Cisco WS-3560G Problems
So about 2 months ago I have a Cisco WS-C3560G-24PS-S switch die on me. It was working then it just stopped. So I power cycled it. Fans turn on and the lights on the front turned on, then went off, but not much else happened, no boot and nothing from the console at all.
So I replaced it with a spare we had (exact same model), and everything was happy. The dead switch sat on my desk for about a week until I managed to get the time to call Cisco to see if they would replace it, and while I was on hold I decided to plug it in and remind myself of the issue. Lo and behold the damn thing started up just fine and was working 100%. I had Cisco replace it anyway, since I could not trust it.
Anyway, fast forward to today, and the replacement switch I put in two months ago has died as well. With the exact same symptoms. Plug it in, get the fans spinning, all the light turn on briefly and then turn off and no console data coming from the unit at all.
One switch failing like this is par for the course and nothing to ponder about, but 2 switches, having the exact same issue in such a short period of time? That leads me to think there is something else going in here, but I am at a complete loss as to what that could be.
Any suggestions on where I can look here? I'd like to be reasonably sure that this won't happen to a third switch I put in place.
DustinB3403 last edited by DustinB3403
It could be the power supply (sometimes faults lie in the manufacturing process), or a software issue. Opening the unit would void your warranty.
Is there any logging available for this unit that might indicate what might be the culprit? Is the software out of date on these?
Dashrender last edited by
As Dustin said, there are a lot of things this could be - bad power from the wall/UPS, to bad hardware.
Two isn't enough to really start getting worried about something in my opinion.
Additionally, you mentioned that this second unit was the same as the first - perhaps it has a similar hardware malfunction that the first had. I'd guess heat related since after the first failed switch sat and completely cooled off it worked again - so might this second failed unit.
Have the environmental conditions around the switches changed?
The replacement switch was actually retired from another site, it ran 100% of the time for years, and the original one has also been running for more than a 5 years in the same location. So if there was some sort of manufacturing issue I would have thought this would have surfaced a long time ago.
As to logging on the unit, there is nothing that I could see on the unit that eventually started working again. But I can't get to any logs since the thing won't boot or push anything out via console.
As to the environment changes, there are none that I can think of. The UPS died in there about the same time as the first switch died, so it was the prime culprit back then, but the new one was plugged directly into the wall till we could circle back and get the UPS fixed up (it just needs new batteries). It is also worth mentioning that there is another switch in that IDF that has not had any issue, plugged into the same outlet.
Might just be older units. Cisco devices are not particularly long lived. If you have "real" Cisco, they are pretty good, but switches are just hardware like anything and really do die from time to time. Check the power supply, make sure that the heat isn't spiking at the location. If those things are good, chances are a third switch will live a natural life.
@scottalanmiller - They both were well over 8 years old, and we have the exact same model all over the place (30+ of them) and in the 5 years I've been here the number of failures we have had could be counted on one hand.
It turns out there were some power issues on the grid on Monday night (downed tree on power lines) and that may have been the culprit here since as I mentioned the UPS is offline now (batteries will arrive Monday) so the surge could have taken the thing down.
Dashrender last edited by
8 years old - yeah, Personally I fully expect failures at any time, just expect them, but am not unhappy when they don't happen either.
I put in 40 new VOIP phones 10 years ago, starting around 7 years, when we lost power long enough to take the switches down, we started having phone failures - they wouldn't power back on.
I actually was talking to @scottalanmiller about a related thing - stickion - when you turn off a computer with drives in it that when the drives stop spinning, they don't work anymore because the heads stick to the platters.
I'm not sure how often this really happens anymore, but it is a real thing. Scott said that the recommendation is to not allow drives to spin down in these situation when possible.
These 8 year running devices are just the same as my phones and those HDDs.