Dashboards

Available monitoring dashboards

Axual makes the assumption that each client will have their own monitoring environments already available. Grafana and Prometheus are standard monitoring environments that Axual exports statistics and dashboards for.

Grafana dashboards are automatically provisioned when starting Axual. The following dashboards are available to monitor the platform:

Dashboard name

Screenshot

Contents

Cluster Overview
Cluster overview

A high level dashboard showing the cluster status and metrics

Broker stats per node
Broker stats per node

A dashboard useful to zoom in to the metrics for a single broker node

Discovery status
Discovery status

A dashboard showing the (instance specific) status of the Discovery API and other metrics

Distributor Overview
Distributor overview

Dashboard showing offset, schema and message distributor status and metrics

VM Stats
VM Stats

A useful dashboard to show a VM’s status, e.g. disk, CPU and memory usage

Docker Monitoring
Docker monitoring

Dashboard zooming in on the Docker statistics of the VMs

Prometheus Overview
Prometheus overview

Dashboard used for checking Prometheus metrics

Monitoring Connect
connect overview

Dashboard provides a general overview of Connect service of a given instance

Monitoring Connectors
connect overview

Dashboard which exposes metrics that concern individual connectors

Rest-Proxy Detailed Overview
Rest-Proxy Detailed Overview

Dashboard which exposes metrics that concern to Rest-Proxy

Cluster-API Detailed Overview
Cluster-API Detailed Overview

Dashboard which exposes metrics that concern to Cluster-API

Message Delays
Cluster overview

Shows the message end to end delays experienced inside an own cluster and after distribution

When starting an new installation is advisable to monitor server metrics. The customer should monitor the time of operations and amount of operations performed. Are there times when an operation spikes in reference to resources used or time?

If there are spikes of bad performance then what causes the spikes? Try to correlate the spike with the amount of messages being send into the system and also CPU usages on the system. You should be able to answer the questions.

  • What is the normal system load?

  • Are there any (Cron) jobs running on the system. That slow the system at certain times.

  • Is there a level of CPU, Network Usage or Disk Usage that correlate to bad performance?

  • If the system load has increase then it there a need to add more brokers to handle the load?

  • Is there a need to add more Connect Instances or Rest Proxy instances to handle extra load?

  • Are the servers loaded evenly? Should you think about using a system such as Cruise Control to balance load?

  • What is a peak load that can be efficiently handled by the system.

Please do not only use Grafana to visually search for issues, but in times of good system performance, periodically create baseline statistics for your system. If there are ever performance issues you will not have a good idea of how to improve the system if you have not spent time in creating baseline statistics.