XCP-ng
    • Categories
    • Recent
    • Tags
    • Popular
    • Users
    • Groups
    • Register
    • Login
    1. Home
    2. vit-zikmund
    V
    Offline
    • Profile
    • Following 0
    • Followers 0
    • Topics 1
    • Posts 3
    • Groups 0

    vit-zikmund

    @vit-zikmund

    0
    Reputation
    1
    Profile views
    3
    Posts
    0
    Followers
    0
    Following
    Joined
    Last Online

    vit-zikmund Unfollow Follow

    Latest posts made by vit-zikmund

    • RE: Overview of a logical allocation of thin provisioned SRs

      Hello @olivierlambert, I've put a comment to the issue. Was it what you expected? The issue itself seemed rather abandoned and nothing moved there since the comment. Is that to be expected? I don't want to sound disrespectful, I know all this is best-effort, but you react surprisingly often! Just want to make sure I didn't miss anything. Some of the changes I suggested I could maybe hand over in a (lame) PR, if such an effort would be appreciated.

      posted in Xen Orchestra
      V
      vit-zikmund
    • RE: Overview of a logical allocation of thin provisioned SRs

      Oh, thanks @olivierlambert, there indeed is such an issue: https://github.com/vatesfr/xen-orchestra/issues/3915

      In the meantime I scratched together a little oneliner to figure out the right number:

      SR='sr-name-label=<sr-name-label>'  # or SR='sr-uuid=<sr-uuid>'
      xe vdi-list "$SR" is-a-snapshot=false params=virtual-size | grep '[^[:space:]]' | sed 's/[^:]*: //' | awk '{s+=$1} END {print s/1024/1024/1024 " GiB"}'
      

      It ignores snapshots, but that's OK for me.

      Anyways, I'll try following up in the issue. Thanks again 😉

      etlweather created this issue in vatesfr/xen-orchestra

      open Show allocated space on SR #3915

      posted in Xen Orchestra
      V
      vit-zikmund
    • Overview of a logical allocation of thin provisioned SRs

      Hello great gurus of XCP-ng and XO!

      Today I got myself into a "little" trouble by accidentally overcommitting my thin provisioned SR, because I falsely believed there's enough space for the theoreticaly-full VDIs I was migrating into it. While that eventually succeeded, later some VDIs grew and the whole space got depleted, and so the VMs ended up with a ro filesystems 😞

      Call me gullible, but I did so because I believed (for reasons I blame myself) this little green graph shows the sum of VDI sizes and not the real filesystem allocation.
      8e089112-abc1-44e3-9fe2-62f195bb242a-image.png

      Take e.g. the VDI I hovered over - kube-demo01-worker03, where it shows 12.32 GiB. When I switch to the "Disks" tab, I see its size to be 100 GiB:
      3189bcde-138a-425b-9930-ae2717da568d-image.png

      I think that it's nice to be able to see what's the real allocation sum of the thin provisioned VDIs, but I believe I also need to know when I will overcommit the space when adding more VDIs. A sum of the VDI full sizes would be a big help, but I haven't found it. Did I miss it somewhere? If not, could that be possibly added? I hope we could elaborate on how.

      Thanks!
      Vit Zikmund

      posted in Xen Orchestra
      V
      vit-zikmund