January 21, 20251 yr After a healthy fight getting my 3900X to behave in Unraid and get the parity back to a valid state, I went ahead and updated to Unraid 7.0 in hopes that it would solve some of the stability issues with the new CPU. It has! However, while updating my containers, I started getting errors installing any update where it immediately failed claiming there is already a container with the same name (didn't catch the exact error unfortunately). Gave the server a restart, and now Docker is dead entirely, showing the error "The Docker service failed to start." I've attached a diagnostic pull. Any help is appreciated. Thanks! mnemosyne-diagnostics-20250120-1941.zip
January 21, 20251 yr Community Expert You may neeed to install the amd reset buf fix plugin. Per your diag and docker settingand per the release note. I have found that using a xfs docker iamge. I see that you have the unbalance plugin installed. Make sure it didn't break the docker image file. Als make sure your "/mnt/user/system/" under teh sahre tab only has a primary storage set no secodary storage... Under docker settings you will need to remake the docker.img file. Recomended options: As you are using vfio if pcie slots have moved verifio that iommu and correct devices as set under system tool... I'm not seeing anythin in the syslog that would indicate a failure so I ahave to assume docker image coruptions...
January 21, 20251 yr Author Hello! Appreciate the response. I went to double check that Unbalance or even the Mover didn't break the Docker image and I found the solution was silly and simple... the drive that the Docker image is stored on had run out space entirely. Thought I had a 10GB free space buffer but apparently not, I only had a few kilobytes free and starting Docker was throwing I/O errors on disk3, the drive the image is on. After freeing up several gigs, were back in action. Appreciate the help!
January 21, 20251 yr Community Expert Solution 4 hours ago, GazaIan said: Hello! Appreciate the response. I went to double check that Unbalance or even the Mover didn't break the Docker image and I found the solution was silly and simple... the drive that the Docker image is stored on had run out space entirely. Thought I had a 10GB free space buffer but apparently not, I only had a few kilobytes free and starting Docker was throwing I/O errors on disk3, the drive the image is on. After freeing up several gigs, were back in action. Appreciate the help! If the pool is also being used for caching purposes, then you should make sure that you set a sensible Minimum Free Space value. Typically something like twice the size of the largest file you intend to cache (or more). This will mean that the 'cache' functionality starts by-passing the pool when free space falls to this value so avoiding such issues.
January 21, 20251 yr Author Yeah, I'm going through all my shares now and for whatever reason all of my minimum free space values are missing for every single share. Not sure why but given all the random troubleshooting I've had to do in general, I'm sure it just got caught up in the mix. Thank you for the help!
January 21, 20251 yr Community Expert 1 hour ago, GazaIan said: Yeah, I'm going through all my shares now and for whatever reason all of my minimum free space values are missing for every single share. Not sure why but given all the random troubleshooting I've had to do in general, I'm sure it just got caught up in the mix. Thank you for the help! The default for has traditionally been 0. The latest Unraid releases try to set a more sensible default if it is spotted to be 0 for a share. However the setting on shares is primarily about writing to the main array. Each pool also has its own setting to stop the pool over-filling and the default there remains at 0.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.