I have a problem with a TRIRIGA environment. Due to an inefficient integration, we have a large number of records with a negative Object ID ready to be cleaned by the Cleanup Agent. But it seems that the Cleanup Agent can’t clean that many records.
Can I configure something or execute something into the database to clean everything in parts? I guess that I cannot simply execute a delete into the database, where the object_id < 0, since there must be more things to clean.
No, you should not go into the back-end and execute a delete on the database. There is a lot more to clean up. What do you mean the Cleanup Agent cannot clean that many records? Are you getting an exception? The OBJECT_ID must be negative, the OBJECT STATE should be null, and the UPDATED_DATE must be over 24 hours earlier for the Cleanup Agent to pick up the records to delete. It also might take a couple runs of the Cleanup Agent to delete all of the associated records.
[Admin: This post is related to the 02.24.16 post about cleaning up after retiring a record, and the 07.14.15 post about cleaning up millions of workflow instance records. As a side note, starting with version 3.4, the Cleanup Agent name was changed to Platform Maintenance Scheduler.]