Hello fellow unraiders,

 

I found myself in a bit of a pickle, where I have one missing drive and one that is unmountable. Since I'm an idiot and I restarted the server a couple of times I assume, getting the logs wont't be much help. I will attach them anyway.

 

But let me tell you how the events that caused that problem occurred:

A couple of days ago I had a power-outage (yes I do not have an UPS) which seemed to have no effect on the server. A couple days later one disk is missing. Since I had that problem before I thought it was just another bad sata-cable. So I powered down, exchanged the cable (which did not help) and now I also have a drive that is unmountable(xfs) and two others  where the smart status telling me they are on their way out.

 

So I started the array in maintenance mode and did the filesystem check on the unmountable disk. Since I am very new to Unraid and the unix world in general and also the logs don't tell me anything, I thought I need to get help here. Before I mess anything more up. 

 

Should I write the correntions to the filesystem? It seems like the only option anyways.

How do I proceed frome here? I already have 4 new drives ready to exchange for the bad ones. Will I be able to fix this without data loss? 

 

Thank you for taking the time to help me.

cheers

 

 

 

 

 

Edited December 3, 20178 yr by Heciruam

Server is not in very good shape, you're likely going to lose some data, with some luck, only a little.

 

Disk5 besides being unmountable has several pending sectors, since another disk is disable and you can't rebuild it you may try to fix the the filesystem by running xfs_repair without the -n (no modify) flag, if it works you'll want to copy all data from that disk to others.

 

SMART for disk6 looks mostly OK, so instead of rebuilding, and as long as there's no new data there once it got disable, best way forward may be to try and do a new config without disk5 (after copying all data) or use a new disk in it's place, if the parity sync fails copy all data from disk6 to another disk(s).

 

All your 3TB disks have been over 60C, 2 of them over 70C, the failed disk5 has reached 76C!! Surprised it lasted this long.

 

 

Thank you very much for your help. Yes I had heat problem and these old drives got very hot.

Fixing the filesystem did not work. How do I create a new config without disk5? Is there a guide or a wiki I can follow?

 

What was the xfs_repair error?

 

Was the array data unchanged since the various issues? If yes there's one thing you could try, it may or not work.

Quote

Phase 1 - find and verify superblock...
Phase 2 - using internal log
        - zero log...
xfs_repair: read failed: Unknown error -5
empty log check failed
zero_log: cannot find log head/tail (xlog_find_tail=-5)

fatal error -- ERROR: The log head and/or tail cannot be discovered. Attempt to mount the
filesystem to replay the log or use the -L option to destroy the log and
attempt a repair.

 

Yeah the data should be unchanged. The last added data isnt that important anyway if that is the problem.

5 minutes ago, Heciruam said:

The last added data isnt that important anyway if that is the problem.

This procedure only works if all data on the array is unchanged, disk5 looks to be in real bad shape, there was a read error during xfs_repair, also you won't be able to rebuild the disable disk anyway, so and hoping disk6 is OK you can try this since you don't have much to lose, though if you want, you can first try to rebuild disk6 to a spare (don't use the old disk) and see if it works, but it's not likely, so if you want to try do this:

 

-Tools -> New Config -> Retain current configuration: All -> Apply
-assign any missing disk(s), assign all disks as they were, including the old disk6
-check both "parity is already valid" and "maintenance mode" before starting the array
-start the array
-stop array, unassign disk5
-start array, check if disk6 and emulated disk5 mount and contents look correct
 

If yes powerdown, replace disk5 with a new spare, power back up, assign it and start the array to begin rebuild.

 

If the emulated disk5 or disk6 are unmountable run xfs_repair first before doing anything else.

 

So at the moment disk 6 still appears to be missing. I havent done anything yet. Should I try the above even though the disk is displayed as missing? When the array is started Disk 5 is not emulated, but disk 6 is.

Edited November 30, 20178 yr by Heciruam

Follow the instructions above carefully, you need to re-assign disk6 after the new config.

 

 

 

 

Everything seemed to have worked. I didn't have to do the xfs repair. Disk 5 is rebuilding right now and should be ready in 19hrs. At first glance, it looks like I didn't loose any data.

Thank you very much! I could not have done it without your help.

Good, but you're not out of the woods yet, the rebuild may fail if there are issues with any other disk, if it succeeds you should replace some of other disks as soon as possible, mostly those 3TB Seagates that are know to fail a lot in the best of conditions, much more likely after getting so hot.

Yes I will replace them as soon as possible. 

It looks like I lost the majority of the data on disk 5. There are only 100gb left on it. But it doesn't seem to be important data. Should I do any follow up actions, besides changes the bad drives?

 

Thank you again for helping me out!

 

Edited December 1, 20178 yr by Heciruam

Nothing special, try to keep more on top of possible issues, if they affect multiple drives it's likely too late, also don't forget that unRAID is not a backup, make sure you backup your irreplaceable data.