Invalid mount path. Following service(s) uses this path: `Snapshot Task

[rubick]

Hi!
I have an app with persistent storage of Host Path type that was working fine.
I enabled a periodic snapshot task for the parent dataset that was host path mounted and when I stopped and started the application I got the following error:

Code:

Error: Error response from daemon: invalid volume specification: '/mnt/ssdm2/configs/media/transmission:/config': Invalid mount path. /mnt/ssdm2/configs/media/transmission. Following service(s) uses this path: `Snapshot Task`.

As soon as I disable the periodic snapshot of ssdm2/configs, the application start without problems.

I don't understand why the host path became invalid for enabling periodic snapshots over that dataset and is preventing the pod from starting.
is it incompatible snapshots of a dataset and app storage of host path type?

 

[rubick]

This happens with official and truecharts

Code:

Error: Error response from daemon: invalid volume specification: '/mnt/ssdm2/configs/homeassistant:/config': Invalid mount path. /mnt/ssdm2/configs/homeassistant. Following service(s) uses this path: `Snapshot Task`.

 

[Capitol]

22.12-BETA.1. Same happens with docker images, but also with replication tasks. Un-enabling the task let the application start.

 

[rubick]

Yeah, that's what I did, but I'm not sure if it's a bug or if it's some kind of incompatibility between features. I think a scheduled snapshot should not make the dataset busy/locked

 

[mndsm]

I am running into this issue as well. This seems to be an issue with whatever storage provider Truenas uses with Kubernetes, because Kubernetes will not be aware of shared services like NFS/SMB and shapshotting.

This is a bug and needs to be addressed. Anybody know where we can log it?

Error: Error response from daemon: invalid volume specification: '<path>:<mntpath>': Invalid mount path. <path>. Following service(s) uses this path: `NFS Share`.

 

[mndsm]

The accommodating thread which outlines the problem and bug reports can be found at https://www.truenas.com/community/threads/docker-nfs-bug.103994/

 

[thepaintsaint]

In Applications, go to Settings > Advanced Settings, and uncheck Enable Host Path Safety Checks. Save, your containers will restart, and you won't have this conflict anymore.