Repository navigation
[BUG] wrong image is set to latest causing issues with upgraded data #541
Description
Activity
Thanks for opening your first issue here! Be sure to follow the relevant issue templates, or risk having this issue marked as invalid.
Unfortunately looks like nextcloud messed up their release as they reversed v34 to a pre-release and our
:latesttag reverted to v33. Little we can do about this.Also, this is an example for why we don't recommend automatic updates and to pin tag versions rather than running latest.
Reacted by sausje, Jens van Almsick, Sid, Martin Kibera, maTTi, EiTiFuzzi, Peter Pirker and zinnderellasame issue here although v34 has locked me into maintenance mode as well :/
Confirming using tag "34.0.0-ls435" worked for me!
Reacted by Peter Pirker, Martin Kibera, mrboobek, maTTi, sausje, EiTiFuzzi, ArtJames and Jon ChristensenAlso, this is an example for why we don't recommend automatic updates and to pin tag versions rather than running latest.
Your're right, but it's also not a good solution to retagging a older version to "latest".
I agree, Nextcloud should have a proper release pipeline. We just relay what their newest non-prerelease is, when they pull a release, a downgrade will happen. It is not the first time Nextcloud has done this great move.
Sigh...I fell for it again and ended up with a new Nextcloud major release on day one...
...when will I learn?Anyhow - my comments here have nothing to do with the docker image.
Sharing for those, like me, who are waking up to an unpleasant surprise.I went straight to
34.0.0-ls437and avoided the shenanigans specified above.
After update, web interface appeared stuck in maintenance mode - but it was not maintenance mode. It was a login error screen.
# grep "ERROR_ID" /data/nextcloud.logto figure out what's happening.
See error related to incompatible apps that the nextcloud update process didn't disable as pasrt of the upgrade process.# occ app:disable camerarawpreviews # occ app:disable suspucious_loginI suspect those encountering the same issue are using suspucious_login.
I think that's enabled by default in most installs.I had the exact same issue after the update and wanted to share what fixed it for me.
The container itself looked healthy, and Nextcloud reported:
version: 34.0.0.12 maintenance: false needsDbUpgrade: falseThe browser only showed a generic "Internal Server Error" page, while the mobile app continued to work normally.
I checked the Nextcloud log with:
tail -100 /data/nextcloud.log
and analyzed the error output with a local LLM. In my case, the actual problem was not Nextcloud itself but the
files_3dmodelviewerapp.The log contained errors similar to:
Call to undefined method OCP\AppFramework\Http\EmptyContentSecurityPolicy::allowEvalScript()originating from:
apps/files_3dmodelviewer/I fixed it by disabling the app:
occ app:disable files_3dmodelviewer
and then running:
occ maintenance:repair
After that, Nextcloud started working normally again.
If anyone else is seeing internal server errors after upgrading to NC34, it may be worth checking
/data/nextcloud.logfor references to incompatible apps before rebuilding the container or rolling back versions.Thanks @PnjDbq & @HartLander
I also had the same "Internal Server Error" page after logging into Nextcloud on version 34.0.0.12 (34.0.0-ls437).
In my case it was the
suspicious_loginapp causing the issue. Disabling that app with the following line fixed it:occ app:disable suspicious_loginNo
occ maintenance:repairor reboot was necessary on my end.This issue has been automatically marked as stale because it has not had recent activity. This might be due to missing feedback from OP. It will be closed if no further activity occurs. Thank you for your contributions.
Is there an existing issue for this?
Current Behavior
Hello. It looks like the "latest" tag here is causing issues for users that have "latest" as their image tag. Just making sure someone is aware because it took down my NextCloud instance and I needed to revert to "34.0.0-ls435" to fix the error. I've pasted the logs from before switching my image from latest manually
Expected Behavior
It looks like the CI bumped 33.0.5-ls436 as the latest or perhaps 34.0.0-ls435 shouldn't be latest yet?
Steps To Reproduce
I experienced this because it seems that Watchtower pulled "linuxserver/nextcloud:latest" when 34.0.0-ls435 was the latest version, then 2 hrs later the latest reverted a major version.
Manual recreation:
For those not doing auto-updates, you are likely unaffected.
Environment
CPU architecture
x86-64
Docker creation
nextcloud: container_name: NextCloud environment: PGID: '1000' PUID: '1000' TZ: America/Chicago image: linuxserver/nextcloud:latest networks: proxy: null ports: - published: 32779 target: 443 restart: unless-stopped volumes: - /var/lib/nextcloud/config:/config:rw - /mnt/container_data/nextcloud/data:/data:rwContainer logs