Uploaded image for project: 'Observium'
  1. Observium
  2. OBS-2837

VMWare ESXi Device Processors Overview

Details

    • Bug
    • Resolution: Fixed
    • Minor
    • None
    • Professional Edition
    • OS, Web Interface
    • None

    Description

      In the device overview, VMware servers list each processor individually. Instead of the average.

      Is this intended or can you do something there? Because with 40 threads you can scroll, scroll scroll ... to see the Memory / Disks stats. 

      Attachments

        Activity

          [OBS-2837] VMWare ESXi Device Processors Overview

          It works perfectly! Thanks for the quick support. 

          kuhner Richard Kucera added a comment - It works perfectly! Thanks for the quick support. 

          I've made this change, but it won't work on everything. It'll only work on devices where the string before the @ is short enough that it doesn't get clipped by VMWare's SNMP agent.

          adama Adam Armstrong added a comment - I've made this change, but it won't work on everything. It'll only work on devices where the string before the @ is short enough that it doesn't get clipped by VMWare's SNMP agent.

          I also found nothing to change the output at the host. Maybe it would be possible to remove everything from the @, like the unique number. If it is possible to remove it, the output would be clean and identical to the other OS or Hardware.

          This issue is only present with VMware hosts, so i think you can point it to VMware OS, like the unique numbers.

          kuhner Richard Kucera added a comment - I also found nothing to change the output at the host. Maybe it would be possible to remove everything from the @, like the unique number. If it is possible to remove it, the output would be clean and identical to the other OS or Hardware. This issue is only present with VMware hosts, so i think you can point it to VMware OS, like the unique numbers.

          This is nothing to do with us. VMWare exposes the names this way.

          adama Adam Armstrong added a comment - This is nothing to do with us. VMWare exposes the names this way.
          kuhner Richard Kucera added a comment - - edited

          If required I can grant access to our observium or snmp access to our vmware host. 

          kuhner Richard Kucera added a comment - - edited If required I can grant access to our observium or snmp access to our vmware host. 
          kuhner Richard Kucera added a comment - - edited

          Thank you, works great. However, there seems to be a maximum length defined (See image). 

           

          kuhner Richard Kucera added a comment - - edited Thank you, works great. However, there seems to be a maximum length defined (See image).   

          This has been improved as a special case for vmware we now split off the cpu identifier before merging.

          adama Adam Armstrong added a comment - This has been improved as a special case for vmware we now split off the cpu identifier before merging.

          Would it be possible to intercept this from the side of the Observium and delete these unique numbers? Or show only the "Average" use if it is a VMware OS?

          kuhner Richard Kucera added a comment - Would it be possible to intercept this from the side of the Observium and delete these unique numbers? Or show only the "Average" use if it is a VMware OS?

          We aggregate them when they have identical names. In this case VMWare seems to be prepending a unique number which causes them not to be aggregated.

          adama Adam Armstrong added a comment - We aggregate them when they have identical names. In this case VMWare seems to be prepending a unique number which causes them not to be aggregated.

          People

            adama Adam Armstrong
            kuhner Richard Kucera
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: