This page last changed on Aug 19, 2010 by amohdaris.

This page provides a quick overview of the jobs that are scheduled to run regularly in your Confluence instance. More detailed information can be found in Quartz configuration file schedulingSubsystemContext.xml located under confluence/WEB-INF/lib/confluence-x.x.x.jar. Where x.x.x is your Confluence version number.

Job Name Description Behaviour Schedule
backupJob Performs a site backup Run once per cluster Fire at 2am everyday
mailQueueFlushJob Sends notifications that have been queued up Run once per cluster Fire in every minute
referralQueueFlushJob Referrals to Confluence pages are queued up. This job writes this referrals to the database Run in every node Fire in every minute at the 15th second
taskQueueFlushJob Flushes the task queue Run in every node Fire in every minute
localTaskQueueFlushJob
Flushes the local task queue
Run in every node
Fire in every minute
cleanTempDirectoryJob This cleans up temp files created in Confluence home temp directory (created by exports etc.) Run in every node Fire at 4am everyday
dailyReportJob Sends out an email summary of all changes in Confluence to all subscribers Run once per cluster Fire at 12am everyday
clearOldMailErrorsJob Notifications that fail to send due to errors are added tot he mail error queue. This job resets this clear periodically. Run once per cluster Fire at 3am everyday
indexQueueFlushJob Each content update to Confluence needs to be updated in index so search results are accurate. This job flushes changes to the index. Run in every node Fire in every minute
indexOptimizerJob Index optimization is performed to compact the index and maintain searching performance. This task is expensive and does not need to be performed too regularly. If you see Confluence performance deteriorate around 3pm, you can try scheduling this job for 3am only and check that search performance remains reasonable. Run in every node Fire at 3am and 3pm everyday
indexQueueCleanJob This job is responsible for periodically triggering an Index Queue clean to ensure that size of the index queue does NOT grow indefinitely. Run once per cluster Fire at 2am everyday
mailPollJob Polls POP accounts on all spaces that have them configured. Run once per cluster Fire in every minute
clusterSafetyJob Ensures that only one cluster is ever writing to the database at one time. For non-clustered instances, this job is still useful for alerting customers that have accidentally deployed two instances of Confluence against the same database Run once per cluster Fire in every half a minute
updateDidYouMeanIndexJob Each content update to Confluence needs to be updated in Did You Mean index. This job flushes changes to the index. Run in every node Fire in every 2 hours, starting at 12 am
Document generated by Confluence on Oct 12, 2010 20:44