I said "drive failures" because I think my drives are actually fine. I just didn't know if anyone else has run across this.
Yesterday my server emailed me saying it had array errors, 3 disks with read errors (Parity disk, Parity disk 2, and Disk 3). It disabled the Parity Drive as well as Disk 3, but things kept running. I was thinking the chances of three drives having errors all at once out of nowhere seemed a bit low, so I doubted it was actually bad drives.
Turns out the following below is not the problem. Problem just came back and shows 11 drives with errors. I'm at a loss and have diagnostics if someone smarter than me can make sense of them.
To get to it, I am wondering if this is the problem somehow..
I use the docker ShinySDR for a SDR dongle I use. I unplugged it from my server a day prior as I was getting a longer antenna cable for it. The docker file had the usb device set as /dev/bus/usb/003/002 (which was correct prior to me unplugging it). and the docker was set to automatically start. Somewhere in there I rebooted the server and I think this is where the issues started. I rebooted numerous times, shut down all docker containers, shut down the one vm I run, and tried to remove all plugins I felt I didn't need trying to find what the issue might be. I forced the server off a couple of times as it was just unresponsive as well. The server actually emailed me yesterday afternoon saying I had 9 disks with read errors. Well I opened the terminal and ran lsusb to see what it had connected and /dev/bus/usb/003/002 was now "Bus 003 Device 002: ID 058f:6387 Alcor Micro Corp. Flash Drive" - This is my Unraid USB drive... I am wondering if this cold have been the cause. I didn't know if the docker container could be trying to access the usb drive in such a way as to spew out all of these read errors and disable my drives.
I did run tools -> diagnostics several times, but I now know that every time you reboot you might miss something important. These files along with the syslog did show errors, but I'm hesitant to believe it as it has since rebuilt the parity drive, and is currently 66% through rebuilding disk 3. The syslog currently shows only the errors for the disabled drives, prior to me removing them and adding them back.
thoughts?
thanks,
John