[Config Support]: Database or disk is full #15661
Replies: 1 comment
-
I continued trying to troubleshoot things and clean other things up. I went through the process to remove unused docker images (older versions of running containers with new version images). After doing that, my mountpoint of the in-use space for the running containers (all the /mnt/server references in the docker config) had not changed... it still showed 100 GB free. I'm sure another mount point whose usage I was not monitoring reduced some, but no mount point was so full it couldn't take another several hundred MB of files, and as previously indicated, the DB file in /frigate/config was only around 45 MB. For kicks and grins, not expecting it to work, I started Frigate, and it came up. The current DB is now 25 MB instead of 45 MB, and the 300 GB mount point where I keep the running config files (and media) reduced from around 64% utilized down to 58% utilized. I'm assuming that is because things weren't working for long enough that my 3-day retention got hit, and a lot of stuff got deleted, after the system had been down for a few hours and possibly not functioning properly prior to that for who-knows-how-long. |
Beta Was this translation helpful? Give feedback.
-
Describe the problem you are having
Frigate won't fully start up. I do see a "database or disk is full" message in the logs during startup right after an indication that vacuum is starting.
Frigate was running (mostly) fine and had been running I think at least a week or more without issue. I started having a minor issue where the history was not loading, so I thought I should restart Frigate, and see if that cleaned anything up, but during restart, I noticed this message, and the restart loop began.
I'm running Frigate as a docker container in Ubuntu 24.04 LTS on hardware that also has home assistant and other containers. The 300 GB disk where all of the containers have their data was only 66% full when the issue began, and I've knocked that down to 64% (no change in startup behavior after that). The database is only about 45 MB, so I would expect there would be plenty of space (over 100 GB) for whatever this "vacuum" process does.
I've reviewed a similar article from Feb of this year, and I don't see a resolution on that one, and I believe I have my config set up appropriately (maybe not the most efficient and safe, but workable). I only have three cameras, and only keep 3 days of snapshots and recordings at this time. And, as indicated, with only a 45 megabyte database, 100 gigabytes of free space should be plenty.
Is there a parameter that limits the database to a specific size? I can see in my frigate/config folder, there is a backup.db file from back in Aug that is over 70 MB... so, I'm guessing that might have been a backup that I took during one of the major upgrades.
Version
0.14.0
Frigate config file
Relevant Frigate log output
Relevant go2rtc log output
Frigate stats
No response
Operating system
Other Linux
Install method
Docker Compose
docker-compose file or Docker CLI command
Object Detector
Coral
Screenshots of the Frigate UI's System metrics pages
No response
Any other information that may be helpful
No response
Beta Was this translation helpful? Give feedback.
All reactions