...
| Info |
|---|
The location of the log depends on your configuration. By default you will find it (Linux) /opt/atlassian/jira/logs/catalina.log |
2. Check if webhook queue is full
Set the appropriate logging level in Jira (com.atlassian.webhooks).
Watch out for entries in the logs of the following format:
| Code Block |
|---|
XXXX-XX-XX XX:XX:XX,XXX-XXXX https-jsse-nio-8443-exec-278 url: /rest/greenhopper/1.0/sprint/<ID>/start; user: <user-id> WARN <user-id> <id> n8qvx1 <IP>,<IP> <trigger-event> [c.a.webhooks.plugin.WebHookPublisherImpl] Executor rejected the web hook 'PublishTask{listenerKey=com.atlassian.webhooks.api.register.listener.WebHookListener@2a5116c5, uri=<webhook-url>, body={<webhook-body>} rejected from java.util.concurrent.ThreadPoolExecutor@40453c55[Running, pool size = 3, active threads = 3, queued tasks = 100, completed tasks = 1000002]' |
Explanation:
Jira has an internal queue (size=100). This queue gets processed by a HTTP client with a worker-pool (threads) of 3 (https://jira.atlassian.com/browse/JRASERVER-70950, https://bitbucket.org/atlassian/atlassian-webhooks-plugin/src/atlassian-webhooks-3.3.0/plugin/src/main/java/com/atlassian/webhooks/plugin/WebHookPublisherImpl.java) If all these 3 connections are used, the queue starts to fill up. If a lot of events need to be processed the queue will reach the limit of 100. Any new request which can not make it into the queue will fail and result in the error above.
Resolution:
Try to limit the amount of webhooks setup in Jira. Try generate less webhooks it not only helps to delete / deactivate configured webhooks in Jira but also limit the scope whenever a webhook triggers by applying a strict JQL filter.
4 Check if webhook add-on (and all modules) is activated in JIRA
...