November 2017
Intermediate to advanced
494 pages
14h 29m
English
This is usually caused by one of the three services, such as pvedaemon, pvestatd, or pveproxy crashing or stopping working for any number of reasons. Simply restarting them through SSH will fix this issue. One of the common causes of this issue is if any NFS shared storage gets stuck during an extended backup task. A node reboot will always fix this issue. But reboot is not always possible in a production node. Forcefully unmounting the NFS shared storage under /mnt/pve/<share>, then running the following commands will show everything normally again:
# service pvedaemon restart # service pveproxy restart # service pvestatd restart
Read now
Unlock full access