Can't run cloudron update for an app in error state

[Lanhild]

@ekevu123 I never faced this situation. Maybe try putting it in recovery mode?

[ekevu123]

I can't, because the app is stuck in "unable to pull image" state.

An assumption of what I think might work:

1. Delete the app
2. Re-install it from scratch
3. If it has the same name, previous backups appear?
4. Then restore?

But I'd need to know first if this works.

[Lanhild]

@ekevu123 uninstalling won't preserve the backups

[ekevu123]

The Cloudron team can say for sure, but I believe it does - backups don't get deleted when you uninstall an app. I think I used this method before in a different setting.

[Lanhild]

@ekevu123 I meant, you won't see the previously uninstalled app backups in the newly installed app, but yes backups won't get deleted

[ekevu123]

I found a solution for the problem: Since the app was looking for a docker tag, the solution was to tag the new image with the tag from the missing backup image, which got the app up, and then perform a proper update with the same image.

[joseph]

@ekevu123 said:

The Cloudron team can say for sure, but I believe it does - backups don't get deleted when you uninstall an app. I think I used this method before in a different setting.

this is correct/

[rosano]

@ekevu123 I'm in the same situation now and I'm wondering if your registry is on a separate Cloudron or everything's on one? I have issues whenever my Cloudron makes requests to URLs hosted on the same Cloudron and now it seems I can no longer install because the image can't be reached for that reason

[nebulon]

Is hairpin working on your system? Check with cloudron-support --troubleshoot

[rosano]

@nebulon I don't see any reference to hairpin:

[OK]	node version is correct
[OK]	IPv6 is enabled and public IPv6 address is working
[OK]	docker is running
[OK]	docker version is correct
[OK]	MySQL is running
[OK]	netplan is good
[OK]	DNS is resolving via systemd-resolved
[OK]	unbound is running
[OK]	nginx is running
[OK]	dashboard cert is valid
[OK]	dashboard is reachable via loopback
[OK]	No pending database migrations
[OK]	Service 'mysql' is running and healthy
[OK]	Service 'postgresql' is running and healthy
[OK]	Service 'mongodb' is running and healthy
[OK]	Service 'mail' is running and healthy
[OK]	Service 'graphite' is running and healthy
[OK]	Service 'sftp' is running and healthy
[OK]	box v9.1.5 is running
[OK]	Dashboard is reachable via domain name

[nebulon]

ah maybe I misunderstood your issue then. So you are not able to reach the docker registry to install apps or are you referring to a specific registry used by some community packages? Maybe the thread is mixing issues also, do you have more info and error output what exactly does not work?

[rosano]

@nebulon said:

So you are not able to reach the docker registry to install apps or are you referring to a specific registry used by some community packages?

My registry is on the same Cloudron. I have built images with the Builder app and pushed there to install custom apps in the past. But it seems recently my Cloudron can no longer make network requests to the same Cloudron (maybe I've misconfigured something?).

Maybe the thread is mixing issues

I suppose for this thread I would only want to know if there's a way to force update an app in error state. The network issue is documented here. I'm currently very stuck on updating those custom apps when the registry is inaccessible and would appreciate any help.

[rosano]

rosano said:

I'm currently very stuck on updating those custom apps when the registry is inaccessible and would appreciate any help.

I've solved my issue by pushing an image to an external registry (docker hub) and then tagging the unreachable image:

docker pull alfa/bravo:charlie
docker tag alfa/bravo:charlie docker.cloudron.example.com/bravo:latest
docker push docker.cloudron.example.com/bravo:latest

But I've learned that I need to phase out the self-hosted Docker registry due to those same-server network issues and migrate to Cloudron's CLI's built-in uploading and building.

[joseph]

@rosano some users claimed that this exact same issue was resolved by upgrading to ubuntu 24 . https://forum.cloudron.io/topic/15265/unusable-application