gellux - docker image corrupt

[gellux]

tried to install and get "

Error: open /var/lib/docker/tmp/GetImageBlob678726820: read-only file system"

 

had this problem with a few plugins and dockers, though don't really know what the other options are.

[trurl]

4 hours ago, gellux said:

tried to install and get "

Error: open /var/lib/docker/tmp/GetImageBlob678726820: read-only file system"

 

had this problem with a few plugins and dockers, though don't really know what the other options are.

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your next post.

 

I will probably split this into its own thread since it is likely unrelated to this docker.

[gellux]

12 minutes ago, trurl said:

Go to Tools - Diagnostics and attach the complete diagnostics zip file to your next post.

 

I will probably split this into its own thread since it is likely unrelated to this docker.

 

nas-diagnostics-20190411-1329.zip

[trurl]

Your docker image is corrupt, but it isn't currently full. Have you had issues with filling it up?

 

You will have to delete, recreate, and reinstall your dockers using the Previous Apps feature on the Apps page. While you are recreating it you should also turn on log rotation. We can get into the details of all that later if it still seems appropriate, but there are some larger changes I would recommend for your total configuration.

 

You have an SSD in the parity array. Definitely not recommended for several reasons. It can only be written at the speed of parity, it cannot be trimmed, and there is some debate as to whether it can invalidate parity depending on how it manages unused storage internally.

 

Your cache is a bit small. I recommend not caching any user share writes with such a small cache. Currently only system, domains, and appdata are on cache. Set all other user shares to Use cache disk: No. You need to set these to cache-no before writing any more data to them, or you will wind up with some of it stuck on that small cache.

 

Personally, I think I would use that 500G nvme disk2 you have and assign it as cache, and just forget about that 120G cache. You would have to move all its contents to disk1, but disk1 has plenty of capacity for that. Then you would have a really nice cache with plenty of capacity for dockers, VMs, and caching writes to your user shares.