Last Friday one of our core switches in the data center started showing output errors on all interfaces.
It is the secondary in a vPC pair with another C9364C. The other C9364C showed errors during this period.
This has happened on and off since Friday.
The error counts aren't high but I do notice them and get alerts via my monitoring tools.
I also noticed the CPU going up by about 10% and it has stayed that way since. This is however has bought it inline with the over 3 C9364C in our environment. So before this it was the odd one out......
I have logged a case with TAC and their response was 'Maybe something in the network barfs garbage periodically'
I find it weird that we are getting output errors on physical interfaces that would be caused by a device somewhere within the network.
What generally do others see as the cause of output errors?
We have had a lot of issues with 100Gb SFP's in these switches with links flapping a lot and the only way to fix it has been to RMA the SFP's. But I find it hard to beleive all SFP's would be stuffed at the same time.
I just remembered we had a interface eth1/12 that started flapping during this period, looks like another failed SFP. I wonder is this interface with input errors would then populate through others via outputs?
No comments:
Post a Comment