Monitoring and Observability
PreviousHow to create a Kubernetes cluster for TestgroundNextUnderstanding Testground performance on Kubernetes
Last updated
Last updated
In order to understand what your Kubernetes cluster is doing, you can use Grafana and check the dashboards.
Port forward the Grafana service to a local port. Then access Grafana on http://localhost:3000
Default admin credentials are username: admin ; password: admin
By default the Testground infrastructure playbooks auto-provision a set of dashboards that provide visibility into the Testground infrastructure:
As Testground matures, these dashboards are likely to change.
You can view aggregated resources usage on the whole Cluster with the USE Method / Cluster dashboard.
You can view CPU, memory, network, disk utilisation per node at the USE Method / Nodeedashboard.
In order to understand what your test run is doing, you can use Grafana and view some of the metrics emitted by it to InfluxDB, while it is running, such as the:
Life-cycle events
Diagnostics (i.e. go runtime metrics)
Redis is an integral part of the sync service used to provide synchronisation and coordination between test plan instances. You can check it's utilisation on the Redis dashboard.
WeaveNet is used for the data plane in Testground - all test plan instances communicate with each other over WeaveNet. You can check statistics from the usage of the network at the WeaveNet dashboards.
