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

    High Thread count on XCP-ng 8.0

    Scheduled Pinned Locked Moved Compute
    6 Posts 2 Posters 473 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.
    • bjerrecsB Offline
      bjerrecs
      last edited by

      Dear XCP-ng Forum.

      We are currently experiencing an XCP-ng 8.0 host having 11k threads compared to other hosts.
      This server has been in production for some time now (3 months) and the high thread count started roughly a week ago.

      At the same time, the SMlog in /var/log is getting ~50MB bigger each day and counting
      I will be able to post some of the SMlog if requested but at this time i'm just looking for imput on this issue

      a7836721-095c-4ce7-9313-3e0f4fca7644-image.png
      Why would this one server be so overloaded?

      Thanks in advance!

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

        Hi,

        Is it the master host?

        1 Reply Last reply Reply Quote 0
        • bjerrecsB Offline
          bjerrecs
          last edited by

          Yes. the host is currently the master in the pool of 3 hosts

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

            So it's the host doing all the coalesce operations on your shared SRs, managing a lot of things slaves aren't.

            bjerrecsB 1 Reply Last reply Reply Quote 0
            • bjerrecsB Offline
              bjerrecs @olivierlambert
              last edited by bjerrecs

              The host currently manages around 18 SR.

              But the underlying questing is that are 11k threads to mutch? How do we define what's good/ not good?

              This comes from a monitoring standpoint. Where we need to be able to define a safe/warning/critical state

              We have another pool of only Citrix Xenserver 7.1 Cu2 hosts where the master only have 1715 threads open

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

                There's no universal answer, check per process what are those threads 🙂

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