Will stopping my login/controller instances created by a slurm blueprint potentially unlink shared filesystems?

[noahharrison64]

Hi,

I've encountered a strange issue after stopping my login and controller nodes (Trying to save some money when not in use by not having these running), when I SSH back into the login node I can no longer see the shared file system. The file system was a persistent disk that was attached to the controller node and shared with the login node.
The file system was located at
/home/<user_name>/
When I SSH into the controller node, I can see the directory/filesystem is still there. However, on the login node its no where to be seen.

Would stopping both nodes and then re-starting cause this to happen? Can any fixes be suggested?

Thanks

[noahharrison64]

Looks like transient loss of network connectivity was causing this issue. I had a Filestore instance mounted on /home/ and the login node was struggling to connect to it. It appears to be up and running correctly now.