Has anyone ever seen entries in /var/log/messages similar to:
"Jan 21 05:41:32 dm01db02 kernel: RDS/IB: re-connect to 102.***.**.** is stalling for more than 1 min...(drops=12 err=0)" ? I have masked out the true address in case it's legitimate, but we do not recognize it, know how it got there, or why. I thought it might be related to Platinum Support, but they say it's not theirs. Any suggestions on how to determine what process is trying to establish this connection or where to look for additional info would be greatly appreciated. We are running netstat in the background and have not logged any actual connections, only these "re-connect" entries in /var/log/messages.
Yes I see them in the messages file on my systems as well.
They error (or message) is related to infiniband connectivity.
If you ssh to "102.***.**.**" you will probably end up on a storage cell.
As a check to exclude any issue run exacheck and confirm all is OK.