Hitachi Vantara Pentaho Community Forums
Results 1 to 3 of 3

Thread: How to reduce the cancellation time of a query

  1. #1
    Join Date
    Mar 2015

    Default How to reduce the cancellation time of a query


    When cancelling a Mondrian query through the Saiku UI, I notice that the JVM running Saiku & mondrian keeps on occupying my CPU for 30 seconds till 10 minutes. A screenshot of the profiler shows following stack:
    Name:  image2015-8-19 17-51-58.jpg
Views: 206
Size:  18.0 KB

    When running this on a single-cored server, it even take hours before the query execution is cancelled. Because I thought it might be due to multithreading (RolapResultShepherd class), I reduced the amount of SQL, MDX and Cache threads to 1 in the but to no prevail.

    Can someone explain the inner workings of the cancellation procedure? And do you know any options I can try to make sure the query is cancelled in a shorter period of time?

    Thank you.

  2. #2
    Join Date
    Jan 2013


    In general a cancellation should be picked up and handled immediately. There are some places where cancellation is not checked until an operation is completed, however, and if the operation is long running (like your tuple list merge) then cancellation can be delayed. I believe what you're seeing is captured here:

    Some work has been done over the past year to clean up some places where cancellation is not handled correctly, but 2279 is still outstanding. Vote for it-- that'll increase it's visibility.

  3. #3
    Join Date
    Mar 2015


    Thank you for your answer. I up-voted the issue and started watching it.

    We see that our server gets bogged down because of accumulation of cancelled user queries. You know any remedies?

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.