Ceph » CephFS

Milind Changire wrote:

The most common mistake that users tend to do is include the mount point path along with the file-system path when using snap-schedule commands.

Doesn't our documentation mention this clearly? If not, we should fix the doc.

This leads to the snap-schedule module encountering invalid paths as far as the file-system is concerned.

This was discussed as part of design with Jan - users can add schedules to non-existent file system paths, which may be created at a later point in time. Additionally, if the number of consecutive failures hit a certain configured value, the schedule would be deactivated for the path. Are you not seeing that?

Venky Shankar wrote:

Milind Changire wrote:

The most common mistake that users tend to do is include the mount point path along with the file-system path when using snap-schedule commands.

Doesn't our documentation mention this clearly? If not, we should fix the doc.

I'll look at the docs and put in a note there.

This leads to the snap-schedule module encountering invalid paths as far as the file-system is concerned.

This was discussed as part of design with Jan - users can add schedules to non-existent file system paths, which may be created at a later point in time. Additionally, if the number of consecutive failures hit a certain configured value, the schedule would be deactivated for the path. Are you not seeing that?

The schedule deactivation part is there, but there's no track kept of number of consecutive failures. The moment an exception get raised during snapshot creation, due to file-system or db updates, the schedule is set to inactive.