Following pipeline metrics are exposed at controller-service
on port 9090
There are different kinds of exporters available: Prometheus, Google Stackdriver, etc. They can be configured using observability configuration.
Metric name | Metric type | Labels/tags | Status |
---|---|---|---|
tekton_pipelinerun_duration_seconds_[bucket, sum, count] | Histogram | pipeline =<pipeline_name> pipelinerun =<pipelinerun_name> status =<status> namespace =<pipelinerun-namespace> |
experimental |
tekton_pipelinerun_taskrun_duration_seconds_[bucket, sum, count] | Histogram | pipeline =<pipeline_name> pipelinerun =<pipelinerun_name> status =<status> task =<task_name> taskrun =<taskrun_name>namespace =<pipelineruns-taskruns-namespace> |
experimental |
tekton_pipelinerun_count | Counter | status =<status> |
experimental |
tekton_running_pipelineruns_count | Gauge | experimental | |
tekton_taskrun_duration_seconds_[bucket, sum, count] | Histogram | status =<status> task =<task_name> taskrun =<taskrun_name>namespace =<pipelineruns-taskruns-namespace> |
experimental |
tekton_taskrun_count | Counter | status =<status> |
experimental |
tekton_running_taskruns_count | Gauge | experimental | |
tekton_taskruns_pod_latency | Gauge | namespace =<taskruns-namespace> pod = < taskrun_pod_name> task =<task_name> taskrun =<taskrun_name> |
experimental |