pvestatd: status update time (409 seconds)

[Afox]

Hello,

I have a CIFS share mounted from within Proxmox and lately there are some real issues with the webinterface and Backups to that share.

I suspect that the only related Log message is the following:

mkdir /mnt/pve/CIFS-share: File exists at /usr/share/perl5/PVE/Storage/DirPlugin.pm line 136

But I´m not sure though.

The symptoms are that all VMs and Storages get a questionmark symbol at the same time. After a few minutes the symbols normalize. But when for example a Backup-cron is running in this time the job gets stuck. It seems to be a cycle because that symptoms repeat.

For any help I am grateful.

Regards,

Afox

 

[Afox]

A Backup-job gets canceled like so:

pvedaemon: could not activate storage 'CIFS-share': mkdir /mnt/pve/CIFS-share: File exists at /usr/share/perl5/PVE/Storage/DirPlugin.pm line 137

 

[Afox]

Update: The CIFS storage was indeed not reachable/broken. But IMO it is a bug if that breaks the proxmox webinterface.

 

[Afox]

Reason for this crash is that the CIFS storage runs out of memory and gets killed:

kernel: Out of memory: Killed process (smbd)
kernel: oom_reaper: reaped process (smbd)