The other day the SP on my 8440 (4 node) sent an email with the following info:
Abstract: (Major)PORT2:1:1 PEL_ERROR SAS cabling issues on 2:1:1. Run "checkhealth -pelmo Event string: SAS cabling issues on 2:1:1. Run "checkhealth -pelmon -d cabling 2:1:1" for more information.
I actually got two emails, one for port 2:1:1 and one for port 2:1:2.
By the time I got a chance to run a checkhealth on the array (maybe 5 minutes after the email), everything was fine and the maybe 10 minutes later the event was marked as resolved by system.
In terms of the impact, I noticed 1 host in my ESXi cluster logged errors in the veents that seems to indicate it dropped half the paths to maybe 20% of the luns attached to it . Checking 2-3 other hosts (they were more though), they didn't have messages at that time (I did not dig into the vmkernel.log though).
The other thing I noticed was the time the email was sent was earlier (by roughly 20 minutes) that the time of the event on the esxi server. Everything should be synced to the same NTP server but I found it odd the times didn't match. I guess that's something I'll need to check when I'm back in the office if all the times on the hosts/storage match up.
Long story short, how concerned should I be about this error and is there something I should check? I've never seen an error like this before and it seems to have fixed itself just as suddenly as it happened.
Thanks
|