsp mailing list [email protected] https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
