06-30-2006 04:50 AM - edited 03-03-2019 03:52 AM
Hi all,
we got 9 catalyst 6513 switches in the access-layer. All switches running the same
Hard/Software constellation (c6k_sup2 engine + cat os 6-3-5 + 12 100BaseFX Boards), absolutely no problems so far.
But now, only one of these access-switches shows unexpected relaods at sporadic intervalls. The NVRAM log shows the reloads but no reasons. After a permanent Console-Logging the following event seems
to be the failure :
int status = 00000400, 00000000mistral interrupt status register contents: 00000
400, 00000000
table manage data par err happened.
mistral interrupt_source_status register contents: 00000400, 20000802
timer 0 int happened.
inband int happened.
programmable int 0 int happened.
table manage data par err happened.
mistral interrupt_status_10 register contents: 00000000, 00000000
fatal mistral happened
EPC: 8039607C
Stack content:
sp+00: 86014A00 86014A14 86014A00 86014A14
sp+10: 00000005 803CD0A8 81247BB4 80CE1488
sp+20: 00000006 803CC540 87FB2658 86014A00
sp+30: 00000009 000000C9 02000000 00000000
sp+40: 82C0AB30 0000000F 86014A00 000000C9
sp+50: 02000000 82CFF81A 00000023 82CFF81D
sp+60: 00000001 8063D9E4 00040009 00000000
sp+70: 00000000 00000000 00000000 00000000
sp+80: 00000000 00000000 00000000 811E71F8
sp+90: 00000000 00000000 00000000 00000000
sp+A0: 00000000 000003FF 87FB2638 00000000
sp+B0: 00000000 00000000 00000000 00000001
sp+C0: 8002D63C 3404F822 00000007 00000007
sp+D0: 00000002 00000000 87FB2648 80784B04
sp+E0: 00000007 00000000 00000007 B91237F5
sp+F0: 00040009 000F0000 00000000 00000000
Register content:
Status: 3404F823 Cause: 01000020
AT: 813905D8
V0: 00000001 V1: 00000001
A0: 81247BB4 A1: 80CE1489
A2: 00000006 A3: 00000001
T0: 00000000 T1: 34040021
T2: 00000001 T3: 85E3CC4C
T4: 00000000 T5: 0000000C
T6: 01000000 T7: 00000000
S0: 000000B4 S1: 00000042
S2: 00000042 S3: 87FB2658
S4: 00000003 S5: 00000000
S6: 81247BB4 S7: 81247BBD
T8: 00000000 T9: 85E3CC28
K0: 30419001 K1: 80CAB2E8
GP: 811DE4B0 SP: 87FB2568
S8: 87FB2568 RA: 805B35D4
HIGH: 00000000 LOW: 00000F00
BADVADDR: 82D1291A ERR EPC: 800218B8
The CSCdz65338 Bug Detail discribes a very similar situation, and i realy would like to know what happend there, before updating sw and replace hw ???
Will a sw update truly help or shall i change the complete sup2 ???
What are your experiences ?
THX for your efforts !
Maik
06-30-2006 06:14 AM
Hi Maik,
The first couple of lines seem to sugesting a parity error. This could be as simple as faulty memory on the sup.
However, the bug is very similar to your problem, although it does mention it being an IOS bug and you are running CATos.
I would run the online diags and get the crashinfo file from bootflash (if there is one there) and see if TAC can verify that you have ran into that bug.
Or...you can just upgrade the software anyway and see if it fixes it.
Cheers
Shaun
06-30-2006 06:20 AM
Hi Maik,
Looks like you're hitting a mistral interrupt. In 6.3(5), mistral interrupts were not saved to the NVRAM log. I feel that a software upgrade would definitely help here. The 6.3 train continues into the 6.4 train (it's essentially the same train, but due to hw support for a new blade, we had to change the release name for the sake of consistency), and the latest maintenance release of 6.4 will give you all of the diagnostic enhancements and bug fixes that have been integrated into this train. The latest maintenance release of 6.4 is currently 6.4(21) and is GD code.
HTH,
Bobby
*Please rate helpful posts.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide