RH-Che (OSIO) has this feature already implemented, where a <ws_master url>/metrics endpoint can be exposed to a port, providing Prometheus compatible metrics, allowing us to observe JVM behavior inside che-master container in real time and collect data for zabbix.

This feature is enabled via environment variable during deployment phase.

This is something that would be really helpful for performance testing both for Che QA team (Maksym Musiienko) and RH Che QE team (Tibor Dancs) to observe real-time accurate behaviour of the che-master under test (e.g.: CPU utilization, memory consumption, disk/gluster activity,...).

RELATED_ISSUES:

• RH Che QE:
  • https://github.com/redhat-developer/che-functional-tests/issues/409
  • https://github.com/redhat-developer/che-functional-tests/issues/383
• Che QE:
  • https://issues.jboss.org/browse/CRW-68

p.s.: Please don't crucify me, this is the first time I have created a JIRA issue