Opened 10 years ago

Closed 10 years ago

#164 closed enhancement (fixed)

job archive database housekeeping at intervals

Reported by: ramonb Owned by: ramonb
Priority: normal Milestone: 1.1
Component: jobarchived Version: 1.0
Keywords: Cc:
Estimated Number of Hours:

Description

when jobarchived is started, a check is performed for any stale or out dated information in the database. For example jobs whose ( start time + requested time ) > current time. Those jobs are then 'closed' in the database.

This can happen when jobs started running while jobarchived is running, but finished when jobarchived was no longer running.

Theoretically this can still occur while jobarchived has not been stopped, but if for example jobmond has stopped running.

The housekeeping checks done at startup of jobarchived should also be performed at an regular interval, to prevent any dummy information in the database. This only becomes an issue for very big systems and after long periods.

Change History (2)

comment:1 Changed 10 years ago by ramonb

  • Owner changed from somebody to ramonb
  • Status changed from new to assigned

comment:2 Changed 10 years ago by ramonb

  • Resolution set to fixed
  • Status changed from assigned to closed

In 857:

jobarchived.py:

  • split checkStaleJobs() into checkTimedoutJobs() and checkStaleJobs()
  • check stale and timedout jobs on startup
  • implemented regular database Housekeeping: after every 20x JobXML iterations: check for timed out jobs
  • changed RRD store thread timewindow from 360-640 seconds to 300-600 seconds
  • closes #164
Note: See TracTickets for help on using tickets.