Issue with restoring backup on Kubernetes 1.15.4

Infos:

Issue with moving and restoring a backup from our old docker based (setup with docker-compose) installation to a kubernetes based (setup with helm chart)

  • Used Zammad version:
    Image: zammad/zammad-docker-compose:zammad-3.1.0-10
  • Used Zammad installation source: helm chart zammad-1.2.1
  • Operating system: Kubernetes 1.15.4

Expected behavior:

  • Make a backup from a docker based installation (setup with docker-compose)

  • Running the zammad_restore.sh in the kubernetes based installation with DB & Files backups in /tmp/zammad_backup creates a working system

Actual behavior:

What’s up with those “zammad: unrecognized service” lines?

Steps to reproduce the behavior:

  • Get a backup from a dockersystem setup by docker-compose
  • Setup a new installation with helm on Kubernetes
  • Copy files to /tmp/zammad_backup, rename config.dist in /opt/zammad/contrib/backup
  • Run zammad_restore.sh

Without knowing how kubernetes works: I guess it’s simply not a service in kubernetes as it’s not required. The restoration script simply tries to start the Zammad service.

Shouldn’t be a real issue, if you can start your containers, you should be fine.

Sorry I didn’t make that clear in the first post: the System isn’t restored at all :sweat_smile:.

Starting with an unconfigured system(right after Install - the getting started page) and the “Hello World”-Setup: no restore goes through, both Systems remain untouched, despite the Script finishing with “restored successfully”.

Is it a mistake to start with all containers running?
Do I need to stop all or specific containers?
I start the script in this state:

Update: The restore skript doesn’t reach the database, because it can’t handle a connection over 2 containers. It is in the main container and needs to be executed there (at least it seems like that), while the database is in a seperate container.