SnapRAID Fix No space left on device.

GDawg · Jun 13th 2025

I have a SnapRAID-setup with four data-disks and two parity-disks of the same size.
I just now had two disk failures (one data and one parity) and replaced them both.
Uppon running "Fix", I had the following error regarding one of the old, working disks:

Bash: snapraid fix

Reading data from missing file '/srv/dev-disk-by-uuid-a1b90601-c219-4ade-bcd9-16f98fe1458e/docker_storage/app_stack/nextcloud/gdawg/files/filename' at offset 4866179072.
Missing data reading file '/srv/dev-disk-by-uuid-d75fec5f-4bf4-470a-b3e7-2395cc49a136/snapraid.2-parity' at offset 21228158976 for size 262144.
Error writing file '/srv/dev-disk-by-uuid-a1b90601-c219-4ade-bcd9-16f98fe1458e/docker_storage/app_stack/nextcloud/gdawg/files/filename'. No space left on device.
WARNING! Without a working data disk, it isn't possible to fix errors on it.
Stopping at block 80979

   18565 errors
   18563 recovered errors
       1 UNRECOVERABLE errors
DANGER! There are unrecoverable errors!

END OF LINE

Display More

I now seek advice on the best next course of action.
I would like to run "Sync" as soon as possible to be safe, but I am unsure if that is wise while that error remains or if i should clean it out first - and if so, how?
(I do have a backup of the 1 file that seems to be concerned by the issue, so there would be no issue in an action that would lead to losing it.)

gderf · Jun 13th 2025

You need to follow the recovery procedure exactly as described in the SnapRAID manual and FAQ available here: https://www.snapraid.it/

Running a sync before completing a recovery will eliminate any chance of recovering the lost data.

GDawg · Jun 19th 2025

Thank you for your answer!

Sorry for taking so long to answer, I wanted to do more research and try more things out (and some "fix" runs took literal days) before answering. Also sorry if some of my questions seem very basic, but I can't for the life of me even just find the meaning of some of these messages online.
I followed the "Step-by-Step guide in case of Disk error" on this page: https://forum.openmediavault.o…te-replace-disk-v1-2-pdf/

and it seems to me that it corresponds to the "4.4 Recovering"-Section in the Manual.

"Fix" on new Disks

The only difference seems to be that it tells me to specify the replaced disks, which results in no errors for the new replacement data disk:

Bash: snapraid -d (new data disk name) fix -U

Initializing...
Selecting...
Fixing...
Nothing to do
Everything OK

But lots of errors on the new parity disk:

Bash: snapraid -d (new parity disk name) fix -U

#On the first run:
Missing data reading file '/srv/dev-disk-by-uuid-d75fec5f-4bf4-470a-b3e7-2395cc49a136/snapraid.2-parity' at offset 75767218176 for size 262144.
#(on the new data disk)
Missing file '/srv/dev-disk-by-uuid-a1b90601-c219-4ade-bcd9-16f98fe1458e/filepath'.
#(on the same pre-existing data disk that is showing the error in the first post in this thread)

#After multiple runs:
error:3787683:data4:filepath: Data error at position 697, diff bits 63/128
entry:0:block:known:bad:data4:filepath:697:
#(data4 being the same pre-existing data disk that is showing the error in the first post in this thread - many such cases)
msg:fatal:
msg:fatal: Stopping for interruption at block 3861532
sigint:3861532: SIGINT received
msg:status:
msg:status:   446831 errors
msg:status:        0 recovered errors
msg:status:        0 unrecoverable errors
msg:status: Everything OK
summary:error:446831
summary:error_recovered:0
summary:error_unrecoverable:0
summary:exit:recovered

Display More

"Fix" on pre-existing disks

As I also had issues on the already pre-existing disks, i ran "fix" as well for the pre-existing parity disk, and for the pre-existing disk that had the error in the first post of this thread.

Bash: snapraid -d (pre-existing data disk name) fix -U

Initializing...
Selecting...
Fixing...
Error writing file '/srv/dev-disk-by-uuid-a1b90601-c219-4ade-bcd9-16f98fe1458e/docker_storage/app_stack/nextcloud/gdawg/files/filename'. No space left on device.
#(same file as in the original post)
WARNING! Without a working data disk, it isn't possible to fix errors on it.
Stopping at block 80979

   18564 errors
   18563 recovered errors
       1 UNRECOVERABLE errors
DANGER! There are unrecoverable errors!
FAIL

Display More

Bash: snapraid -d (pre-existing parity disk name) fix -U

Reading parity data outside range at extra offset 2537951592448.
100% completed, 2369989 MB accessed in 10:02    

11583917 errors
       0 recovered errors
       0 unrecoverable errors
Everything OK

Summary

All it says in the manual about fixing is that:

Quote

If you are not satisfied of the recovering, you can retry it as many time you wish.

I can hardly tell when to be "satisfied" and what I should do if I am.

I don't understand why snapraid tells me that most errors are not "unrecoverable", but still doesn't seem to recover them.

So what should I do to recover the errors that I can?
And how do I manage the errors that I accept I can't recover from to return to a stable state?

gderf · Jun 19th 2025

I suggested that you follow the recovery procedure available in the SnapRAID manual exactly. But you decided not to. I can't offer any different help.

SnapRAID Fix No space left on device.

ryecoaaron Jun 13th 2025

"Fix" on new Disks

"Fix" on pre-existing disks

Summary

Participate now!

Tags