Hello,

 

Just tried for the second time to update my unraid server from latest 6.12. version to newly released stable 7.0.0., and sadly I have a problem with my cache drive.

After a reboot it says, that my cache drive (for system, appdata and domains) is not mountable, and it wants me to format it to make it work, but after that I will lose all of my data, so is there any solution for this situation?
I know, that I can always backup my system, but that's not how it supposed to be right?

Below screens and diagnostic file attached.

 

Thank You

 

unraid-diagnostics-20250110-1403.zip

Edited January 10, 20251 yr by rogales

Blkid is not detecting a valid filesystem, post the output from:

 

wipefs /dev/nvme0n1

 

And please note that despite the name, this command as written, won't wipe anything

 

 

That's the output:

Edited January 10, 20251 yr by rogales

Sorry, need the output from the partition, not the device, it should be

 

wipefs /dev/nvme0n1p1

 

That's not a problem. Here:

 

root@Unraid:~# wipefs /dev/nvme0n1p1
DEVICE    OFFSET  TYPE  UUID                                 LABEL
nvme0n1p1 0x3     ntfs  773F1B2F62FEC1E9                     
nvme0n1p1 0x10040 btrfs a98d1a74-74e2-42da-9e6d-759d22128d15 
nvme0n1p1 0x1fe   dos  

 

Edited January 10, 20251 yr by rogales

Yes, there are multiple filesystem signatures, so blkid doesn't retune any.

 

You can just delete the extra signatures with wipefs, this should be perfectly safe, but just in case, recommend making sure backups are up to date of any important data.

 

If you need to backup, you can mount the pool manually with:

mkdir /x

mount -t btrfs /dev/nvme0n1p1 /x

 

You can then see the pool contents under /x and copy the data somewhere else.

 

 

To correct the issue type:

 

wipefs --backup -a -t ntfs,dos /dev/nvme0n1p1

 

After that, confirm only the btrfs signature is left, and if yes, v7 should now mount the pool.

 

 

 

 

 

Should I proceed? If so, with argument?

--force

 

wipefs --backup -a -t dos --force /dev/nvme0n1p1

?



And if that's the proper command, what should I do next? Just restart SERVER?

Edited January 10, 20251 yr by rogales

Before doing that, post the output from

 

blkid /dev/nvme0n1p1

 

In case the dos sig is not causing issues

 

i also updated from latest 6.12 to 7, and now the array is not starting "Wrong Pool State-cache - invalid expansion" should i open another thread?

Just now, neo404 said:

i also updated from latest 6.12 to 7, and now the array is not starting "Wrong Pool State-cache - invalid expansion" should i open another thread?

 

2 minutes ago, JorgeB said:

Before doing that, post the output from

 

blkid /dev/nvme0n1p1

 

In case the dos sig is not causing issues

 

 

root@Unraid:/# blkid /dev/nvme0n1p1
/dev/nvme0n1p1: UUID="a98d1a74-74e2-42da-9e6d-759d22128d15" UUID_SUB="be4a9779-ad86-4243-a4cb-8915add09dbb" BLOCK_SIZE="4096" TYPE="btrfs" PTTYPE="dos"

 

Edited January 10, 20251 yr by rogales

See if it mounts now, it probably will.

It worked.
Thank You for Your time, and help.

Hello...

 

I´ve got a problem (wrong pool state).. Anyone know how to fix??

@Mattti1912check this link: 

 

@JorgeB, one last question; Can You explain to me, why something like this happened? I'm just curious, why (as for now) only I have had such a problem.

It usually means the device was not correctly wiped when it was formatted, so the signature for the previous fs was still present, I've seen this happen before, mostly with array devices.

Hi first-time poster and a noob.

It seems since I tried to update to Unraid 7, I have had a similar problem to the above with my cache pool. I managed to downgrade and get the tower running.

What should I do?

Thanks in advance.....

44 minutes ago, Crazy_B said:

What should I do?

Please post the diagnostics.

1 minute ago, JorgeB said:

Please post the diagnostics.

Thanks for getting back.

tower-diagnostics-20250116-2240.zip

Jan 16 03:18:06 Tower kernel: BTRFS error (device sdb1): read time tree block corruption detected on logical 25265799168 mirror 1

 

This means there is filesystem corruption on the pool, so it aborts, was 6.12.14 the release you were using before?

Yes.

I tried to update to 7 and couldn't get Docker containers to start. After several reboots, I couldn't log back into the server. Therefore, after reading a few community posts, I restored the flash drive and downgraded it to 6.12.14, hoping it would fix the issue. But it didn't, and I found this post.

 

3 hours ago, Crazy_B said:

I restored the flash drive and downgraded it to 6.12.14

If this is the release you were using before updating, and it still doesn't mount, you can try using these recovery options:

 

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/#findComment-543490

 

I will try it tonight. Thanks.

If I still want to update to version 7, which I did initially and ran into the same issue, is there anything I could do?

2 hours ago, Crazy_B said:

If I still want to update to version 7, which I did initially and ran into the same issue, is there anything I could do?

 

If you manage to backup the pool, you can then reformat it and restore the data, and it will no longer be a problem.