Disk error, what do i do now?

[nmills3]

I just got an array error on disk 3, i've had some issues with errors on drives connected to the on board sata controller before but i fixed that by connecting to a hba instead, my drives are all only 1-2 years old so i'd be surprised if my drive is dead

tower-diagnostics-20221129-2110.zip

[nmills3]

ok, so i stopped the array, un assigned the drive, started the array, stopped it again, then went to stop it again but it would stop so i had to restart the pc, now i got it running again and i'm rebuilding onto the same drive but the emulated contents of the drive seem to be completely empty, none of the data that was on the drive is showing anywhere

[nmills3]

Also on the "main" page, where it normally shows the drive capacity/used size is just saying "Unmountable wrong or no file system"

[nmills3]

so the disk got 8% into a rebuild while having the previously mentioned issues and now stopped and has about 380 errors. i've attached the new logs. i have no idea what to do now and i'm very worried about losing data. what do i do?

tower-diagnostics-20221129-2243.zip

[JorgeB]

Problem with disk3 looks power/cable related, check/replace the cables for that disk, then start the array and before rebuilding check filesystem on the emulated disk3.

[nmills3]

currently it's showing this

Phase 1 - find and verify superblock...

couldn't verify primary superblock - not enough secondary superblocks with matching geometry !!!

attempting to find secondary superblock...

and then it just has a line of many many dots

[JorgeB]

Are you using the GUI or running the command manually?

[nmills3]

the gui. I followed the instructions for checking the drive on the link you sent. It says it's still running, is it meant to take this long?

[JorgeB]

1 minute ago, nmills3 said:

is it meant to take this long?

Usually not unless the superblocks are damaged, if they are it will scan the complete disk looking for a backup, post new diags in the meantime.

[nmills3]

tower-diagnostics-20221130-1108.zip

[JorgeB]

According to those you didn't reboot yet, and still lots of errors on disk3, check/replace cables for disk3 and post new diags after array start.

[nmills3]

i rebooted last night, left the array rebuilding and it failed about 7% in, then this morning i tried testing it with no disk assigned to disk 3 which didn't work and then i got the reply from you so i restarted the array in maintenance mode and the test has been running on the emulated disk 3 since. I assume i should wait until after the test is done to start changing cables

[JorgeB]

Above I mentioned to replace the cables first.

[nmills3]

My mistake. I don't think i have a spare cable for that one to hand, it's a sas to 4x sata cable. If i leave the drive unassigned, should i be able to get the emulated disk working until i can get another cable?

[JorgeB]

If you leave it unassigned it should not be a problem, but fist reboot anyway to clear the failed rebuild.

[JorgeB]

And you can at least disconnect and reconnect the cables.

[nmills3]

should i do that right now, or should i wait until after the repair thingy finishes trying to find a second superblock on the emulated drive?

[JorgeB]

If it still didn't find one unlikely to find it later, so do it now.

[nmills3]

ok disk removed, so should i start in maintenance mode and run the xfs repair tool on disk 3 again?

[nmills3]

[nmills3]

tower-diagnostics-20221130-1155.zipalso here's another diagnostics file

[JorgeB]

12 minutes ago, nmills3 said:

so should i start in maintenance mode and run the xfs repair tool on disk 3 again?

Yep.

[nmills3]

ok, so far the test is doing the same as last time. how long should i let it run for?

[JorgeB]

Let it run until if finishes, because if xfs_repair cannot fix the filesystem, and assuming no backups exist, the only solution to try and recover the data would be running a file recovery util like UFS explorer.

[nmills3]

ok, assuming it takes as long to scan the entire drive as a parity check, this should take about 8 hours to run