3 Replies Latest reply on Jan 7, 2010 8:52 AM by robertcbrowning

    Pollng engine question


      What is the proper way technically of shutting down the polling engines. IT seems like the way I have gotten it done is a hassle and was wondering what the SOP is for something like this. I am asking because I am in the middle of a rollout of NPM and 4 polling engines and a hot-swap engine. 

        • Re: Pollng engine question

          .....IT seems like the way I have gotten it done is a hassle.....



          Does "hassle" means that when you trying to shutdown all services, the NPM stays at "stoping" mode?

          When I rollout or apply SP or upgrade, I did the following;

          1. Open Task Manager, under Process Tab, Highlight NetPerfMonService.exe. Move the windows task manager window on one side of the desktop

          2. Open Orion Service Manager. Start > All Programs > Solarwind Orion > Advanced Features > Orion Service Manager

          3. From Orion Service Manager. Shutdown Everything

          4. If I see that the Solarwinds Network Performance Monitor Freezes at stopping mode, I go to the task manager and kill the NetPerfMonService.exe. That's it

          5. Then I can do whatever Rollout I wanted to do. :)






            • Re: Pollng engine question

              We do the same thing.  It seems the more devices you have the longer it takes the service to stop.  (I've never actually waited for it to stop! )

                • Re: Pollng engine question

                  I've got 5 engines, a aux web site & am just loading a HotStandby & each upgrade involves taking the entire complex down. We operate 24x7x364 so it does hurt. I use a twin screen PC & open a whole load of Remote Desktops on one screen sizing them to fit, position the Orion Services manager in a visible portion of each TS with the TaskManager showing the NetPerfMonService & its memory. At a glance I can see the status of all my services on all my Orion servers.

                  I make sure that each server's services are stopped in a controlled sequence using StopEverything. The NetPerfMon can take over an hour to close when it starts to get busy & if you let it, but I normally kill it in TasMgr after the other services have stopped. You can see the NetPerfMon slowly releasing memory in TaskMgr. I use the other screen to do whatever work is required, keeping one of my screens dedicated to showing all of my services. As an SP or product is installed, it automatically tries to stop & start services & this often means that services are re-started before upgrades are completed on the other servers, so I have watch carefully to re-stop services so that I can bring the farm back up in a controlled way.