Clustered job may run on more than one node if execution time is longer then interval.

Summary

Clustered job may run on more than one node if execution time is longer then interval.

For example:

• Job configured as interval job with 1 min interval, job is a 3-minute-thread-sleep.
• Job will scheduled at node1
• And than 1 min later at node2 while still running on node1

Environment

• JIRA datacenter

Steps to Reproduce

1. Schedule the long interval Job (see above)
2. Wait and monitor the jobs status

Expected Results

Only single job is executed across the cluster.

Actual Results

Two instances of the job will be running in the cluster.

Workaround

Implement cluster lock at Job level. See details: Plugin guide to jira high-availability and clustering