Proxmox VE 4 supports groups of up to 32 physical nodes. La gestión centralizada Proxmox hace que sea fácil de configurar todos los nodos disponibles desde un solo lugar. There are many advantages if using a Proxmox cluster e.g. es fácil migrar una máquina virtual de un nodo a otro nodo. You can use 2 Proxmox servers for a multi-node configuration, but if you want to configure Proxmox for high availability, you need 3 or more Proxmox nodes.
In this tutorial, we are going to set up a Proxmox 4 cluster with 3 Proxmox servers and 1 NFS storage server.
Network scheme:
Requirements:
proxmox (Master)
IP : 192.168.1.230
FQDN : proxmox.redorbita.com
proxmox01
IP : 192.168.1.231
FQDN : proxmox01.redorbita.com
proxmox01
IP : 192.168.1.232
FQDN : proxmox02.redorbita.com
openfiler (Storage)
IP : 192.168.1.220
FQDN : openfilter.redorbita.comCreate volume
Primero debemos crear un volumen y exportarlo mediante NFS en el servidor de almacenamiento, yo en mi caso voy a utilizar openfilter, tu puedes realizarlo mediante otro tipo de cabinas de almacenamiento (freenas, openmediavault, netapp..etc) o mediante un servidor NFS
Crear volumen y exportarlo mediante NFS en openfilter
Configure hosts file
We must configure the hosts file of the three nodes as follows:
cat /etc/hosts
192.168.1.231 proxmox01.redorbita.com proxmox01 pvelocalhost
192.168.1.230 proxmox.redorbita.com proxmox pvelocalhost
192.168.1.232 proxmox02.redorbita.com proxmox02 pvelocalhostTo create a cluster in proxmox we must predefine one of the nodes as master, in our case we have chosen the server: proxmox.redorbita.com – 192.168.1.230
root@proxmox:~# pvecm create proxmoxclusterWe access the other two servers and execute the following command:
PROXMOX01
root@proxmox01:~# pvecm add 192.168.1.230
The authenticity of host ‘192.168.1.230 (192.168.1.230)’ can’t be established.
ECDSA key fingerprint is 72:ed:38:4e:5f:35:32:c4:a7:5f:5e:5c:7f:49:1f:11.
Are you sure you want to continue connecting (yes/no)? yes
root@192.168.1.230’s password:
copy corosync auth key
stopping pve-cluster service
backup old database
waiting for quorum…OK
generating node certificates
merge known_hosts file
restart services
successfully added node ‘proxmox02’ to cluster.PROXMOX02
root@proxmox02:~# pvecm add 192.168.1.230
The authenticity of host ‘192.168.1.230 (192.168.1.230)’ can’t be established.
ECDSA key fingerprint is 72:ed:38:4e:5f:35:32:c4:a7:5f:5e:5c:7f:49:1f:11.
Are you sure you want to continue connecting (yes/no)? yes
root@192.168.1.230’s password:
copy corosync auth key
stopping pve-cluster service
backup old database
waiting for quorum…OK
generating node certificates
merge known_hosts file
restart services
successfully added node ‘proxmox02’ to cluster.We check if we have the nodes correctly joined to the cluster:
root@proxmox:~# pvecm status
Quorum information
——————
Date: Tue Mar 29 11:37:53 2016
Quorum provider: corosync_votequorum
Nodes: 3
Node ID: 0x00000001
Ring ID: 32
Quorate: Yes
Votequorum information
———————-
Expected votes: 3
Highest expected: 3
Total votes: 3
Quorum: 2
Flags: Quorate
Membership information
———————-
Nodeid Votes Name
0x00000001 1 192.168.1.230 (local)
0x00000003 1 192.168.1.231
0x00000002 1 192.168.1.232Once joined, we access the graphical interface through our web browser and check that our nodes are joined.
NFS Configuration
To configure the shared service we must go to Storage > Add > NFS
ID: Name of the share in our PROXMOX cluster
Server: NFS server (openfiler)
Export: Shared resource to export
Content: What are we going to use this resource for?
Nodes: the nodes that we are going to include.
Once configured, we check that the file system is correctly mounted on all nodes.
root@proxmox01:~# df -h
Filesystem Size Used Avail Use% Mounted on
udev 10M 0 10M 0% /dev
tmpfs 401M 5.7M 395M 2% /run
/dev/dm-0 3.9G 1.1G 2.6G 31% /
tmpfs 1001M 60M 941M 6% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
tmpfs 1001M 0 1001M 0% /sys/fs/cgroup
/dev/mapper/pve-data 8.4G 20M 8.4G 1% /var/lib/vz
tmpfs 100K 0 100K 0% /run/lxcfs/controllers
cgmfs 100K 0 100K 0% /run/cgmanager/fs
/dev/fuse 30M 24K 30M 1% /etc/pve
192.168.1.220:/mnt/vol_cluster_proxmox/vol_cluster_proxmox/vol_cluster_proxmox/ 108G 33M 108G 1% /mnt/pve/nfsproxmoxEnable High Availbility to a VM and a CT
first we must add the VM or CT
Syntax:
ha-manager add vm:<VM ID/CT>
root@proxmox:/etc/pve# ha-manager add vm:100
root@proxmox:/etc/pve# ha-manager add vm:101Once added we must enable it
Syntax:
ha-manager enable vm:<VM ID/CT>
root@proxmox:/etc/pve# ha-manager enable vm:100
root@proxmox:/etc/pve# ha-manager enable vm:101If we access through our browser and select the server we can see in the Summary tab if it is enabled correctly.
High-availability testing
We are going to proceed to turn off the proxmox01 node and check if it detects its failure and proceed to migrate VM 100 (rredorbita01)
If everything goes correctly, as you see in the log, it should detect the node down and migrate to another node.
All the best.
:wq!
Comments