PVE 4.3 two node cluster does not start after reboot

[gkovacs]

After updating a two node cluster to 4.3, I have rebooted the nodes one by one (not at the same time). After reboot none of the VM's were running, trying to start them on any node gave a cluster error:

root@proxmox2:~# qm start 111
cluster not ready - no quorum?

Checking the cluster showed it's not running at all:

root@proxmox2:~# pvecm nodes

Membership information
----------------------
    Nodeid      Votes Name
         2          1 proxmox2 (local)

root@proxmox2:~# pvecm status
Quorum information
------------------
Date:             Fri Sep 30 02:33:52 2016
Quorum provider:  corosync_votequorum
Nodes:            1
Node ID:          0x00000002
Ring ID:          2/1812
Quorate:          No

Votequorum information
----------------------
Expected votes:   2
Highest expected: 2
Total votes:      1
Quorum:           2 Activity blocked
Flags:

Membership information
----------------------
    Nodeid      Votes Name
0x00000002          1 10.10.10.2 (local)

I had to manually start the Proxmox services on both nodes to have a functioning cluster and reach quorum:

root@proxmox2:~# /etc/init.d/pve-cluster start
[ ok ] Starting pve-cluster (via systemctl): pve-cluster.service.
root@proxmox2:~# /etc/init.d/pvedaemon start
[ ok ] Starting pvedaemon (via systemctl): pvedaemon.service.
root@proxmox2:~# /etc/init.d/pve-firewall start
[ ok ] Starting pve-firewall (via systemctl): pve-firewall.service.
root@proxmox2:~# /etc/init.d/pve-manager start
[ ok ] Starting pve-manager (via systemctl): pve-manager.service.
root@proxmox2:~# /etc/init.d/pveproxy start
[ ok ] Starting pveproxy (via systemctl): pveproxy.service.
root@proxmox2:~# /etc/init.d/pvestatd start
[ ok ] Starting pvestatd (via systemctl): pvestatd.service.

root@proxmox2:~# pvecm nodes

Membership information
----------------------
  Nodeid  Votes Name
  1  1 proxmox1
  2  1 proxmox2 (local)

root@proxmox2:~# pvecm status
Quorum information
------------------
Date:  Fri Sep 30 02:42:20 2016
Quorum provider:  corosync_votequorum
Nodes:  2
Node ID:  0x00000002
Ring ID:  1/3036
Quorate:  Yes

Votequorum information
----------------------
Expected votes:  2
Highest expected: 2
Total votes:  2
Quorum:  2
Flags:  Quorate

Membership information
----------------------
  Nodeid  Votes Name
0x00000001  1 10.10.10.1
0x00000002  1 10.10.10.2 (local)

Previous Proxmox versions were able to reboot, reconnect to the cluster and start VM's without any manual intervention.