0.15.3 includes native prometheus support
The irondb-prometheus-adapter is deprecated by native IRONdb support. Users are encouraged to switch to native ingestion which has improved performance. This adapter will receive minimal support going forward.
Prometheus Adapter to IRONdb.
Also check out our launch blog post!
Docker Image Pull
You can now install the irondb-prometheus-adapter container image from dockerhub with the below command:
docker pull irondb/irondb-prometheus-adapter docker run -p1234:8080 irondb/irondb-prometheus-adapter:latest -addr :1234 -log debug -snowth http://127.0.0.1:8112
Included is a Makefile which will perform all the build tasks you will need in order to build this service. Below is a quick outline of typical build cases:
make clean # clean and remove prior exe make build # will build `irondb-prometheus-adapter` exe make docker # will build a docker scratch container with irondb-prometheus-adapter inside
After building, you can run in two ways, directly or through docker:
docker run -d -p8080:8080 irondb-prometheus-adapter:<commit_id> # the above will run irondb-prometheus-adapter in a container docker run -p1234:8080 irondb-prometheus-adapter:860a64f4e9d793beaef25196e36a35da1480d88b -addr :1234 -log debug -snowth http:127.0.0.1:8112 # the above shows an example of specifying custom command with non-default args ./irondb-prometheus-adapter # the above will run irondb-prometheus-adapter outside of a container
The above will set the server to listen on port 8080.
The above will set the server to use debug logging. Log levels supported are:
The above will setup two base snowth endpoints that the service will use
to communicate with. You can specify multiple
-snowth params if you know
of more than one in a given topology. The server will interrogate the
listed snowth servers to figure out other nodes in the topology.
There are three endpoints currently:
POST /prometheus/2.0/write/:account/:check_uuid/:check_name POST /prometheus/2.0/read/:account/:check_uuid/:check_name GET /health-check
/health-check endpoint will return with a "message", "commitID" and
"buildTime" in the response.
write endpoint will take in prometheus encoded metric data, and convert
it to IRONdb encoded flatbuffer data and submit to the snowth nodes.
read endpoint will pull IRONdb encoded metric data from the snowth nodes
and convert it to prometheus encoded metric data for response to the caller.