Forcing migration in HA

[spitf1r3]

I have 3 Proxmox hosts: 2 rather powerful machines (s01proxmox02, s01proxmox03) with more RAM, and a low power one (s01proxmox01), on which I virtualize a router and a few containers, all of them are in one PVE datacenter. The smaller one only has 1 data disk, while the bigger guys have 2 or 3.
Most of VMs are being replicated between the big ones, while some smaller ones that would be able to fall back to s01proxmox01.

I was playing around with HA so I could automate migration between them and somehow 2 of my VMs got migrated to s01proxmox01, while their storage remained on the big ones (02 and 03). Now when I want to move them back on migration they display an error "storage 'local-zfs_2' is not available on node 's01proxmox01' (500)" (which is true). How do I move them back

 

[fiona]

Hi,
you should be able to recover by moving the configuration files manually (from /etc/pve/nodes/<nodeA>/qemu-server/<ID>.conf to the other node /etc/pve/nodes/<nodeB>/qemu-server/, for containers the directory is lxc instead of qemu-server).

If you want HA to not migrate guests to a certain node, you can use an HA group with the restricted flag: https://pve.proxmox.com/pve-docs/chapter-ha-manager.html#ha_manager_groups

 

[spitf1r3]

Hi,
you should be able to recover by moving the configuration files manually (from /etc/pve/nodes/<nodeA>/qemu-server/<ID>.conf to the other node /etc/pve/nodes/<nodeB>/qemu-server/, for containers the directory is lxc instead of qemu-server).

If you want HA to not migrate guests to a certain node, you can use an HA group with the restricted flag: https://pve.proxmox.com/pve-docs/chapter-ha-manager.html#ha_manager_groups

Thank you!

I've already created several HA groups and assigned them to VMs as needed, but that migration happened before I did.