Instances Enter a Paused State on the Hypervisor

Problem

Instances running on a hypervisor go into a paused state. The following error is observed in the /var/log/pf9/ostackhost.log.

Javascript

2021-08-31 09:40:12.849 ERROR nova.compute.manager [...] [instance: c67fc199-c897-4b42-8305-8a90f2473a23] Cannot reboot instance: Failed to read or write disk info file: [Errno 28] No space left on device: '/opt/pf9/data/instances/c67fc199-c897-4b42-8305-8a90f2473a23/disk.info.tmp': nova.exception.DiskInfoReadWriteFail: Failed to read or write disk info file: [Errno 28] No space left on device: '/opt/pf9/data/instances/c67fc199-c897-4b42-8305-8a90f2473a23/disk.info.tmp'

Environment

• Platform9 Managed OpenStack - All Versions

• Nova

Cause

The instances entered a paused state because there is no available disk space left on the instances mountpoint. When this happens, the instance state is reported as ACTIVE in the UI but the instances are in a paused state on the hypervisor.

Resolution

1. Provision additional space to the instances mountpoint on the hypervisor.

2. Run the following commands for each instance present on the affected hypervisor to bring the instances back into a running state.

Javascript

# openstack server pause <server_uuid>
# openstack server unpause <server_uuid>