Average Time In Status Gadget Causes High CPU Utilization

Description

Running an 'expensive' JQL on the Average Time in Status gadget can consume resources and cause long running threads and can cause JIRA to become unresponsive.

Steps to Reproduce

  • Create a gadget on a test dashboard for "Average Time in Status"

  • Configure the gadget with the following settings:

    • Filter: that returns ALL issues

    • Statuses: select ALL status

    • Period: Hourly

    • Days Previously: 300 (which appears to be the maximum)

    • Refresh Interval: Every 15 mins

Expected Result

Results show

Actual Results

JIRA server CPU utilization spikes and accessing administration pages is unresponsive.

The below is an example of the size of an instance that has experienced this:

1 2 3 4 5 6 7 8 9 10 ___ Database Statistics ____________________ Issues : 258704 Projects : 438 Custom Fields : 131 Workflows : 48 Users : 5767 Groups : 9626 Attachments : 280911 Comments : 1077221

Is there some type of boundary limit or check that we can implement to prevent these 'expensive' queries from running?

Environment

None

Testing Notes

None

Status

Assignee

Unassigned

Reporter

ChrisA

Labels

None

Add-on Type

None

Team

None

CC

None

Risk factor

None

QA Kickoff Status

None

QA Demo Status

None

Components

Affects versions

1.20

Priority

Major