Hi! I could not quickly find a guide or best practice for the FC storage. I have a blade system with 5 blades connected to a shared FC storage. What is the best way to organize a shared storage for HA if I need snapshots? I read that CLVM is an option, but you can't make snapshots there. CEPF?
FC bestpractice
- Thread starter Donner
- Start date
The reason you cant find a guide or best practice for FC storage is that its always vendor specific. The best practices for your SAN should always come from the manufacture of the device.
Your best route is to find vendor's document on connecting storage to Linux , or if they have it - to Ubuntu.
Once you have your storage connected and disks are visible, you will have to decide how to present these disks to the application - PVE/Proxmox.
Since you have existing legacy storage and snapshot are a requirement you only have one option:
Use any of the available _clustered_ file systems to present a common file system across your nodes. Again, you wont find guides on how to set it up in Proxmox documentation - its out of the scope. You can start here for a common list: https://en.wikipedia.org/wiki/Clustered_file_system.
For snapshots you can either go with QCOW storage and use snapshots in conjunction with Proxmox, or you can utilize the native snapshot functionality of your chosen Clustered File system. However the latter will be out of band of Proxmox and will need to be managed manually.
Using CLVM is not going to buy you snapshot support in Proxmox:
https://access.redhat.com/documenta...olume_manager_administration/snapshot_volumes
If you were building a new environment and open to modern solutions, I'd recommend taking a look at Blockbridge - we support shared storage across multiple clusters. Snapshots, clones, HA are all natively supported - https://kb.blockbridge.com/guide/proxmox
Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
Your best route is to find vendor's document on connecting storage to Linux , or if they have it - to Ubuntu.
Once you have your storage connected and disks are visible, you will have to decide how to present these disks to the application - PVE/Proxmox.
Since you have existing legacy storage and snapshot are a requirement you only have one option:
Use any of the available _clustered_ file systems to present a common file system across your nodes. Again, you wont find guides on how to set it up in Proxmox documentation - its out of the scope. You can start here for a common list: https://en.wikipedia.org/wiki/Clustered_file_system.
For snapshots you can either go with QCOW storage and use snapshots in conjunction with Proxmox, or you can utilize the native snapshot functionality of your chosen Clustered File system. However the latter will be out of band of Proxmox and will need to be managed manually.
Using CLVM is not going to buy you snapshot support in Proxmox:
https://access.redhat.com/documenta...olume_manager_administration/snapshot_volumes
Code:
Note
LVM snapshots are not supported across the nodes in a cluster. You cannot create a snapshot volume in a clustered volume group.If you were building a new environment and open to modern solutions, I'd recommend taking a look at Blockbridge - we support shared storage across multiple clusters. Snapshots, clones, HA are all natively supported - https://kb.blockbridge.com/guide/proxmox
Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
Ceph is not recommended in your case, its not meant to run on shared storage that is backed by RAID. Ceph was designed to work on local disks of independent nodes
Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox