What happened if compute node machine restarts or stop working ??
Hello, I wonder what is happening when a machine that runs computation node with active VMs is shutdown due hardware malfunction or power outage and after some time restarting and returns back? Does OpenStack somehow manage "to move" VMs that were configured to that node to run on another node? What happened to networking between VMs on other nodes trying to reach VMs that were running on the shutdown node?
Responses
The instances on the down compute can be evacuated off the compute host. Per default this is NOT an automatic task. Evacuation rebuilds the instances with the same metadata (IP,...) on a different compute host. Depending if shared storage is configured (like ceph) the instances get started from the current disc on shared storage. If no shared disk is available the instance gets recreated from the glance image used by the old instance.
Automatic evacuation could be enabled using [1]
[1] https://access.redhat.com/documentation/en-us/red_hat_openstack_platform/10/html-single/high_availability_for_compute_instances/index
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
