Hello everybody,
I'm newbie to Proxmox.
I only have 2 servers to mount my Proxmox platform (i do not need the HA).
I installed the Proxmox distribution (Virtual Environment 4.4-5 / c43015a5) on these 2 servers and I use DRBD as storage space. I managed to create some VMs (LXC, KVM) and run them. I found, however, the following:
1) When I stop one of the nodes and I restart it, all the machines of it does not start !
if I try to migrate the machines, from the interface, from the primary node (node1) to the secondary (node2) it does not start! DRBD indicates:
<code>
root@node1:~# drbd-overview
102: vm-102-disk-1/0 Connected (2 *) Secondary (2 *) UpToDa / UpToDa
</code>
... the disk does not appear on the secondary node (node2) in "/dev/drbd/by-res /" and that's why it does not start !
Also, if Indeed, I was forced to do this with DRBD to restart the machines:
For example for vm-102:
root@node2:~# drbdadm disconnect vm-102-disk-1
root@node2:~# drbdadm secondary vm-102-disk-1
root@node2:~# drbdadm connect --discard-my-data vm-102-disk-1
</code>
root@node1:~# drbdadm disconnect vm-102-disk-1
root@node1:~# drbdadm connect vm-102-disk-1
root@node1:~# /etc/init.d/drbd restart
[ ok ] Restarting drbd (via systemctl): drbd.service.
root@node1:~#
</code>
I saw that on the site of Linbit (DRBD) it was necessary to install the plugin "drbdmanage-proxmox".
I had not done this before because I used "drbdmanage", available from Debian repositories.
What does this plugin add?
Is that why I have these problems mentioned ?
2) If I want to use the Ceph storage space, does it work with only 2 servers ?
With my thanks and best regards
I'm newbie to Proxmox.
I only have 2 servers to mount my Proxmox platform (i do not need the HA).
I installed the Proxmox distribution (Virtual Environment 4.4-5 / c43015a5) on these 2 servers and I use DRBD as storage space. I managed to create some VMs (LXC, KVM) and run them. I found, however, the following:
1) When I stop one of the nodes and I restart it, all the machines of it does not start !
if I try to migrate the machines, from the interface, from the primary node (node1) to the secondary (node2) it does not start! DRBD indicates:
<code>
root@node1:~# drbd-overview
102: vm-102-disk-1/0 Connected (2 *) Secondary (2 *) UpToDa / UpToDa
</code>
... the disk does not appear on the secondary node (node2) in "/dev/drbd/by-res /" and that's why it does not start !
Also, if Indeed, I was forced to do this with DRBD to restart the machines:
For example for vm-102:
- On the secondary node (where the machines were not active):
root@node2:~# drbdadm disconnect vm-102-disk-1
root@node2:~# drbdadm secondary vm-102-disk-1
root@node2:~# drbdadm connect --discard-my-data vm-102-disk-1
</code>
- On the primary node (where the machines were operating):
root@node1:~# drbdadm disconnect vm-102-disk-1
root@node1:~# drbdadm connect vm-102-disk-1
root@node1:~# /etc/init.d/drbd restart
[ ok ] Restarting drbd (via systemctl): drbd.service.
root@node1:~#
</code>
I saw that on the site of Linbit (DRBD) it was necessary to install the plugin "drbdmanage-proxmox".
I had not done this before because I used "drbdmanage", available from Debian repositories.
What does this plugin add?
Is that why I have these problems mentioned ?
2) If I want to use the Ceph storage space, does it work with only 2 servers ?
With my thanks and best regards