6 Replies Latest reply on Jan 15, 2013 8:42 PM by njoylif

    NPM 10.4 Questions


      I have a couple of questions regarding lines in the release notes that I'd like some clarification on.


      Improved handling of interruptions in the connection between polling engines and the database.

      Could someone explain exactly what this means?  If we momentarily lose the DB will polling queue up and then write back to the DB upon restoration?


      It is now possible to configure the maximum application pool size, making additional SQL connections available.

      How is this done?  I looked through the admin guide, core components guide, and the managing orion performance technical reference and can't find anything in relation to pool size.

        • Re: NPM 10.4 Questions

          Bueller, Bueller.

          • Re: NPM 10.4 Questions

            Good questions.....where IS Bueller??

            • Re: NPM 10.4 Questions

              The polling engines can now support connection interruptions from the database.  They basically use a store and forward, so when the connection is lost they will queue the data until the connection is restored.  They should be able to handle a connection interruption of around 45-50 minutes.


              To configure the maximum application pool size you'll need to add ";Max Pool Size=250" to the end of the connection string setting within the SWNetPerfMon.DB file and restart the services.



              • Re: NPM 10.4 Questions

                If I may interject a couple of related questions on this subject please.


                On the Polling Engines page, what is the difference between the Last Database Update and the Last Database Sync?


                For each, what are the "normal" values? Thank you.

                  • Re: NPM 10.4 Questions

                    Hi patriot,


                    Last Database Update - is time difference between current time and the latest synchronization on Nodes, that mean last time when node was polled and result data are stored back to DB. This value is global for Database and valid value is lower than minimal poll interval (by default 120 seconds). You can check your latest Node update with this SQL statement: SELECT TOP 1 DATEDIFF(SECOND, LastSync, GetDate()) AS TimeDiff  FROM Nodes WITH(NOLOCK) ORDER BY LastSync DESC


                    Last Database Sync - is time difference between current time and the last period update of "KeepAlive" column in "Engines" table, this updates are periodically done by BusinessLayer (60 seconds by default). Value is specific per Polling Engine and valid value is lower then defined period interval. You can check your period interval when these writes are done with with this SQL statement: SELECT CurrentValue FROM Settings WHERE SettingID = 'SWNetPerfMon-Settings-Keep Alive Timeout'