App monitoring can give you a deeper insight into what apps are doing in your instance. This can be useful when troubleshooting issues with a specific app, or to help you determine whether an app may have contributed to a drop in overall performance or stability.
You'll need a copy of the Java agent and the Prometheus JMX Exporter.
You'll also need a running instance of Prometheus and Grafana. Details for setup and configuration can be found in the EAP documentation.
Overview dashboard- This provides a high-level overview of your instance. It contains metrics such as CPU and Memory utilisation, as well as basic alerting e.g. Memory Utilisation is over 80% for a 5 minute duration. (Note: Database connection pool graphs will only display if a JNDI Datasource is configured)Cache clearing- This provides insights into how apps are utilising their caches. It also provides insights into apps enabling/disabling as these events can result in clearing of caches.Cluster locks- This identifies which apps are holding and waiting for specific locks.Database load- This provides insights into how apps are querying the database through the Atlassian AO and SAL frameworks.Response times- This shows how long apps are taking to serve requests. It includes aspects such as condition evaluation (e.g. permissions), web panel rendering and http rest request times.
Contributions to DC App Monitoring Dashboards are welcome! Please see CONTRIBUTING.md for details.
Copyright (c) 2021 - 2022 Atlassian and others. Apache 2.0 licensed, see LICENSE file.
