[c-nsp] Cisco cards failing

[Chris Roberts]

Heya,

 

We’ve recently lost about 6 GEIP+ cards in our 7507/7513s and now one PA-GE
in our 7204VXRs. I’ve checked the usual – environmentals, heat, etc. One
router is particularly bad having lost 3 cards (all to memory parity errors
or PMA PCI Bus Retries, which the TAC tell me is also a symptom of a memory
fault), however these are in 5 routers, across two sites. They were all
purchased about the same time (3 years ago), and we’ve not been notified of
a recall on these cards. Apparently all of these cards used to be at one
single site, which did have a heat problem a long time ago (1 year+),
although none of our kit was seriously affected.

Power was the other consideration we had, but this is obviously very hard
for us to test I believe (we don’t own the colo center).  Obviously heat
damage can shorten the life of transistors, but I find it hard to believe
that a heat issue 1 year ago would cause all of these cards to start going
at exactly the same time.


Anyone else had any experiences like this, and got any useful/anecdotal
ideas?

Cheers,

Chris.


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.762 / Virus Database: 510 - Release Date: 13/09/2004