3 Replies Latest reply on May 20, 2011 9:35 AM by byrona

    Virtualization Monitoring Scheduler Problems?

    byrona

      It seems that there may be problems with the scheduler responsible for the virtualization monitoring. 

      When ever Orion crashes and I have to recover the system (which I have had to do two times in the last two weeks due to database crashes) the virtuazliation monitoring doesn't resume. To recover the virtualization monitoring I need to go into VMWare Settings, disable polling on all of my VMWare systems and then re-enable polling on all of those systems.  Doing this gets the polling jobs re-scheduled properly and things begin working again.

      My question; is this a known issue?

        • Re: Virtualization Monitoring Scheduler Problems?
          lukas.belza

          Hi byrona,

          The all polling jobs for VMware systems are rescheduled when the "SolarWinds Orion Module Engine" service is (re)started on main poller. There is one known issue with rescheduling nodes on additional pollers. If "SolarWinds Orion Module Engine" service can't connect to job scheduler on additional poller (poller is down or services are stopped) then the rescheduling failed for nodes on this additional poller. Then your approach with disabling/re-enabling polling is necessary.

          This situation can also happen for nodes on main poller if "SolarWinds Job Scheduler" services is stopped when "SolarWinds Orion Module Engine" service starts, but Configuration Wizard or Orion Service Manager application start the Scheduler service before Module Engine service. So this situation shouldn't happen if the "SolarWinds Job Scheduler" service isn't stopped manually.

          Could you confirm one of these situations?

          Thank you,

          Lukas Belza (SolarWinds development)