4500 Chassis - Spurious PSU Failure Messages

[Mark Richards]

Hi,

Hoping to get an answer to an issue we're seeing across multiple 4500 Switches ever since upgrading the IOS.

Every now and again and by no means regularly I see the following log message

011489: Oct  1 06:12:58: %C4K_IOSMODPORTMAN-4-POWERSUPPLYBAD: Power supply 1 has failed or been turned off

011490: Oct  1 06:12:58: %C4K_CHASSIS-3-MIXINPOWERDETECTED: Power supplies in the chassis are of different types (AC/DC) or wattage

011491: Oct  1 06:13:03: %C4K_IOSMODPORTMAN-6-POWERSUPPLYGOOD: Power supply 1 is okay

011492: Oct  1 06:13:08: %C4K_IOSMODPORTMAN-6-POWERSUPPLYOUTPUTINCREASED: Power supply 1 output has increased

011493: Oct  1 06:13:08: %C4K_IOSMODPORTMAN-6-POWERSUPPLYINPUTON: Input 1 on Power Supply 1 has been turned on

I did search the forums but the only answers I found were to "Change the PSU"

I got suspicious because we are seeing this message on multiple devices, quite rarely to be honest but only on devices which are running cat4500-entservicesk9-mz.122-53.SG2.bin. Our other 4500's are running cat4500-ipbasek9-mz.122-46.SG.bin and haven't as yet given the error. If you notice the timestamp the error seems to clear in 5 seconds in all cases which is quite puzzling.

Is this an issue with the IOS?

[glen.grant]

   The first thing to do is to check the connection going into all power supplies. These connections are notoriously hard to get inserted correctly,  they are extremely tight going in and they may not be seated correctly.

[mvsheik123]

Hi,

We have #3 4500 on the same image (more than 10months) and never received any such messages. As Glen suggested check the connection ot try reseating the PS#1.

hth

MS

[Mark Richards]

Thanks for the comments but I'm not convinced, it just seems too co-incidental.

We've got 48 of these devices and only 9 are running the newer code, of that 9, 5 have given the failed PSU message and 1 of them has done so twice. All the 4500's have 2 of the 4200W dual input PSU's installed and all have the cable clamps fitted. These Catalysts have been in situ for 3 years without exhibiting this error before, we've had a few failed PSU's but the error stays until the failed part is replaced.

I appreciate that it could be a loose connection or a reseat required but why is it only the new code and why exactly 5 seconds everytime?

[mvsheik123]

Hi Mark,

If this seems co-incidental and there is not really any issue with hardware or connections, this may be a cosmetic bug.

I recommend you to reachout to TAC to investigate further.

Thanks

MS