Symantec Management Platform (Notification Server)

 View Only
  • 1.  Measuring Queue Performance Improvement

    Posted Nov 15, 2015 09:57 PM

    We are looking for details on how to measure the current Queue processing to properly assess the improvement in that processing after implementing the following article.

     

    Increasing Queue processing in SMP 7.5 and 7.6

    https://support.symantec.com/en_US/article.TECH211788.html

     

    Before making such a change, it is important to better understand the current situation and assess what the improvement result is.  We have a single server NS solution serving 2600 endpoints.  We are actively deploying patch management and software delivery.  We have a separate DS 6.9 server that is used for realtime software delivery and other operational tasks.  We do need to retire that environment and migrate the functionality into the existing ITMS.  We need to expand what we do with this ITMS and to do that we need to ensure it is performing well. 

     

    Thanks.

     

     



  • 2.  RE: Measuring Queue Performance Improvement

    Posted Nov 16, 2015 11:14 AM
    You can check queue processing performance by running the "Event Queue Statistics" report from Reports > Run Reports > Notification Server Management > Server. Check that the Event Queues are clearing after a reasonable period of time following Inventory. What can kill event queues is collecting too much software inventory. If you expand the scope of the software inventory keep an eye out for this.


  • 3.  RE: Measuring Queue Performance Improvement

    Posted Nov 16, 2015 01:57 PM
      |   view attached

    Thanks for the response.  I have viewed this report previously but perhaps the understanding of what I see in this report is the issue.  I have attached the report for today that spans back to the 9th of Nov.  In my opinion, this does not appear to be having any issues but again I am not certain of what I am looking for.

    I see a very consistent 30 minute set of fast events and a few standard events.  No slow, large or failure entries.  I did notice 1 period that had 40 priority events.  So what I see is that this is working just fine.  I do not however see any information that would suggest how long any given queue processing was taking so I don't see a method for comparing.

     

     

     

    Attachment(s)



  • 4.  RE: Measuring Queue Performance Improvement

    Broadcom Employee
    Posted Mar 17, 2016 06:54 AM

    Maybe these reports will help you to see appropriate information of nse processing?