XCP-ng
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login

    Backup, CR and snapshots question

    Scheduled Pinned Locked Moved Xen Orchestra
    2 Posts 2 Posters 694 Views 1 Watching
    Loading More Posts
    • Oldest to Newest
    • Newest to Oldest
    • Most Votes
    Reply
    • Reply as topic
    Log in to reply
    This topic has been deleted. Only users with topic management privileges can see it.
    • V Offline
      VoipDude
      last edited by

      Hello world,

      I've been using XenServer for years and the combination of XCP-NG and XenOrchestra is a breath of fresh air.

      I'm running the following setup, with two VMs running on three separate XCP-NG hosts (no shared storage):

      • Site 1:
        • Host 1: 1TB local array, running VM A
        • Host 2: 1TB local array, running VM B
      • Site 2:
        • Host 3: Offsite, 1TB local array, running no VMs.

      Both VM A and B have a single 300GB disk created for them.
      I then use continuous replication of VM A from host 1 to hosts 2 and 3 and VM B from host 2 to host 1 and 3.
      The idea is to be able to survive a catastrophic failure of their main hosts and also if the main site goes offline.

      Given that XenServer was a bit peculiar in the past with storage (ex: not able to reclaim disk space online after deleting a snapshot in older versions if I remember correctly), I have a few questions to double-check and I won't run into limitations in the future.

      I'm also running with local storage with thick provisioning, the local storage is mechanical SAS drives so I don't want to lose performance with fragmentation all over the place.

      My questions are:

      1. Is it "safe" to use continuous replication for this kind of scenario and could this be done hourly with no long-term issues? I.e.: will 4 years from now (when XCP-NG LTS 8.2 will near EOL 😉 ) will this not cause to run into weird errors because the VM has been snapshotted/imported/exported tens of thoudsands of times. From what I understand in your documentation, coalescing fixes this so it should be fine.

      2. Knowing that each host has 1TB local storage, will it be enough place to continue using CR and even snapshots from time to time?
        Let me explain:
        In the past, I tried labbing with 400GB storage VMs, but it refused to create snapshots (from what I understand, XCP-NG wanted to see the whole 400GB space available). (400GB VM + 400 GB CR = less than 400GB remaining)
        Now I use 300GB disk space VMs, so 300GB VM + 300GB CR = more than 300 GB remaining

      Is there any other limitation that I should be aware?

      I also want to use temporary snapshots before major upgrades on the VM, just in case. But those snapshots would be quickly deleted after I confirm the upgrade went correctly. I suppose the disk space would be reclaimed after deleting the snapshot, right?

      Thanks!

      1 Reply Last reply Reply Quote 0
      • olivierlambertO Offline
        olivierlambert Vates 🪐 Co-Founder CEO
        last edited by

        1. Correct, no problem to do that
        2. No other limitation.
        3. Yes, space will be reclaimed thanks to coalesce after that.
        1 Reply Last reply Reply Quote 1
        • First post
          Last post