[pve-devel] Could we please get a solution for HA and backup locked VMs?

Thomas Lamprecht t.lamprecht at proxmox.com
Mon Jul 24 17:51:59 CEST 2017


Hi,

On 07/24/2017 05:32 PM, Harald Leithner wrote:
> Hello,
> 
> I know that there was already one or more conversations about locked VMs 
> that can't be started in case of a node reboot/crash/failure.
> 
> But it's a pain in the ass if you have a HA cluster and your HA is 
> broken because you make backups.
> 
> My backups run several hours, maybe 6 hours or more. So 25% of the day I 
> don't have a automatically failover for at least one VM. And this to a 
> time frame where HA is most important ;-)
> 
> Is there anything I can do to change this for me, or how I can motivate 
> to change the mind of the proxmox team?
> 

What PVE version are you running `pveversion -v`?

This should be already implemented and working in PVE 4.3 and newer, 
see: 
<https://git.proxmox.com/?p=pve-ha-manager.git;a=commit;h=5dd3ed863a162bf78e59c5dc550f2f708b3cbfe2>

The crash/failure aspect should be covered. Gracefully rebooting during 
running backups sounds strange.

cheers,
Thomas




More information about the pve-devel mailing list