Cluster Proxmox Migrate Failed

[Totof]

Hello everybody,
My english is not great, so I would use google translate
I'm having trouble with proxmox. I rent a dedicated server or proxmox is install, and I have a second server self host. I set up a cluster with these two machines, until all is well. I am going through a VPN in order to do it.

Node1 will be the dedicated server rent
Node2 will be the self-hosting server

When I try to migrate Node2 -> Node1 it's good
But when I try to migrate Node1 -> Node2 it's a failure.

Indeed when I look at the summary of the cluster, I see the Node 1 with the public ip, however for the node 2 I see it with its internal ip.

So necessarily when the node1 tries to connect with the internal ip of the node2 it does not work.

Is there a possibility to specify an FQDN for node2 in the cluster configuration?


I hope that google translate is doing its job well and that it will be quite understandable.

thank you

 

[karnz]

What its mean "self-hosting server"?

How you can join node2 with internal IP to the cluster? it should be on different subnet? doesn't get error while joining?
For FQDN, you must set it before adding node to the cluster because after the nodes are in the cluster, it's quite impossible to change an IP or node's name. (may have to leave the cluster then re-install the node and re-join it)

All nodes in the cluster should be updated to the same proxmox version to avoid any conflict.

 

[Totof]

By self-hosting I meant that it is a server where I installed proxmox and it is behind a box adsl.

To make the cluster, I first set up a vpn so that both servers can communicate in multicast

And there is no problem for the cluster as such.

root@pve1:~# pvecm status
Quorum information
------------------
Date:             Sun Jun 10 19:23:03 2018
Quorum provider:  corosync_votequorum
Nodes:            2
Node ID:          0x00000001
Ring ID:          1/216
Quorate:          Yes

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

Membership information
----------------------
    Nodeid      Votes Name
0x00000001          1 10.9.0.1 (local)
0x00000002          1 10.9.0.2
root@pve1:~# pvecm nodes

Membership information
----------------------
    Nodeid      Votes Name
         1          1 serveur1-vpn (local)
         2          1 10.9.0.2

to join the cluster, I used this command. Namely that the names server1-vpn and server2-vpn are in the hosts

pvecm add serveur1-vpn -ring0_addr 10.9.0.2

However, where I think this is a problem, is that when I go to Datacenter -> Summarize I see both node pve1 with his public ip and pve2 with his private ip.

On the node rented at ovh, the interface enp4s0 is the public ip while on node 2 the ip is 192.168.1.2

So when I start a migration of node 2 -> node 1 everything is OK (I guess it's because in the summary we can see node1 = Public IP)

But when I start a migration of node 1-> node 2 I have the error below

2018-06-10 19:30:49 # /usr/bin/ssh -e none -o 'BatchMode=yes' -o 'HostKeyAlias=pve2' root@192.168.1.2 /bin/true
2018-06-10 19:30:49 ssh: connect to host 192.168.1.2 port 22: No route to host
2018-06-10 19:30:49 ERROR: migration aborted (duration 00:00:03): Can't connect to destination address using public key
TASK ERROR: migration aborted

Migration Node2 --> Node 1 :

2018-06-10 21:32:36 starting migration of CT 101 to node 'pve1' (91.XXX.XXX.XXX)
2018-06-10 21:32:36 found local volume 'local:101/vm-101-disk-1.raw' (in current VM config)
Formatting '/var/lib/vz/images/101/vm-101-disk-1.raw', fmt=raw size=4294967296

attached, two screenshot of what I have


13/5000
Regards,