I'm seeing a lot of problems in 4.9.1 where monitord is not correctly updating the database on large workflows. monitord seems more flaky that in previous releases and I have to run pegasus-monitord --replay on the log files.

It would be good if pegasus can add a job at the end of the workflow (or at the end of the top-level pegasus-dagman) to re-run monitiord and clean up the workflow database.