1

were facing 5 Cisco CAT 3550 crashes now, we send out configuration to 3rd party company CISCO Gold partner and did environment power checking with the server site, no clue as yet what would be the cause of these crashes.

The password-recovery mechanism is enabled.
384K bytes of flash-simulated NVRAM.
Base ethernet MAC Address: 00:23:04:D1:6B:00
Motherboard assembly number: 73-5700-12
Power supply part number: 34-1705-01
Motherboard serial number: FDOXXXXXXXXX (removed)
Power supply serial number: DTH1217603G
Model revision number: R0
Motherboard revision number: A0
Model number: WS-C3550-24-DC-SMI
System serial number: FDOXXXXXXX (removed)

!!! WARNING: The switch is not usable !!!

Does anyone face the same problem with CAT 3550 24-DC?

petrus
  • 5,287
  • 25
  • 42
Syed
  • 69
  • 1
  • 2
  • 11

4 Answers4

3

My guess: it's a hardware problem.

I presume it's not under guarantee. Do you have some kind of service agreement or do you know somebody who could physically check that switch?

splattne
  • 28,348
  • 19
  • 97
  • 147
  • no service agreement as we didnt anticipated such high failure rate, just bought few spares just incase. – Syed Jun 10 '09 at 16:12
  • I thought the 35xx series have life time warranty. Your Cisco Gold partner should know that. I had some similar issue with an 6y old 3548XL. As far as I remember it was caused by broken memory. Are there any Error codes starts with an "%..." ? – sam Jun 10 '09 at 19:32
  • will take sometime to ship it to CISCO, anyhow a 3rd party vendor says something wrong with our config and want us to pay huge money then they would let us know, but CISCO already have our config and didnt tell us we have config problem. – Syed Jun 11 '09 at 10:01
  • I don't know if you can sent it directly back to Cisco. The normal way is to request a RMA by your reseller/Distributor. I do not beleave that its cause by a messed up config. Nerver that a config failure could force the switch to a broken state. Well you could play with conf-register setting or erase the IOS and stuff like that... but in any case you would end at least in the ROMON mode... – sam Jun 11 '09 at 20:31
2

One of the great things about Cisco is that they'll let you buy smartnet coverage after the failure occurred. That means on devices that aren't super-duper critical you can pay them their money for the support contract after the failure occurs and they'll fix it for you. We had to do this recently and the cost was pretty good (a lot better than buying new hardware, and the fix ended up being a hardware replacement, so it was worth the money).

jj33
  • 11,038
  • 1
  • 36
  • 50
  • 1
    +1 - this looks suspiciously like bad hardware. Take out SmartNet cover, open a TAC case, and let Cisco diagnose the failure directly. – Murali Suriar Jun 10 '09 at 20:53
1

If it re-occurs, do a "show version" once the switch is back up. That should tell you why it restarted (power-on, crash or whatever). It's the line that starts with "System returned to ROM by". That should tell you if you're looking at power issues.

If you see something that seems crash-related, you may well have memory issues, but it may take a couple of crashes to identify that using this method.

Vatine
  • 5,390
  • 23
  • 24
1

You may have a corrupt software image. Assuming you have a CCO account, get the latest version of software for that switch (or copy it from another switch if you do not) and copy it down. You may need to use xmodem to do this, which is going to take a long time but increase the bps to 19200 on the console port first to speed this up a bit.

Peter
  • 5,403
  • 1
  • 25
  • 32