Description of problem: Official solution for remediation of issues where /var/lib/cinder/mnt goes to 100%. Unpause VMs in Paused State after NFS storage for Cinder hit 100% After expand the NFS Filesystem space we ran the below command on every compute node to resume the VMs from the libvirt layer #virsh list | grep -i pause | awk '{print $2}'|xargs -i virsh resume {} Looking for an official document to fix this using openstack commands as - nova reset-state [--all-tenants] [--active] <server> [<server>] - nova resume <server> etc... Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info:
I'm assuming here that this happened because the Cinder volumes aren't preallocated? Need to discuss with the team how we think Nova should act here.
The agreement was that Nova should notice that the VM is paused and move it to the paused state.