help me proxmox (ERROR CLUSTER)

[juliano caserta]

People need a help, what happens I have a proxmox server in production and I created a secondary.

I went into the shell and typed pvecm create sunway "cluster name", it created everything normal, normal status, but now I can not back up the vms it error, I can not create vm it gives the following error, I will put everything Please, please help me, because if you put a stick on the server, I can not backup these vms in production. I need your help if possible, THANK YOU.

PS: IM NOT SPEAK ENGLISH... LOL..

 

[dietmar]

You created a cluster with two nodes, but one node is offline (vm2). So your cluster is not quorate, and write activity is blocked.

If there is only one node (vm), you can set expected votes manually there:

# pvecm extected 1

Then remove the second node (vm2) ...

 

[juliano caserta]

THANKS MANNNNNNNN. YOU ARE THE BESTTTTT

 

[juliano caserta]

Dietmar .. I did not quite understand how to do the cluster process because when I had done that I gave this error I did this procedure.

200.7.120.253 - primary server (master)
I put the command

Pvecm create sunway (create cluster)

Pvecm Status
Quorum information
------------------
Date Tue Mar 14 13:44:25 2017
Quorum provider: corosync_votequorum
Nodes: 1
Node ID: 0x00000001
Ring ID: 1/8
Quorate: Yes

Votequorum information
----------------------
Expected votes: 1
Highest expected: 1
Total votes: 1
Quorum: 1
Flags: Quorate

Membership information
----------------------
Nodeid Votes Name
0x00000001 1 200.7.120.253 (local)


Now in secondary
200.7.120.66 - secondary server (client)

What do I do to connect this secondary server to the master?

Does anything need to be done?

 

[dietmar]

see https://pve.proxmox.com/wiki/Cluster_Manager

 

[juliano caserta]

see https://pve.proxmox.com/wiki/Cluster_Manager

Thanks, but my question is, would it have to change something on these two screens?
Other than that if you have everything correct in these images just send it to the master, right?

Being the master 200.7.120.253
And the secondary 200.7.120.68

 

[dietmar]

Sorry, I don't understand your question. There is no master/secondary relation ..

 

[juliano caserta]

Sorry, I don't understand your question. There is no master/secondary relation ..

OK.

Are Settings for Clustering Correct?

Being the "vm" the master connection of the cluster, and vm3 which will cluster in the "vm".

 

[udo]

OK.

Are Settings for Clustering Correct?

Being the "vm" the master connection of the cluster, and vm3 which will cluster in the "vm".

Hi,
vm3 must be empty! all VM-configs are gone during clustering - the data from "vm" will overwrite on vm3.
But do you think it's an good idea to run the cluster on the official IPs? directly in the internet?? I would not recommend this!

Anyway - before you start you should take a backup of /etc/pve (for savety) on vm first (like "tar cvf /root/etc_pve.tar /etc/pve" ).
Join from vm3 the excisting cluster vm (and test multicast-ping before).

Udo

 

[jucaserta]

Hi,
vm3 must be empty! all VM-configs are gone during clustering - the data from "vm" will overwrite on vm3.
But do you think it's an good idea to run the cluster on the official IPs? directly in the internet?? I would not recommend this!

Anyway - before you start you should take a backup of /etc/pve (for savety) on vm first (like "tar cvf /root/etc_pve.tar /etc/pve" ).
Join from vm3 the excisting cluster vm (and test multicast-ping before).

Udo

Udo, thank you for your attention.

So what happens, can I do it this way?

Being
"VM" MASTER CLUSTER
"VM3" CLIENT CLUSTER

Could I call them via vmbr1?

If yes what is the best procedure, since the "vm" is in production.

Ps: need to change something in /etc/hosts

 

[udo]

Could I call them via vmbr1?

Hi,
yes - an internal network is the right choice.

If yes what is the best procedure, since the "vm" is in production.

Ps: need to change something in /etc/hosts

/etc/host must changed - pvelocalhost (and vm) must set to the private network address.

See also here:
https://pve.proxmox.com/wiki/Proxmox_VE_4.x_Cluster
https://pve.proxmox.com/wiki/Separate_Cluster_Network

Udo

 

[juliano caserta]

Hi,
yes - an internal network is the right choice.

/etc/host must changed - pvelocalhost (and vm) must set to the private network address.

See also here:
https://pve.proxmox.com/wiki/Proxmox_VE_4.x_Cluster
https://pve.proxmox.com/wiki/Separate_Cluster_Network

Udo

All, I was reading and uploaded a question, I created a cluster as it is currently in / etc / hosts and the ips, I would need to remove this cluster and create a new one or not ??

Sorry for the questions, but I'm a little proxmox and I wanted to understand how it works so as not to make a crap on the production server.

Changing the ip in / etc / hosts would I have to restart proxmox?

What's the difference between these two links?

 

[udo]

All, I was reading and uploaded a question, I created a cluster as it is currently in / etc / hosts and the ips, I would need to remove this cluster and create a new one or not ??

Hi,
you can remove the cluster (stop corosync), change /etc/hosts and create an cluster again. Or you change /etc/hosts and edit /etc/corosync/corosync.conf from host vm and use instead of names the IP and restart corosync - then your cluster use the right IPs. But this is not really recommended for beginners...

Sorry for the questions, but I'm a little proxmox and I wanted to understand how it works so as not to make a crap on the production server.

Changing the ip in / etc / hosts would I have to restart proxmox?

no

What's the difference between these two links?

the content?!

perhaps you should find an buddy in your town who know proxmox/corosync?

Udo