7 Replies Latest reply on Jan 17, 2005 2:30 PM by vsop

    Network Peformace Monitor HUGE DB

      I am having an issue where after only a few days of running my DB will go from a compacted size of 256MB to nearly 1GB.  If left running with out compacting for a week or so it gets to nearly 2GB. As soon as I run compact it goes back down to 256MB or so.  Anyone have any ideas? I figure something has to be wrong.  This is version 7.

      Thanks
        • Re: Network Peformace Monitor HUGE DB
          I don't think there is a problem, the new verion (7) of NPM is capturing alot more then the previous version.  It would be nice if there was an option to allowing the data to be stored in SQL server.  I too have the same issue as you.  I have just made it part of my routine to compact the database every friday.

          Lockheed Martin
          Atlas V Launch Operations
          Network Engineer
          • Re: Network Peformace Monitor HUGE DB
            If it was compacting from 500MB down to 256MB I would agree with you but the fact that its going from 2GB to 256MB makes me think something has to be wrong.  I hope so anyway but I guess I like you, I will have to compact that thing weekly.
            • Re: Network Peformace Monitor HUGE DB
              The problem is related the database engine driver (Jet 4.0) from Microsoft. If will only recover unused space when the database has been closed by every process.

              Network Performance Monitor routinely deletes detailed records and summarizes them into hourly and daily records. This causes unused space in the Jet database that can only be recovered by closing the application and performing a compact. Version 7.0 made this problem more evident because it can collect a lot more statistics from each device.

              SolarWinds development is currently looking for a better solution to this problem.
              • Re: Network Peformace Monitor HUGE DB
                I am experiencing this same issue. Solarwinds is working on it, but we have not found a solution yet. I realize Version 7 has the ability to collect data on more devices (Server Volumes for example) but I have not increased the number of interfaces or nodes since Version 5. With the previous Version 5, I could run months before compacting the database. With Version 7, I can't run 2 weeks before reaching 2GB. I have fine tuned Network Performance Monitor to reduce how often it polls and collects statistics and have increased summarization to every 4 days, but this has not made much difference.
                • Re: Network Peformace Monitor HUGE DB
                  Can someone point me to proper process of compressing db?

                  Thanks
                  • Re: Network Peformace Monitor HUGE DB
                    If this "feature" is likely to continue in the next version of the toolkit, I'd like to see a function built in to schedule a shutdown of the Network Performance Monitor, then (when it's completed shutting down - which can take a couple of minutes on my box) to run a compact and repair on the database, then when that's finished to start NPM up again.  That way, I can have automated database tidies running once a week.

                    Cheers,
                    Rob.
                    • Re: Network Peformace Monitor HUGE DB
                      looks like they released a fix for this problem...

                      EE-NPM-Database-Growth-Fix.zip

                      contact tech support and see if this is the fix you are in need of