Portainer CE to BE upgrade failed, it's now in a CrashLoopBackOff state

Question

I just wanted to try the self-service 3-nodes trial of portainer BE on my cluster, and apparently something went wrong.

All I really did on that cluster so far was installing Portainer CE, spinning up some application with it, manually adding a environment (second kubernetes node) and then upgrading to Portainer BE (using Portainers own functionality in-web-app) in hopes of setting up RBAC, a LB and a third node using it's features.

These are the logs I get when the container starts:

2023/12/01 01:29PM INF github.com/portainer/portainer/api/cmd/portainer/main.go:369 > encryption key file not present | filename=portainer
2023/12/01 01:29PM INF github.com/portainer/portainer/api/cmd/portainer/main.go:392 > proceeding without encryption key |
2023/12/01 01:29PM INF github.com/portainer/portainer/api/database/boltdb/db.go:125 > loading PortainerDB | filename=portainer.db
2023/12/01 01:29PM FTL github.com/portainer/portainer/api/datastore/migrator/migrate_ce.go:101 > the Portainer database is set for Portainer Business Edition, please follow the instructions in our documentation to downgrade it: https://documentation.portainer.io/v2.0-be/downgrade/be-to-ce/ |

Please ignore the encryption part as the cluster is not used in prod/only accessed by 2 people so far.

How can I rescue this? Should I just setup the portainer namespace again using BE from the start?

Answer 1

Shortly after asking this, i tried updating and reapplying the helm repo over the existing one and fixed it that way.

The generic command would be:

helm repo update
helm upgrade -n portainer portainer portainer/portainer --set enterpriseEdition.enabled=true

But since I am using microk8s in my case it would be:

kubectl helm3 repo update
kubectl helm3 upgrade -n portainer portainer portainer/portainer --set enterpriseEdition.enabled=true

This seems to have fixed the issue. I will try to get in touch with portainer to see if I can provide any logs regarding this.