0

We have three small business PoE switches, SG220-50P purchased from two places. They are each randomly rebooting at different times every few days, and sometimes only after a few hours. They have been doing this basically since we purchased them. I initially thought the devices were defective, but I haven't seen anything regarding this model, and I wouldn't expect to get 3 bad ones from two different sources.

I've enabled flash logging, and nothing abnormal is recorded, it's not even showing power down. They just come back up as if it were a normal boot. I updated the firmware two times now, and that hasn't helped. Resetting to factory seems to help for a while, but then the problem comes back. The devices are on a UPS that isn't overloaded. They are each consuming about 35W on PoE, with a max of 375 available.

What would you recommend to troubleshoot this issue?

I'm new to networking admin, so sorry if my question is unrefined.

  • Do the reboot at the same time? If so, you should look at the power, UPS or not since the UPS could be the problem. You can change the logging level to try to capture more. You could also set up SNMP with traps, but you will need to set up a server to capture that. You may need contact Cisco about this problem. – Ron Maupin Mar 2 '16 at 19:24
  • How do you know/confirm that they rebooted ? Is there a uptime shown ? – Pieter Mar 2 '16 at 20:48
  • They do not reboot at the same time. I've put them on two separate UPS and both sets still had the issue. I put logging level to 5 on flash and 6 on RAM. Uptime is shown, and the system time reverts to 1999 in the logs until it gets the current time from NTP server. – Yanzzee Mar 2 '16 at 20:58
  • Can you disconnect one of the switches from all network connections, so it is only connected to power, and see if it still reboots? Would maybe give a hint if some strange traffic pattern/bug causes the problem, or if they have a generic issue not related to any traffic. – xpac Mar 2 '16 at 22:24
  • I've thought about that, but the test would take up to 4 weeks or more depending on when the next reboot happened. Supposing it is strange traffic, how would I track that down? snmp traps? observium or something? – Yanzzee Mar 3 '16 at 0:10
0

I eventually got Cisco to RMA two of the three switches. The new ones were revision V01 instead of V00, so I was hopeful for an improvement. However the new ones were doing the same thing after a few days, one of them worse than the other but still did it. I'm basically using these as unmanaged switches, so it's definitely not a config problem.

I read on a Cisco forum that the new firmware 1.1.1.2 has resolved the problem. I'm upgrading now from version 1.0.0.19. I will update if this hasn't resolved it.

  • You should accept your answer so that the question doesn't keep popping up forever, looking for an answer. – Ron Maupin Jan 30 '17 at 13:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.