Repository navigation
|
Hi, I am in a loop, the following is in the docker log of the nextcloud container: 2026-11-11T08:24:03.947008804Z [nextcloud-aio-nextcloud] 2026-11-11 09:24:03.945218+01 So I restored the backup last night, restarted the containers, all working again and this morning it starts all over (3 restores allready, so it keeps repeating). |
Replies: 2 comments 1 reply
Why the loop happens (and why the log looks empty after restoring)In
How to find the root cause and break the loop
|
|
I've found the cause: Strangely enough the filesarchive app was inactive and thus also reported in the logs, removing fixed the upgrade. I am not on the beta channel, so it looks like 35.0.2 has been released for a broader audience, and more users will face this. Side question: is it possible to leave the choice of moment in upgrading by the admin instead of surprising the admin with not a riskfree upgrade? Is a notification possible that the upgrade is pending approval of the admin? In my opinion this would prevent unwanted outages. |
Why the loop happens (and why the log looks empty after restoring)
In
Containers/nextcloud/entrypoint.sh, when Nextcloud AIO runs an update:update.loginside your Nextcloud data directory (exec > >(tee -i "/var/www/html/data/update.log")).update.failedin$NEXTCLOUD_DATA_DIRand exits on every subsequent container start.skip.updateso the container starts cleanly on the backed-up version—and the restore overwritesupdate.log, which is why it looks empty afterward.image_version > installed_version, retries the…