I was upgrading a 250GB drive I had in my array with a 1.5TB one when one of the other disks failed during the rebuild. I tried putting the old 250GB drive back in since it still has everything on it but unraid sees that the drive is smaller than the one I was replacing it with and refuses to start the array. This sounds like it could be fixed since the data on the 250GB drive hasn't been touched at all and only one drive failed.

The old 250GB drive in its old slot, the drive I was going replace it with is going in the dead drive's position (yes it's bigger than the old one)

And Unraid complaining about there being too many wrong disks

If the old drive is OK you can do a new config with it, but post the diagnostics first.

Wouldn't that automatically wipe the party and make me lose all of the data on the failed drive?

1 hour ago, JorgeB said:

If the old drive is OK you can do a new config with it, but post the diagnostics first.

Edited Wednesday at 08:35 AM4 days by peterpeter

The new config doesn't wipe any of the data drives, since parity should be mostly valid; you could also check "parity is already valid" and then run a correcting check.

The procedure, again assuming the old disk is healthy, would be to go to Tools -> New Config -> Preserve current assignments: All -> Apply

Assign the old disk as disk2, I assume there wasn't a disk3, so leave that unassigned.

Check "parity is already valid" start the array and run a parity check

21 minutes ago, JorgeB said:

Assign the old disk as disk2, I assume there wasn't a disk3, so leave that unassigned.

Check "parity is already valid" start the array and run a parity check

The failed disk was disk2, the one I was going to upgrade was disk3.

jankserver-diagnostics-20260826-1253.zip

That changes things; I misunderstood the initial issue. You cannot do a new config.

The diags posted don't show the disk failure; it also looks like you are trying to assign the disk that failed as disk3 as disk2???

Please post the details of how the array was originally configured, or better yet, post the diags after the issue.

Original Config before failure:
Same Parity drive, disk1 is the same, disk2 is the failed drive, disk3 is the old drive I was going to upgrade.

What happened: I upgraded disk3 to a bigger drive and while the rebuild for the new drive was going - disk2 failed.

I still have the old disk3 which all the data on it intact.

What I'm trying to do:

Replace disk2 with the drive I was going to upgrade disk3 with; return the old unupgraded disk3 to its old slot. Unraid throws an error about there being too many wrong disks.

Edited Wednesday at 10:24 AM4 days by peterpeter

7 minutes ago, peterpeter said:

Replace disk2 with the drive I was going to upgrade disk3 with; return the old unupgraded disk3 to its old slot. Unraid throws an error about there being too many wrong disks.

You cannot do that, post new diags with all the drives connected and assigned as they were when the disk2 failed.

I connected the failing drive back to the system and made everything the way it was. I can try starting the array to hopefully save the data but I won't do anything for now.

jankserver-diagnostics-20260826-1357.zip

Disk2 is not disabled; by "failed during disk3 upgrade," I assume you mean there were read errors. SMART does show a pending sector.

Disk3 also doesn't look very healthy; my first suggestion would be to run an extended SMART test on both and post new diags after.

I did a short smart test on disk2 before and it was spitting out read errors. During the rebuild it was making weird noises (not quite clicking but close enough) and it was incredibly slow. I'll do an extended SMART test on both drives and post diagnostics.

Also I gotta say that I am extremely thankful for the work you (and the other magicians in the forum) do for the community. <3

23 minutes ago, peterpeter said:

I did a short smart test on disk2 before and it was spitting out read errors.

No point in doing the extended test if the short test fails. If the short test fails then the drive needs replacing ASAP, whereas passing the short test is not definitive and then you need the extended test to get a more definitive answer.

The diagnostics after doing SMART tests on both drives:

jankserver-diagnostics-20260826-1523.zip

Edited Wednesday at 01:38 PM4 days by peterpeter

It failed for disk3, confirming that disk is not good. There's no new result for disk2; maybe it's still going?

I tried running the test on disk2 again, it seems like it finishes without errors but I think that the drive is so far gone that it can't even initialize a SMART test.

If there weren't any writes done after the disk was replaced, you can try this, but it will only work if parity is still valid, or mostly valid, with the old disk:

-Tools -> New Config -> Preserve current assignments: All -> Apply

-Go to main and correct the assignments: Parity, disk1 and disk2 should be the current disks, assign the old disk3 in its slot

-IMPORTANT - Check both "parity is already valid" and "maintenance mode" and start the array (note that the GUI will still show that data on parity disk(s) will be overwritten, this is normal as it doesn't account for the checkbox, but it won't be as long as it's checked)

-Stop array

-Unassign disk2

-Start array (in normal mode now), and post new diagnostics

Alright, i'll do exactly that.

Did it, stopped the array after I did the diagnostics

jankserver-diagnostics-20260826-1852.zip

Check filesystem on the emulated disk2; disk3 is also not mounting due to a different problem, but let's ignore that for now, to see if disk2 can be emulated.

It's checking, hope to report with good news soon-ish

That's a rebuild, not a filesystem check. I would recommend seeing if the filesystem can be fixed before attempting a rebuikd, since there's not much point if it can't.

It is a check, it just looks like a rebuild.

4 hours ago, peterpeter said:

It is a check, it just looks like a rebuild.

It is a read check that is checking parity, not a check filesystem that checks the filesystem on the drive.

oh i completely misread the docs... thanks for correcting me, I'll run a proper filesystem check now

jankserver-diagnostics-20260827-0036.zip

Edit: I haven't done anything yet - the docs say that I can click the fix button but im not doing anything unless instructed by you guys

Edited Wednesday at 09:40 PM4 days by peterpeter