As a ONAP developer, I would like to improve the runtime monitoring capabilities in policy components. So that necessary alerts can be easily triggered and possible outages can be avoided in production systems.

Following are the high level work items planned for it:

• Expose application level metrics in policy components. So that an end user can plugin a prometheus instance and start listening to the metrics exposed by policy components and either raise alerts or show them on a Grafana dashboard for operations team to keep monitoring the health of the system.
• Current api & pap components will be migrated to spring boot framework to facilitate the metrics exposition and easier integration/maintenance.

• Improve the policy api & pap readiness probes to handle database failures. So that api & pap k8s pods are marked ready only if the policy database pod is ready.

 Metrics are documented here - https://wiki.onap.org/display/DW/Policy+Framework%3A+Key+metrics+to+monitor