Problem Definition

Currently there is no way to monitor the thread pool queue. When there is an issue, we can see an error message indicating pool size, active threads, queued tasks, etc.:

Caused by: java.util.concurrent.RejectedExecutionException: Task com.atlassian.sal.core.executor.ThreadLocalDelegateRunnable@1ef6143e<com.atlassian.sal.core.executor.ThreadLocalDelegateRunnable@1ef6143e> rejected from java.util.concurrent.ThreadPoolExecutor@3a65b61f[Running<java.util.concurrent.ThreadPoolExecutor@3a65b61f[Running>, pool size = 10, active threads = 10, queued tasks = 200, completed tasks = 30604323]

Before getting the error, we don't have a way to proactively monitor these values.

Suggested Solution

Add an option for live monitoring of the above values (pool size, active threads, queued tasks, and completed tasks) as JMX Monitoring Beans for example, and in that way, we could suggest using Live Monitoring Using the JMX Interface.

Why this is important

There is currently no other way for Jira administrators to monitor the thread pool successfully and that would be useful for them to resize the values based on concrete information of their environment if needed.

Workaround

n/a