Creating a cluster will not erase any VMs on the first node.
All other nodes should be emptied (Backups made to NAS).
PS: The Shared Storage for Container and DiskImage should both be on RAID10, the others on RAID5 is OK…
VZBackupDump and DiskImage / Container should NOT be on the same storage Volume / Disks!
I would recreate ContainerTemplate on RAID1, and remove it from RAID10. RAID10 should only be used for VMs…
I usually limit ContainerTemplate to 100GB, Containers to 300 or 500 GB, ISOimages to 300 GB…
DiskImage 2TB (or more), VZBackupDump to 2 or 3 TB…
Adapt as suitable / needed.
I always create the cluster from the console, when I started the GUI option was not available, and I find the console simpler:
————————————————————
Cluster
————————————————————
Create the cluster
Login via ssh to the first Proxmox VE node. Use a unique name for your cluster, this name cannot be changed later.
Create:
hp1# pvecm create YOUR-CLUSTER-NAME
pvecm create PVE-CLUST
To check the state of cluster:
hp1# pvecm status
Adding nodes to the Cluster
Login via ssh to the other Proxmox VE nodes. Please note, the nodes cannot hold any VM. (If yes you will get conflicts with identical VMID¥s - to workaround, use vzdump to backup and to restore to a different VMID after the cluster configuration).
WARNING: Adding a node to the cluster will delete it's current /etc/pve/storage.cfg. If you have VMs stored on the node, be prepared to add back your storage locations if necessary. Even though the storage locations disappear from the GUI, your data is still there.
Add the current node to the cluster:
hp2# pvecm add IP-ADDRESS-CLUSTER
For IP-ADDRESS-CLUSTER use an IP from an existing cluster node.
To check the state of cluster:
hp2# pvecm status
Aha! I was wondering about that nomenclature.
AW (your name)
“R7”… for a moment I tough it was for “Raspberry” or “Release”, but wrong!
I (almost) use magical creatures names for my servers; but the good ones are few; and the mythical names from México like Quetzalcoatl, they are very complex. But some would be very funny: Nagual, Pejelagarto, Xoloitzcuintle.
I had a client who insisted on using planets for his servers / pcs.
His MS SBS Server was called Pluto…
When they downgraded Pluto, his world fell apart…
His next one was called SOL, you can’t downgrade the Sun (SBS2008 / R2)…
Then MS killes SBS…
He now uses NethServer, and my naming conventions…
The normal limit for name length is the lowest common denominator, and that’s Windows, which allows max 15 chars… So some of your mexican names are close or over the limit!
Naming conventions become more important, as soon as several sites are involved. My way you never get duplicate names…
Many things had to be done without a previous plan, due to the need; and also the lack of experience did not allow me to plan years in advance, especially the range of IPs.
I plan to reinstall both proxmox servers. And using your naming convention gives more space to work and organize: AVO1-PVE[1,2] (my job place, office 1)
First the emergent one, added to the cluster (IP 192.168.16.21)
… checking.
There is no trace of the NFS share.
Yes, seems like the reboot can solve this.
I need to wait a few hours, maybe there is still staff working.
I have a PC that need to run some jobs at 6:00-6:30pm or it was 6:00-7:00pm… this is fun, I tend to forgot little things.
Move (first a test) some VMs to → DiskImage (right?)
I’ll start with the VM 301, it runs and I need to evaluate the sql performance, for my next project, migrate our payroll system to a new windows+sql server as VM in Proxmox.
If all works, the VM 100 is the next big step
Then move the other Proxmox (for tests) to the cluster and run both VMs just to see how the live migrations works now.
Edit
At this moment, I’m testing the migration of the VM 103, the VM 301, but can’t be moved because it have snapshots. I need to investigate how to move the VM including the snapshots, if this can’t be done I’ll delete the snapshots, to continue my testing for migrating sql server from one domain (from SBS2000 to NethServer)
So, I need to “bite the bullet” and migrate the VM 100, but first, I need to to a backup (on the NAS)
I wonder if the slow backup from the proxmox to the NAS is “normal”? and I hope that when the VM is in the NAS this process can be fast. I need to buy some “cache card” for this NAS.
The easiest and most painless way for a migration from local storage to shared storage is using Backups, then shutdown and erase. Restore from the NAS using the same VM number (or change nr if that fits better!)
You can massively improve the speeds once you have shared storage, simply by adding networks to Proxmox.
Best would be 4 Networks:
Public Network (The company’s normal LAN)
Cluster Network
Storage Network
Backup Network
You could combine the cluster network with the backup network, meaning you’ld need three networks.
These two illustrations should make this more understandable…
Your big NAS has 4 NICs, two connecting to your LAN, two connecting to a storage network, using a seperate switch or a vLAN segmant of the main switch.
I use the GUI from proxmox, delete old snapshots, take a recent backup up and then move the disk to the “DiskImage” share, and finally run the VM… and it works!
