ZFS Pool Replication Across Cluster Nodes?

[pr0927]

Hi all, I previously asked this, but perhaps I couched it in too much detail and didn't get much of a response.

Also, this time after doing some digging I have some more focused questions that might help simplify the answer here.

Long story short:

1. 3 nodes in a cluster, using ZFS.
2. CTs and VMs are replicated across nodes via GUI.
3. I want to replicate data in ZFS pools (which my CTs and VMs use - CTs through bind-mounts, VMs through NFS shares) to the other nodes.
4. Currently using Sanoid/Syncoid to make this happen from one node to two others via cronjob.

So three questions:

1. If I do Sanoid/Syncoid on all three nodes (to each other) - is this stupid, and will it fail - or will each node recognize snapshots for a ZFS pool and incrementally update if needed (even if the snapshot was made on/by a different node)?
   1. As a sub-question to this - and kind of the point of my overall thread and the previous one - is this even a sensible way to approach this, or is there a better way?
2. For the GUI-based replication tasks, since I have CTs replicating to other systems, if I unchecked "skip replication" for the bind-mounted ZFS pools - would this accomplish the same thing? Or would it fail outright? I seem to remember needing to check this for some reason.
3. Is this PVE-zsync suitable for my situation? I see mention of no recursive syncing, which I don't fully know what that means, and I don't know if that's a dealbreaker. I suppose if this is the correct choice - then I need to delete my current GUI-based CT/VM replication tasks?

 

[pr0927]

Gentle bump for assistance!

 

[pr0927]

I've posted here and on Reddit and don't seem to get much of a bite - for multiple threads. Is there somewhere else I should be seeking assistance?

 

[lucius_the]

I am also interested in knowing if I can use pve-zsync together with storage replication... I opened another thread here for that specifically: https://forum.proxmox.com/threads/c...cation-and-pve-zsync-at-the-same-time.159194/

 

[pr0927]

Hmm I see no luck on your thread either. I've unfortunately had this experience here, on Reddit, and even on the LevelOneTechs forums. Really unsure where to turn for guidance...

 

[lucius_the]

Strange. Does nobody here has plans for a failover site ? Perhaps nobody uses Proxmox (with local ZFS pools, at least) for that.

Well, there are some options:
- we can try and see for ourselves (I currently don't have time for that)
- there's stuff like Veeam, that can replicate VMs across pretty much anything (it's expensive and I don't know how well it works here, since their support for PVE is rather newish)

I'm considering Veeam because I've used it before and it's a solid backup solution as well as replication solution. And is faster than PBS. PBS is great, but in some operations, like verify and restores, it's performance is inexplicably sluggish even on modern hardware - and I had no luck trying to persuade Proxmox staff to look into it. So far they seem rather uninterested in this issue.