Hello everyone,

 

Been using unRAID for a few months, with no issues until recent. After having done a CA backup, where it had an error where it said that it failed to complete backup, I was successful to backup my appdata, and flash after a second attempt.

 

After a few hours, an issue I was noticing occur was with Sonarr/Radarr where in their logs, from the docker page, where they mention they only have read access, and was crashing their applications. Following that, Plex has been having the issue where it will respond back with the issue of Permission Denied (included below labeled as PlexLivelog).

 

These issues were yesterday, where shutting down Docker, removing image, and rebuilding the containers seems to have fixed the Sonarr/Radarr problem. Then in the morning, I received a notification of " Call Traces found on your server ", and was recommended to upload the diagnostics to the forum.

 

I've also gone ahead to include a screenshot of Hardware, plugins (if relevant), and another screenshot of all active Containers, with the unRAID version shown (6.3.5).

Fairly certain I have included all immediately relevant files, and information, but please let me know if there is anything further I should include, or wish for clarification.

 

Only thing I can think of that might still be relevant, is the logs from Plex, however I am not sure if just the Crash Report would be useful, or a zip of the entire container within the appdata.

 

Much appreciated for any, and all assistance,

Proxy

 

primarytower-diagnostics-20171029-1230.zip

The docker image is corrupt, delete and re-create.

I can certainly do it again, however yesterday afternoon, I had already done that once I had noticed Sonarr/Radarr was misbehaving.

That leaves me wondering if this will be a re-occuring issue, or what specifically within the image got corrupted a second time within 24 hours.

 

*edit* Once I delete, and re-create the docker image, would it be beneficial to you, for me to supply an updated diagnostics zip?

Edited October 30, 20178 yr by ByteMyProxy
Updated with question

The image is getting corrupt because you're running out of space, your cache filesystem is fully allocated and that's no good, see here:

https://forums.lime-technology.com/topic/54533-624-cache-drive-reporting-quotno-space-left-on-diskquot/#comment-532931

Understood, and thank you for the fast response. Following along with the btrfs commands from earlier in that thread, this is the output from my server:

btrfs fi show /mnt/cache
Label: none  uuid: 68ca2f39-565f-40c0-b3f0-04f355247cc8
        Total devices 2 FS bytes used 25.17GiB
        devid    1 size 476.94GiB used 476.94GiB path /dev/nvme1n1p1
        devid    2 size 476.94GiB used 476.94GiB path /dev/nvme0n1p1

btrfs fi df /mnt/cache
Data, RAID0: total=947.81GiB, used=24.78GiB
System, RAID1: total=32.00MiB, used=96.00KiB
Metadata, RAID1: total=3.00GiB, used=396.20MiB
GlobalReserve, single: total=56.28MiB, used=0.00B

Then continuing from what you have recommended with balancing the drive, here is the outcome:

btrfs fi show /mnt/cache
Label: none  uuid: 68ca2f39-565f-40c0-b3f0-04f355247cc8
        Total devices 2 FS bytes used 24.95GiB
        devid    1 size 476.94GiB used 68.00GiB path /dev/nvme1n1p1
        devid    2 size 476.94GiB used 68.00GiB path /dev/nvme0n1p1
btrfs fi df /mnt/cache
Data, RAID0: total=129.94GiB, used=24.58GiB
System, RAID1: total=32.00MiB, used=16.00KiB
Metadata, RAID1: total=3.00GiB, used=381.78MiB
GlobalReserve, single: total=41.94MiB, used=0.00B

Already I can see the difference, and see that the issue shouldn't be as often in the 6.3 versions. After deleting, and re-creating the Docker image, is there anything else I can do proactively to see that this doesn't happen again? Only thing I see as an option is checking the Cache allocation, then running a re-balance on it, hopefully before it breaks anything again.

This issue should be fixed once we get to kernel 4.14, but until then I recommend running this balance weekly, you can use the user scripts plugin to schedule it:

 

btrfs balance start -dusage=75 /mnt/cache

 

Sweet, I have that set up now. Then as a follow up question just to be certain, when I am recreating the images, is it still fine for me to use the user-templates for each container, or is it best practice for me to go into the community apps, and build from a blank slate?

 

I still have that backup of the appdata from the second time I ran it a few days ago, however I'm not certain restoring from that won't cause any issues. Not that I see any reason why it would, but from reading online, people seem to recommend a few different ways to rebuild after deleting the docker image.

 

Much appreciated for all of your help!

28 minutes ago, ByteMyProxy said:

is it still fine for me to use the user-templates for each container

 

Yes, that's fine.