PVE 5.x - Is HA possible with RRP?

[Romsch]

Hi all!

I tested rrp in our test environment with ceph. Two physical separate network switches, one for cluster network (the nodes 10.10.10.0) and one for ceph network (172.16.0.0).
I had the same experiences like the others described; if cluster network on one node is broken or offline, the vm still are running (thats good) but not available at the moment. Also the HA configured VMs are not changed to another host - is this possible now to configure? It would be very good, if HA configured VMs automatically move to a node wich is available in the cluster network.

With the ceph network if one ceph connection is broken or offline, the running VMs on this node are available, but in read only mode - is here also maybe a way, that the configured HA VMs automatically change to another host?

Best regards

 

[Binary Bandit]

Hi Romsch,

My understanding is that out of the box the network needs to be made redundant in hardware. I've read about other options like Pacemaker but the recommendation is to go with hardware.

Put another way, the cluster will protect you from server failure but now network failure.

Connect your Ceph network to two switches coming from two separate NICs in each server. Look into MLAG for the switches or if this is too pricey, the Debian Linux OS that Proxmox runs on supports active / backup links into standard switches.

Yes, VMs can be made to automatically move when there is a host server failure ... just configure a cluster HA group and put the VMs in it. Assuming that everything else (storage and network) is configured correctly the cluster will restart VMs on another host when the VM's current host fails. Check into hardware watchdogs to make this reliable.

best,

James