Team Calendars seems to have an unidentified process kicking off a weekly job that overloads the application
Platform notice: Server and Data Center only. This article only applies to Atlassian products on the Server and Data Center platforms.
Support for Server* products ended on February 15th 2024. If you are running a Server product, you can visit the Atlassian Server end of support announcement to review your migration options.
*Except Fisheye and Crucible
Summary
Every day or every week, around the same time, Team Calendars seems to have an unidentified process kicking off a weekly job that overloads the application and sometimes crashes the app
Environment
- Confluence Server
- Team Calendars
Diagnosis
On a scheduled basis, you notice Confluence running slower and crashing sometimes. In the logs, you can notice the following logs:
2020-10-07 15:24:11,590 WARN [http-nio-8090-exec-391] [analytics.client.listener.ProductEventListener] rejectedExecution Analytics event not be processed (most likely due to too many events in queue for single-threaded event processor).
– url: /rest/calendar-services/1.0/calendar/export/subcalendar/private/6cea42815cedc59b6887d939b47e0faa38677b86.ics | traceId: 9a9e52cedb465609 | userName: anonymous
2020-10-07 15:24:11,592 WARN [http-nio-8090-exec-431] [analytics.client.listener.ProductEventListener] rejectedExecution Analytics event not be processed (most likely due to too many events in queue for single-threaded event processor).
– url: /rest/calendar-services/1.0/calendar/export/subcalendar/private/6cea42815cedc59b6887d939b47e0faa38677b86.ics | traceId: fe079b26be0c69a0 | userName: anonymous
2020-10-07 15:24:11,598 WARN [http-nio-8090-exec-433] [analytics.client.listener.ProductEventListener] rejectedExecution Analytics event not be processed (most likely due to too many events in queue for single-threaded event processor).
– url: /rest/calendar-services/1.0/calendar/export/subcalendar/private/6cea42815cedc59b6887d939b47e0faa38677b86.ics | traceId: 6fbba9fc956f0ebe | userName: anonymous
2020-10-07 15:24:11,600 DEBUG [http-nio-8090-exec-421] [extra.calendar3.calendarstore.AbstractCalendarDataStore] getSubCalendarContent Getting sub-calendar 2028534b-f0b2-48a3-98d5-3dbb958f4a61 events with ActiveObjects.
2020-10-07 15:47:49,317 ERROR [http-nio-8090-exec-430] [confluence.util.profiling.ProfilingSiteMeshFilter] doFilter Application context has not been initialized: – url: /rest/calendar-services/1.0/calendar/export/subcalendar/private/6cea42815cedc59b6887d939b47e0faa38677b86.ics | traceId: eff2f7fc8bba4344 | userName: anonymousCause
Someone responsible for the calendar is forcing a refresh on a scheduled basis during work hours.
Solution
- In the atlassian-confluence.log file, find the log entry similar to the one below:
2020-10-07 15:24:11,600 DEBUG [http-nio-8090-exec-421] [extra.calendar3.calendarstore.AbstractCalendarDataStore] getSubCalendarContent Getting sub-calendar 2028534b-f0b2-48a3-98d5-3dbb958f4a61 events with ActiveObjects.
- Fetch the sub-calendar ID (that in the log above is 2028534b-f0b2-48a3-98d5-3dbb958f4a61) and run the following query against Confluence Database:
SELECT NAME FROM "AO_950DC3_TC_SUBCALS" WHERE "ID" = '2028534b-f0b2-48a3-98d5-3dbb958f4a61';
- Based on the name of the Calendar, engage the respective owner to update their process/job, not during a business time, to avoid overloading your instance.