Hi
We have been doing some testing on HA with PVE and Ceph, which has given very possitive results in terms of our goals with exploring the setup.
From our testing, we have found two issues that we had not really expected
1. When a host has failed and comes back online, VMs are not migrated back. We asumed that they would be migrated back, but we are actually happy that this is not the case as it would not be a good idea to migrate services on to a unhealthy node
2. Resources have to be manually configured for HA. Is there a way to, like with backups, to automatically have all resources be under HA, so that we do not have to remember adding a new VM/Container to HA each time we add a new resource? We would ultimately like to have an HA setup, where all resources are under HA, unless we configure an exception for it.
We have been doing some testing on HA with PVE and Ceph, which has given very possitive results in terms of our goals with exploring the setup.
From our testing, we have found two issues that we had not really expected
1. When a host has failed and comes back online, VMs are not migrated back. We asumed that they would be migrated back, but we are actually happy that this is not the case as it would not be a good idea to migrate services on to a unhealthy node
2. Resources have to be manually configured for HA. Is there a way to, like with backups, to automatically have all resources be under HA, so that we do not have to remember adding a new VM/Container to HA each time we add a new resource? We would ultimately like to have an HA setup, where all resources are under HA, unless we configure an exception for it.