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

    CPU Usage Mismatch

    Scheduled Pinned Locked Moved Xen Orchestra
    6 Posts 3 Posters 514 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.
    • S Offline
      SomeFixItDude
      last edited by

      Hi I am pretty new to xcp-ng and xoa. The cpu reported in the graphs doesn't match the actual. Or maybe I don't understand something? I also went on to the host and verified the cpu usage matches htop's values. The XOA graph says a pretty consistent 38% usage of one of the cpus.

      alt text

      Not sure how to get the image on here but here is the link: https://ibb.co/t2Bx8VC

      Thanks!

      1 Reply Last reply Reply Quote 0
      • jedimarcusJ Offline
        jedimarcus
        last edited by

        What's your issue? That it says 38 on the left an 31 on the right?

        1 Reply Last reply Reply Quote 0
        • S Offline
          SomeFixItDude
          last edited by

          Thanks for responding @jedimarcus

          I am just missing something possibly... The procs say cpu 1, 2, 3 and they are all 1.3% or lower. If you are referring to the lower area, in the table. That is 30% out of 300% I believe. Again I could just be misunderstanding how its being reporting. This is why I am asking. What i was expecting was, the breakout of cpu 1, 2, and 3 at the top left of htop to match the xoa chart, but these are different values?

          Thanks!

          1 Reply Last reply Reply Quote 0
          • jedimarcusJ Offline
            jedimarcus
            last edited by

            I'm not sure why htop summary says 3x < 1.3% but on the bottom table, you can obviously see that there is a process using >30% CPU...

            I guess this is a htop question, not a XOA or XCP-ng question.

            1 Reply Last reply Reply Quote 0
            • B Offline
              Biggen
              last edited by Biggen

              Did you have more cores/less cores assigned at some point to the VM? I noticed something similar some time ago. It took a host reboot to flush the RRDs which got XO(A) reporting CPU utilization accurately again for me.

              1 Reply Last reply Reply Quote 0
              • S Offline
                SomeFixItDude
                last edited by

                I actually had more cores assigned to the vm earlier. Maybe that caused it.

                1 Reply Last reply Reply Quote 0
                • First post
                  Last post