That's a good idea but be aware that it may not fix your current problem, though it may help, but hopefully it will prevent more in the future.

Do you think that new failed drive 4 caused corruption while rebuilding 13 if it was failing during that?

I believe the file system corruption was probably caused by the constant call traces/crashes related to the SASLP, they are know to have issues with unRAID v6 and should be avoided when possible (same for the newer SAS2LP).

 

I swapped out the controllers with the perc's today.  now im back to the drive 13 being unrecognized

 

attached new diags

hoser-diagnostics-20170913-2145.zip

Looks like the disk is now being correctly identified, since you still have a disable disk you can do this, you need to have a spare disk connected for disk4:

 

-Tools -> New Config -> Retain current configuration: All -> Apply
-assign new spare in disk4 slot and any other missing disk(s)
-check both "parity is already valid" and "maintenance mode" before starting the array
-start the array
-stop array, unassign disk4

-start array

 

If both the emulated disk4 and disk13 are still unmountable try running xfs_repair again on them.

the old disk 4 showed back up, i can reassign that.  i also have a new disk i bought.  should i try the orig one first and see if it throws tons of errors again?

4 minutes ago, jhanda said:

should i try the orig one first and see if it throws tons of errors again?

 

It would be best to use a new disk, if the emulated disk is still unmountable and xfs_repair can't fix it you can try to recover the data from the old disk.

Did that, disk 13 still gives Structure needs cleaning errors.  no disk4 shows up at all emulated or otherwise

xfs repair on disk13 seems to be failing again been searching for an hour

7 hours ago, jhanda said:

no disk4 shows up at all emulated or otherwise

 

Then you didn't follow the procedure correctly.

 

5 hours ago, jhanda said:

xfs repair on disk13 seems to be failing again been searching for an hour

 

This is what I expected, whatever the cause of the filesystem corruption is it happened in the past.

what do i do now?  I just want it back up and save whatever i can off the failed drives

If xfs_repair can't fix disk13 not much else you can do, you can still try to fix disk4 by doing what I posted earlier:

 

22 hours ago, johnnie.black said:

-Tools -> New Config -> Retain current configuration: All -> Apply
-assign new spare in disk4 slot and any other missing disk(s)
-check both "parity is already valid" and "maintenance mode" before starting the array
-start the array
-stop array, unassign disk4

-start array

 

If that doesn't work try recovering data from the old disk4.

I did that, it allowed the disk13 to go back into the array but no disk 4

 

just tried it again, no disk 4 after i unassign and restart

is there a way i can restart the thing without 13 or 4 so it just works and then try to copy off what data i can and then reformat those disks and add them back in?

1 minute ago, jhanda said:

just tried it again, no disk 4 after i unassign and restart

 

Then you're not doing it right, you need to use a new spare disk to assign to disk4 slot.

1 minute ago, jhanda said:

is there a way i can restart the thing without 13 or 4 so it just works and then try to copy off what data i can and then reformat those disks and add them back in?

 

Yes, do a new config, and assign all disks except disks 4 and 13, then start a parity sync (or don't assign parity for now).

Edited September 15, 20178 yr by johnnie.black

3 minutes ago, johnnie.black said:

 

Then you're not doing it right, you need to use a new spare disk to assign to disk4 slot.

trying it one more time.. pretty sure i did it just as you said.

nada..

 

so unassign 13 and 4, new config parity valid?

 

then plug in the corrupted drives and see what i can get?

38 minutes ago, jhanda said:

nada..

 

Again, you're doing something wrong, you may end up with an unmountable/unfixable emulated disk4, but if done correctly you have to end the procedure with an emulated disk4, lets try this one last time, tell me at what step you can't follow/or end up without a disk4:

 

1-Tools -> New Config -> Retain current configuration: All -> Apply
2-assign new spare in disk4 slot and any other missing disk(s)
3-check both "parity is already valid" and "maintenance mode" before starting the array
4-start the array
5-stop array, unassign disk4

6-start array

Edited September 15, 20178 yr by johnnie.black

it happens at step 6.  after i unassign 4 in step 5 it says missing disk, check here to start array with missing disk, insert a replacement asap.  then when i start it it just comes up valid with no disk 4 instead of in a degraded state with an emulated disk

I give up.

I'm ready to give up too, I really appreciate all the help.  If anything I can tell the thing works alot better with the perc's in there.  I used to have to reset a few drives every time it booted.  Was I correct in that I just do a new config without 13 or 4 and say parity valid?  Am I able to plug those other drives in then and just attempt to copy what I can back to the array?

 

I think I will shrink the array by 1 and add a 2nd parity drive if thats possible

The only way the procedure is not working is if you're doing something wrong or some other unusual problem on your system.

 

2 minutes ago, jhanda said:

Was I correct in that I just do a new config without 13 or 4 and say parity valid?

 

Not correct, if you remove any disks(s) parity is no longer valid.

This is what happens

Ok, it's working, you have the emulated disk, disk4 is unmountable and that is expected since it already was before, now try running xfs_repair on it.

 

Stop the array, start the array in maintenance mode, SSH into your server and type:

 

xfs_repair -v /dev/md4