[c-nsp] C6500 C6KERRDETECT-SP-2-FIFOCRITLEVEL

[Sukumar Subburayan (sukumars)]

Also, I don't see any specific bug fixes related to this error msg. so,
I don't think upgrading the code will do anything. As part of upgrade,
you would reset the switch, which might clear some temporary condition
which was causing this error message, however, I don't see any fixes
which went in the latest release to fix any false positives.

sukumar

-----Original Message-----
From: cisco-nsp-bounces at puck.nether.net
[mailto:cisco-nsp-bounces at puck.nether.net] On Behalf Of Sukumar
Subburayan (sukumars)
Sent: Saturday, October 13, 2007 10:34 AM
To: cisco-nsp at moe.co.uk; Kevin Graham; cisco-nsp at puck.nether.net
Subject: Re: [c-nsp] C6500 C6KERRDETECT-SP-2-FIFOCRITLEVEL

comments inline..

sukumar
 

-----Original Message-----
From: John Imison [mailto:i at moe.co.uk] On Behalf Of cisco-nsp at moe.co.uk
Sent: Saturday, October 13, 2007 7:43 AM
To: Sukumar Subburayan (sukumars); 'John I'; 'Kevin Graham';
cisco-nsp at puck.nether.net
Subject: RE: [c-nsp] C6500 C6KERRDETECT-SP-2-FIFOCRITLEVEL

Hi sukumar,

Thanks for your reply.

> There is actually no need to swap chassis. 
> 
> This error message indicates that the SP-CPU inband port (the internal

> 1 Gig connection) FIFO pointers were not moving for certain amount of 
> time possibly indicating something wrong between the CPU and it's 
> inband connection. All this is internal to the Supervisor.
> 

I've only started receiving this message after upgrading the supervisor
memory on the module, otherwise, it has been fine for months.

Could I have not reseated the MSFC properly onto the SUP?  Could it be
the memory?  Could I have bumped something else?

##sukumars

possible, as you are saying that this started happening only after
upgrading memory.

Are there any commands I can execute to see these fifo pointers you
mention?

##sukumars

there are some asic register commands on the SP-side for the ASIC
(pinnacle), you can look at. But since these are debug type commands, I
would not recommend using them without TAC help.

> If the errors are persistant, the active sup might eventually reset or

> switchover (in case there is standby). As part of the upgrade process,

> the transient condition probably got cleared.
> 

The errors have been going for a while now, however, the unit continues
to switch/route packets without problem.

What would you recommend I do?

##sukumars

For Fifo stuck, we do try auto-recovering by soft resetting the internal
ASIC path. So, if you are not really seeing any problems, I would
suggest you don't have to do anything. If the errors are happening very
very frequently, then I would suggest the below steps.

(I will be trying what's been recommended already such as reseat module
& upgrade IOS)

Thanks,
John
_______________________________________________
cisco-nsp mailing list  cisco-nsp at puck.nether.net
https://puck.nether.net/mailman/listinfo/cisco-nsp
archive at http://puck.nether.net/pipermail/cisco-nsp/