On this page

The monitoring stack is a Linux server toolkit for tracking and visualizing application resource usage in real time. This stack uses a set of open-source tools, including InfluxDB and Grafana, as well as Java Management Extensions (JMX) and various server agent plugins to collect and display metrics. The monitoring stack can be used to evaluate Teamwork Cloud/Magic Collaboration Studio system usage and performance. Metrics for Java operations, database backend processes, and server resource usages can be tracked to isolate performance issues, as well as optimize system resources.

Components

The following components constitute the monitoring node:

• InfluxDB - time series database for collecting the metric data
• Grafana - dashboard visualization layer
• Jmxtrans - Collector to retrieve JMX metrics from remote hosts

The following components are deployed on each Teamwork Cloud/Cassandra node:

• Telegraf - system metrics collector
• Dropwizard metrics-graphite-3.1.2.jar - metrics publishing agent for Java

On a single node installation, all components may be installed on the same node. On multi-node systems, the monitoring node components may be installed on one of the nodes (or on a separate machine), while the other components are deployed on each node.

Ports

The following are default ports used by services running on the monitoring node.

The following sections provide instructions on how to set up the monitoring stack.