Hi everyone,

For some reasons I decided not to wait and upgrade my server to version 7.0.0.

After the upgrade I was able to check and everything seemed good.

But..

This morning, I saw that none of my services hosted with docker were working, upon connection, I see that the docker tab says "Docker Service failed to start.".

When rebooting the server, Everything, including docker starts as normal, but after some time, the docker service will go down again.
The array remains up and I can still access the files shares from clients on my LAN.

Did someone encounter the same type of issues ? I'm looking for a solution other than rolling back which I'll do if needed.
 

millesime-diagnostics-20250111-1941.zip

There's a problem with the pool filesystem, and it's going read-only:

 

Jan 11 18:18:31 millesime kernel: BTRFS critical (device nvme0n1p1): corrupt leaf: block=150432333824 slot=95 extent bytenr=89176625152 len=4096 invalid data ref root value 15731073498405142533
Jan 11 18:18:31 millesime kernel: BTRFS error (device nvme0n1p1): read time tree block corruption detected on logical 150432333824 mirror 1
Jan 11 18:18:31 millesime kernel: BTRFS error (device nvme0n1p1: state A): Transaction aborted (error -5)
Jan 11 18:18:31 millesime kernel: BTRFS: error (device nvme0n1p1: state A) in __btrfs_free_extent:3116: errno=-5 IO failure
Jan 11 18:18:31 millesime kernel: BTRFS info (device nvme0n1p1: state EA): forced readonly

 

With btrfs, recommend backing up and reformatting, nothing to do with upgrading to v7, other than possibly the newer kernel detecting a problem that was going undetected before.

Thanks for the input, how can I do that ? You meant backing up the entire array and reformating the disks ?

No, just the cache pool.