Hello.
I read this page about 2 nodes HA cluster.
As I understood, I should manually install and configure DRBD if I haven't SAN to use 2 nodes cluster failover and some configure of failover domain.
And how work failover if cluster has 3 nodes and more, it also will need SAN or other or it can work itself?
I change cluster.conf and now it' look like this:
but after I reboot first node (on it was created VM<100>) it can't be started! And I have rgmanager: goup joined failed -1 -1 error. If it's important that I also change /etc/default/grub by modify line that: GRUB_CMDLINE_LINUX_DEFAULT="quiet iommu=pt iommu=1" to use pci pass throw, because I want to use usb pass throw but command qm set 100 -usb0 host=0ea0:2168 has no effect on guest system (yes, it modify 100.conf file in qemu-server folder, but guest system WinXP didn't see usb anyway).
And what about failover migration if I will use usb pass throw, I read it's impossible...
I read this page about 2 nodes HA cluster.
As I understood, I should manually install and configure DRBD if I haven't SAN to use 2 nodes cluster failover and some configure of failover domain.
And how work failover if cluster has 3 nodes and more, it also will need SAN or other or it can work itself?
I change cluster.conf and now it' look like this:
Code:
<?xml version="1.0"?>
<cluster config_version="12" name="test-cluster">
<cman two-node="1" expected_votes="1" keyfile="/var/lib/pve-cluster/corosync.authkey"/>
<clusternodes>
<clusternode name="proxmox-d" nodeid="1" votes="1"/>
<clusternode name="proxmox-vb" nodeid="2" votes="1"/>
</clusternodes>
<rm>
<pvevm autostart="1" vmid="100"/>
</rm>
</cluster>And what about failover migration if I will use usb pass throw, I read it's impossible...