Troubleshooting
Problem
Often the number of records in yfs_task_q table becomes huge. It becomes difficult to purge them in a single process. One of the approaches is to purge the records in batches. One of the following techniques can be implemented to purge records in batches.
Resolving The Problem
Below are the options to run the order purge by manipulating Hold Flag.
1. Modify the hold flag to 'Y' in the yfs_task_q for the necessary records which should not be picked up by purge agent. Once the purge agent has completed a full run, modify the hold flag again to 'N' for the necessary records which should get picked up by the purge agent. This needs to be done continually till all the purgeable records are deleted from task_q table.
OR
2. Modify the available date in yfs_task_q to a future date for the records which should not be picked up by the purge agent. Agent will pick only those records whose available date is older than the current date. Set 'incremental' future dates for different sets of records so that you don't have to do the updates again and again. After the updates, run the purge agent continually till all the purgeable records are deleted from task_q.
For both the above approaches, make changes in task_q, only to the order purge records. Also, note that it is recommended that the API 'manageTaskQueue' be used for updating the yfs_task_q records. SQL update should be avoided unless necessary.
Was this topic helpful?
Document Information
Modified date:
11 February 2020
UID
swg21617013