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

    alert for network trafic (perf-alert plugin)

    Scheduled Pinned Locked Moved Xen Orchestra
    8 Posts 2 Posters 255 Views 2 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.
    • L Offline
      LubosMaly
      last edited by

      Hello, I'd like to setup alert for network trafic on VMs. I tried to do it in Settings - Plugins via perf-alert plugin. But in "Alarm type" field there is only cpuUsage and memoryUsage to choose from.
      I use version 5.59.0 where is perf-alert plugin v0.2.1. Is there any chance that if I upgrade I will be able to set this kind of alert ? Or there is no way to do it via perf-alert plugin ? Is there another way how to set this up then ?

      Thanks for your help.

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

        Sure, but we'll need a functional spec. What would you expect in there?

        L 1 Reply Last reply Reply Quote 0
        • L Offline
          LubosMaly @olivierlambert
          last edited by

          @olivierlambert
          Thanks for quick reply. I would like to have alert for throughput on each VM, respectively on each VIF.

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

            Can you be more specific: how would you expect to set it, which values and so on. Like imagine you are in front of the UI and describe it (or do some mock design). Thanks!

            L 1 Reply Last reply Reply Quote 0
            • L Offline
              LubosMaly @olivierlambert
              last edited by

              @olivierlambert
              May be I should describe my problem first. Sometimes happen on our network, that there is very slow response (ping) to some HOST and I would like to know (to be informed) which VM is causing the lag. Which VM sends or receives that big amount of data. And Im looking for the way how to do it in xoa or in xe console by any sort of alarm or whatever.

              The best way for me would be in XOA settings - plugins. The same way as there is posibbility to set alarm for cpuUsage and memoryUsage. Just set some limit in (for example KB/s) and if exceeds the limit value send alarm on email. But I can do it in console as well. It dos't matter. I am just asking if there is any way to provide it. And if there is, can you briefly describe how to.

              Thanks

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

                Yes, there's a way, it's just a matter of knowing the context (your provided it, thanks!) and how you imagine to set it up. So a limit in KiB/s triggering the alarm, in a nutshell like this but replacing cpuUsage by networkUsage:

                aaf66c73-eb4c-462d-8ae1-5d55d407e848-image.png

                However, it's a bit more complex:

                1. traffic in or out, or both, accumulated or not?
                2. what if there's multiple virtual interfaces?
                3. and many other questions you are the only one to answer

                More precise your requirement/how you imagine to do it, easier for us to implement without doing 500 back and forth 😛

                L 1 Reply Last reply Reply Quote 0
                • L Offline
                  LubosMaly @olivierlambert
                  last edited by

                  @olivierlambert
                  My very first question was meant, whether there is any alarm for monitoring network ready to use right now.

                  So as I understand (from our further conversation), there is non at the moment.
                  But you can add it to the future versions of XOA.

                  Am I right ?

                  I do not know exactly options to watch (measure) network behaviour in virtual enviroment.
                  Do not know what is possible and what is not.

                  But I think the more precise it could be the better.

                  So

                  1. both would be great, but accumulated better then nothing
                  2. would by great for all VIFs separately, but overall better then nothing
                  3. I will let the rest on you, because I do not know possibilities (background) as I mentioned above.
                    For me above mentioned is enough.

                  I hope I am not the only one who would appreciate this functionality.

                  Regards

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

                    1. That's correct, that's why I asked for a spec, so it's easier to build it while knowing exactly what's requested
                    2. Okay so we can start simple 🙂 I will make the issue on GH and we'll see how hard it is, to know when we can make it 🙂
                    1 Reply Last reply Reply Quote 0
                    • First post
                      Last post