I ran a parity check on my server because of an issue I had that is most likely unrelated. It got 81.9% done with 65 errors and then a disk went disabled.  I was hoping someone might be able to look at the logs and see if it looks like the disk has failed or if it is a cabling issue. The server has been up for 26 days before I rebooted and ran the parity check and was not having any issues, the idea of a working cable going bad when it is in a sever untouched seems far fetched to me but stranger things have happened.  It won't give me any SMART reports or any information about the disk because it is disabled. I'm sure I could reboot and start it in maintence mode and attach the disk and get the reports (which I will need if the drive is indeed bad and I need to start a warranty claim on it) but I didn't want to touch anything until I had some guidance. I've learned that lesson!

 

The probably unrelated issue is this: The server was running along happily and the power blinked. The UPS kept everything online (except my main switch, oversite when I replaced it which has been fixed) When I got back on my other machine after the blink I noticed that several torrents that were running had gone red and said the files were missing. Sure enough, they were gone. I found them in the recycle bin.  I wasn't anywhere near the files and I know I didn't deleted them and I wouldn't think that even if it was in use the program using it losing connection with server wouldn't cause it to get deleted, but who knows.

 

Thank you!  Diags attached. 

 

 

l33tnas-diagnostics-20240726-1251.zip

Jul 26 09:24:34 L33tnas kernel: ahci 0000:05:00.0: AHCI controller unavailable!

Jul 26 09:24:34 L33tnas kernel: ata12.00: disable device
Jul 26 09:24:34 L33tnas kernel: ata10.00: disable device

Jul 26 09:24:34 L33tnas kernel: md: disk2 read error, sector=25605135336
Jul 26 09:24:34 L33tnas kernel: md: disk4 read error, sector=25605139856

 

Controller problem, so Unraid lost connection with both disks (only one gets disabled since you only have single parity), reboot and post new diags after array start

Thanks for taking a look. And oh boy, I hadn't even thought about the controller.

I have rebooted and am attaching the diagnostics.

l33tnas-diagnostics-20240727-0817.zip

Emulated disk2 is mounting, assuming contents look correct you can rebuild on top, hopefully it was a one time thing, but any more controller issues and you may try using a different PCIe slot, if that doesn't help you may need to replace it.

Just an update, I enabled the drive and ran another parity check, it finished we zero errors.  I've got a monthly check that runs on the first, Just for fun at let it run (even though the last one had finished 2 days before) and it ran with zero errors as well. I will chalk it up as a glitch and if it happens again I will take a look at moving the controller to a different slot or replacing it.

 

Thanks!