Now, the metrics-processor also optionally pushes metrics into an InfluxDB instance. This has to be explicitly enabled, so it does not run in CI pipeline and can be opted-in when wanted. It is enabled by default on docker-compose up.
Log in with the credentials set in the docker-compose.yml
On the left sidebar, go to Boards
Open the Metrics board by clicking on its title
If you're already publishing metrics, you should see something like the following (adjust the time range and refresh interval in the top-right corner and the component to watch in the top-left one).
Notes:
To publish metrics, metrics-processor accesses a shared volume with influxdb in which influxdb writes a token with write access to its bucket
Influx auto-imports a dashboard, so it's already set up. You only need to manually configure the time range you want to see and whether auto-refresh is on or not. Then, simply use the variables to choose which component and metric you want to see.
To publish the lower and upper thresholds, the metrics-processor does some small math that would actually belong in the anomaly-detector, so it's a small duplication of code.
Now, the
metrics-processor
also optionally pushes metrics into an InfluxDB instance. This has to be explicitly enabled, so it does not run in CI pipeline and can be opted-in when wanted. It is enabled by default ondocker-compose up
.To access influx:
docker-compose.yml
Boards
Metrics
board by clicking on its titleNotes:
metrics-processor
accesses a shared volume withinfluxdb
in whichinfluxdb
writes a token with write access to its bucketmetrics-processor
does some small math that would actually belong in theanomaly-detector
, so it's a small duplication of code.