Running fsck/e2fsck on Synology NAS

I opened a ticket with Synology Support because some ‘Hyper Backup’ data backup tasks were failing.  The tech there were errors on the volume, which is odd because it hasn’t been powered off incorrectly and I just created it maybe 2-3 weeks prior.  The recommendation: move the data off the volume, delete the volume and recreate.
No thanks.

It really irks me when they don’t even bother to troubleshoot.

Synology doesn’t provide any command line support so you’re really at the mercy of the tech you get and the community forum.  And since the box lacks lsof and fuser (unless you install ipkg which I’m not interested in doing just yet) it makes tracking down which process is tying up the volume a real pain.

Enter syno_poweroff_task.

TL;DR

This process worked well for me on DSM 6 Beta 2 but your mileage may vary.

I recommend you run e2fsck in read-only mode first before attempting to fix anything.

 

# reboot the NAS
# when it comes up, ssh into it
# login & su to root
# run the following and sit tight; reconnect if you're disconnected
syno_poweroff_task -d

# activate the volume
vgchange -ay

# play it safe - replace with the correct logical volume
e2fsck -fnvtt /dev/vg1/volume_1

# run e2fsck for real
e2fsck -Dfttvy -C 0 /dev/vg1/volume_1

# go order a pizza and watch a movie
# run the command again to be sure everything is clean
# reboot and all should be fine
 # if you're having //serious// issues, you can try the following
# this is more of an overnight + work day + weekend task.
e2fsck -ccDfttvy -C 0 /dev/vg1/volume_1

Running e2fsck cleared the volume errors but the issues I was having continued.  I ended up recreating the data backup task in Hyper-Backup and it worked.

How Did I Get Here?

During my 2014 Thanksgiving I had a massive data loss scare that brought to light my lack of an adequate backup solution for my – well – backup.  I was able to recover and I’ve since implemented a bare bones backup strategy (rsync to a local NAS) and I’m working up to storing on Azure.

After upgrading my DS411+ii to DSM 6 Beta 2, I’ve been kicking the tires on Hyper Backup Synology’s ‘Backup and Restore’ replacement which so far impresses.  And by impresses I mean it seems to be working.  After running the tasks nightly for maybe a week or two, one task failed.  A little while later, another task failed; then another.  Clearly, something is wrong somewhere with the data I’m trying to backup, and because logs are slim pickins, it’s impossible for me to make any real progress.

This is why I opened the ticket with Synology to hopefully have them troubleshoot further but they darn near shotgunned it.

 

After fixing some minor issues, the tasks still failed so I resorted to removing the task, deleting the existing data and starting over.  This worked, so far, but I’m concerned there may be a larger underlying issue here.  All drives were replaced within the last year with one failing 10 days before the 1 year mark.  Hopefully it’s something silly like a strangely named file – we’ll see.

Good Providence to you if you go down this route!

 

5 comments

  1. A little late but many thanks. I was struggling with an I/O error for almost a year on my ds411j, and none of the other solutions I tried worked for me. Yours did!

    Like

  2. Thanks so much for this post. I was able to fix my filesystem error on my Synology 1815+ SHR 50TB raid. I did have to run it a few times but after the reboot it’s back to normal again.

    Liked by 1 person

Leave a comment