When converting an existing program to use the MDR framework the obvious way to help find problems is to run the original program and MDR program side by side to check the results, to remove the multiple processing and intervals as possible issues keep the MDR program you can run it with just one large interval.

Typical bugs introduced from converting a program to use MDR are:

  • Global variables not being cleared at the end of the interval processing subroutine, if you have no problem with one large interval the problem will most likely be around clearing global variables
  • Repeating code in interval processing that is only required once per background job
  • Variables not being stored for the transformation program, don’t forget anything that needs to be displayed in the results needs to be stored into the results and retrieved by the transformation program

Feedback

Was this helpful?

Yes No
You indicated this topic was not helpful to you ...
Could you please leave a comment telling us why? Thank you!
Thanks for your feedback.

Post your comment on this topic.

Post Comment