Details

    • Bug
    • Resolution: Incomplete
    • Major
    • None
    • Professional Edition
    • Poller, Web Interface
    • None
    • OS : CentOS 6.7 (64Bit)
      RAM : 8GB
      CPU : 3
      This is a VM machine running on ESXi6

    Description

      Recently I purchased Enterprise edition. Initially I haven't hit with any issue. But all of sudden , there is some white space[ not plotting] on my graphs. Not sure what happened all of sudden. Even I updated the package with latest version. Though, Still its no luck.

      Note : I couldn't notice any error in the log.

      Moreover , I purchased this product to overcome my existing monitoring system which have less info compare to observium. But this unknown issues are too weird to me. Since My old monitoring system doesn't have any issue in plotting.

      Attachments

        1. 31.png
          31.png
          137 kB
        2. 34.png
          34.png
          102 kB
        3. 35.png
          35.png
          131 kB

        Activity

          [OBS-1729] Intermittent white space in Graphs

          I had this same problem but it started once I added more devices (I have 33 devices now and I have more to add) and after a couple weeks of testing, I found that I needed to change how many poller instances I have running. I started with turnkey observium and that was running one poller instance. I moved over to ubuntu 16.04 server and installed observiem. It ran great with 15 devices on 4 poller instances but once I added more I started to have gaps in the graphs and SMNP errors.

          This page gives you some good info on it. http://www.observium.org/docs/tuning/

          schutzenhoffer Tommy Schutzenhoffer added a comment - I had this same problem but it started once I added more devices (I have 33 devices now and I have more to add) and after a couple weeks of testing, I found that I needed to change how many poller instances I have running. I started with turnkey observium and that was running one poller instance. I moved over to ubuntu 16.04 server and installed observiem. It ran great with 15 devices on 4 poller instances but once I added more I started to have gaps in the graphs and SMNP errors. This page gives you some good info on it. http://www.observium.org/docs/tuning/
          landy Mike Stupalov made changes -
          Workflow Original: classic default workflow [ 12834 ] New: Observium workflow [ 14703 ]
          adama Adam Armstrong made changes -
          Resolution New: Incomplete [ 4 ]
          Status Original: Open [ 1 ] New: Closed [ 6 ]
          kmg Gugan added a comment -

          Thanks for your reply.

          As per your advice , I will check the mentioned points [ Poller overlapping and its statistics for devices , also will check which SNMP queries got failed ]. Will share you my findings. Thanks Again.

          kmg Gugan added a comment - Thanks for your reply. As per your advice , I will check the mentioned points [ Poller overlapping and its statistics for devices , also will check which SNMP queries got failed ]. Will share you my findings. Thanks Again.

          This is caused by missing data, broken snmp queries or overlapping poller runs.

          You've either got some packetloss or other issue between your observium server and the monitored device, or your device is taking too long to poll.

          You can tell which by looking at the poller performance statistics for that device, and looking at the "MIB" section of the device's settings to see what SNMP queries have failed.

          You can sometimes improve this by increasing SNMP timeouts/retries.

          Beyond that, there's little we can change to fix it

          adama Adam Armstrong added a comment - This is caused by missing data, broken snmp queries or overlapping poller runs. You've either got some packetloss or other issue between your observium server and the monitored device, or your device is taking too long to poll. You can tell which by looking at the poller performance statistics for that device, and looking at the "MIB" section of the device's settings to see what SNMP queries have failed. You can sometimes improve this by increasing SNMP timeouts/retries. Beyond that, there's little we can change to fix it
          kmg Gugan made changes -
          Attachment New: 34.png [ 13032 ]
          Attachment New: 35.png [ 13033 ]
          kmg Gugan created issue -

          People

            landy Mike Stupalov
            kmg Gugan
            Votes:
            0 Vote for this issue
            Watchers:
            3 Start watching this issue

            Dates

              Created:
              Updated:
              Resolved: