OpenFaaS comes with built-in Prometheus metrics. We provide a collection of Grafana dashboards to customers to help with monitoring OpenFaaS and individual functions.
- Autoscaling and usage metrics dashboard - Get an overview of all deployed functions and metrics for the OpenFaaS gateway.
- Spotlight dashboard - Get detailed CPU/RAM and Rate, Error, Duration (RED) metrics for individual functions.
- Queue worker dashboard - Monitor the work queue for your async function invocations.
- Builder dashboard - Dashboard for monitoring the OpenFaaS Function Builder API.
Autoscaling and usage metrics dashboard¶
The usage metrics dashboard gives you an overview of all the function that you have deployed. It gives you insights into:
- CPU/RAM usage of functions.
- The function replica count and current load.
- Rate, Error, Duration (RED) metrics for the gateway API and individual functions.
Autoscaling and usage metrics dashboard
The spotlight dashboard lets you filter out the data for a single function that you select from a dropdown. This allows you to zoom in and get a good view of what's going on with one function in isolation.
The JetStream queue-worker comes with the addition of metrics to monitor the behaviour of queues. This dashboard can help you get insights in things like:
- The message ingestion rate for each queue.
- The number of messages waiting in the queue to be processed by functions.
- The processing rate.
Dashboard for the OpenFaaS Pro function builder. The builder dashboard can give you better insights in the performance of your builder pods. These metrics include:
- The average build duration per pod
- The invocation rate
- The number of builds in progress