Redis crashes when adding/removing volumes in a stopped app
I have a server with 2 Nextcloud instances that use Hetzner Storageboxes as storage backend and until today, the directory was mounted manually via
/etc/fstab. Since every restart ended up with a lost race condition which made me manually remove empty directories and remount the storagebox, I wanted to use the built-in SSHFS Volumes to get this done.
So I added the server, stopped Nextcloud, changed the path to the data dir in the
config.phpand wanted to add the newly created Volumes. While reconfiguring it threw the message
An error occurred during the create container operation: Inactive: Error getting IP of redis-xxx-id serviceand went into an erronous state. I had to go to Services / Redis and restart the corresponding redis service and hit retry in the Repair tab - after that the Volume was added successfully. I could easily reproduce the issue with the second Nextcloud instance, so either adding or removing Volumes in a stopped app made redis crash, maybe something to check out for consistency.
Apparently I still have to manually hit the remount Volume button after a restart to get it online, but that beats sshing into the box and rm empty directories
Ah, thanks for reporting. I have opened https://git.cloudron.io/cloudron/box/-/issues/808 internally .