Time Series Metric Collection over HTTP/REST for Jira Mail Queue Observability

Description

Jira Data Center does not have observability for mail queues running independently on different nodes. Monitoring and observability are not possible at all for the mail service job, it is not restarted in case of a failure and troubleshooting is very complicated in case of intermittent or node-specific problems.

Suggested Solution:

Implementing Prometheus compatible or similar HTTP/s and/or REST (JSON) endpoints to expose metric information on the nodes. Option to save the metrics in the Jira logs like performance metrics would be beneficial.

Sample metrics for Mail Queue
sent_emails_last5min
error_emails_last5min
sent_emails – Since the restart
error_emails – Since the restart
mail_queue_status - S/E/R Success Error Restart etc.

Only available examples so far with limited information on mail queue:

• There is an Open Source 3rd Party Solution for this purpose Prometheus Exporter For Jira in the Prometheus Exporters
• This Open Source solution is also provided via a Plug-in from Atlassian Marketplace Prometheus Exporter PRO for Jira

Current limited metrics provided by plug-in

# HELP jira_mail_queue_gauge Mail Queue Gauge
# TYPE jira_mail_queue_gauge gauge
jira_mail_queue_gauge 0.0

The described metrics for the Jira in this plug-in are essential for Jira observability and monitoring and future containerization requirements.

Workaround

none