curious i see all this talk of proxmox and kvm, wondering why nobody is using XCP-NG ofor VMs with management ui, or ven StarlingX for that matter. If your into kubernetes and working on k8s/virtualization.

  • UntouchedWagons@alien.topBEnglish
    1·
    1 year ago

    I find Proxmox a lot easier to learn, you don’t need to worry about deploying a management interface and it has ZFS support.

    • Mezoloth@alien.topBEnglish
      1·
      1 year ago

      You can use ZFS with XCP also. I was using it until recently when I went with NVME storage.

  • Haribo112@alien.topBEnglish
    1·
    1 year ago

    I prefer XCP-ng over Proxmox because my first professional work experience was as an intern at a company that used Citrix XenServer so all I know about hypervisors is rooted in that.

    • Mr_1984@alien.topBEnglish
      1·
      1 year ago

      Same boat there. Years of Zen so I was happy at how similar they were. Although xcp-ng has been much more stable for me. Zen had weird network issues at my old job (but that was probably caused by our odd networking).

  • EtherMan@alien.topBEnglish
    0·
    1 year ago

    Xcp is a lot stricter, for good or bad. For proxmox it’s fine to have networks on each host that looks roughly the same. For xcp, the network has to ne 1:1 equal to be in the same pool. This makes proxmox much easier to deal with administratively. But for xcp the upside is that there’s no guesswork involved, which is much more stable. Not that proxmox is unstable, but it’s not like it never crashes. I have yet to see xcp crash even once over several years I’ve messed with it.

    • bufandatl@alien.topBEnglish
      1·
      1 year ago

      That’s it true. You can have different networks in one pool. I have a Host with 4 NICs and a second with 2 just 2 NICs in one pool but you can’t move VMs between both hosts if your VM has a notiere attached that is not present at the second host.

  • neovb@alien.topBEnglish
    0·
    1 year ago

    I wanted to like XCP-NG because it’s easy to use, but the limitation of 2 terabytes per VHD was the killer for me. I guess for most VMs you won’t need that much storage, but that was a deal breaker.

    There may be some way to get around that limitation (which is an inherent issue with XCP-NGs storage stack) or use software RAID in an OS to create arrays of 2 TiB disks, but that’s a lot of effort for nothing.