Move Data Directory to a CIFS mount end in a lockup

[Kubernetes]

Hi there,

I connected a CIFS Mount (Hetzner Storagebox) to my Cloudron for exclusive use as Data Directory for some Apps.
I have moved some Apps successfully to that new storage.

When I tried to move NextCloud Data Directory it did copy a lot of data (not sure if it is ALL data) to the new location. That took some time (5GB).
Then the App status went to "Not Responding".

When I look in the NextCloud App Logs on Cloudron I see something like this:

Jun 02 14:44:31 files_sharing settings
Jun 02 14:44:31 ==> Copying htaccess
Jun 02 14:44:31 ln: failed to create symbolic link '/app/data/core': Operation not supported
Jun 02 14:44:32 ==> Updating apps
Jun 02 14:44:32 ==> Old apps:
Jun 02 14:44:32 activity files_sharing settings
Jun 02 14:44:32 admin_audit files_trashbin sharebymail
Jun 02 14:44:32 bruteforcesettings files_versions support
Jun 02 14:44:32 circles firstrunwizard survey_client
Jun 02 14:44:32 cloud_federation_api logreader suspicious_login
Jun 02 14:44:32 comments lookup_server_connector systemtags
Jun 02 14:44:32 contactsinteraction nextcloud_announcements text
Jun 02 14:44:32 dashboard notifications theming
Jun 02 14:44:32 dav oauth2 twofactor_backupcodes
Jun 02 14:44:32 encryption password_policy twofactor_totp
Jun 02 14:44:32 federatedfilesharing photos updatenotification
Jun 02 14:44:32 federation privacy user_ldap
Jun 02 14:44:32 files provisioning_api user_status
Jun 02 14:44:32 files_external recommendations viewer
Jun 02 14:44:32 files_pdfviewer related_resources weather_status
Jun 02 14:44:32 files_rightclick serverinfo workflowengine
Jun 02 14:44:32 activity files_trashbin sharebymail
...
Jun 02 14:45:20 => Healtheck error: Error: connect ECONNREFUSED 172.18.19.196:80
Jun 02 14:45:23 ==> Update app: files_pdfviewer
Jun 02 14:45:30 => Healtheck error: Error: connect ECONNREFUSED 172.18.19.196:80
Jun 02 14:45:33 ==> Update app: files_rightclick
Jun 02 14:45:35 ==> Update app: files_sharing
Jun 02 14:45:39 ==> Update app: files_trashbin
Jun 02 14:45:40 => Healtheck error: Error: connect ECONNREFUSED 172.18.19.196:80
Jun 02 14:45:42 ==> Update app: files_versions
Jun 02 14:45:45 ==> Update app: firstrunwizard
Jun 02 14:45:48 ==> Update app: logreader
Jun 02 14:45:50 => Healtheck error: Error: connect ECONNREFUSED 172.18.19.196:80
Jun 02 14:45:51 ==> Update app: lookup_server_connector
Jun 02 14:45:52 ==> Update app: nextcloud_announcements
Jun 02 14:45:54 ==> Update app: notifications

And this is repeating and repeating all the time now. I think the problem could be the failed symbolic link?

I didn't do anything else except looking in the logs, as I don't want to break NextCloud completely.

What is you advise? How to get it up running again and using the CIFS Data Directory?

Thanks for any quick help here...

Best,
Michael

[Kubernetes]

Uhm, I just read here that CIFS is not suitable for Data Directories...

@girish maybe Cloudron should not allow to move App Data to a CIFS share, if it is not suitable?

[Kubernetes]

I was able to move the data back again with the App Configuration in Cloudron and everything seems to be fine again. No help needed anymore, but I still think it would be a good idea to prevent user from doing this.

[girish]

@Kubernetes Yes, true. will make an issue for this. CIFS doesn't support symlinks and proper permissions. So, it is not well suited for data directory of apps.

[girish]

https://git.cloudron.io/cloudron/box/-/issues/827