I've replaced my parity 2 drive a few times because it keeps getting offline uncorrectable errors.

 

I've changed ports and cables, but still seem to be having this issue. Not sure what to look for, but diagnostics are attached.

diagnostics.zip

What did you replace it with?  The one in the diagnostics has 29000 hours on it, and is showing UNC in the error logs.  Run an extended SMART test on it.

You apparently haven't replaced it since Aug 11 because syslog goes back that far. It was showing up as a medium error on parity2 in syslog though so I think it is a problem with the disk. You could run an extended SMART test if you want.

 

Can't say anything about any previous disks you replaced obviously, but that disk is over 3 years old. A few of the others I looked at are much older, so it makes me wonder if you are using old, used disks in your server and whether or not you test them before trusting them in your array.

This server is my backup undraid server, so the drives in it are from the main server. When I upgrade a drive for more capacity, they go in the backup server if needed.

 

This is the third drive I've used as a parity 2 drive in this server, but it did live it's entire life in my main server before being tasked as a parity drive. The other drives are older, but this is the only one I've had "fail" repeatedly. The Reported uncorrect count hasn't gone up in a long time, but this drive randomly has errors, then it'll be fine for a long time. 150 days at this point.

2 hours ago, PanteraGSTK said:

This server is my backup undraid server, so the drives in it are from the main server. When I upgrade a drive for more capacity, they go in the backup server if needed.

Exactly what I do. Just about to do it again after I upgrade a disk in my main server.