This exporter exposes the statistics provided by a specific ArangoDB instance in a format compatible with prometheus.
To use the ArangoDB Exporter, run the following:
arangodb-exporter \
--arangodb.endpoint=http://<your-database-host>:8529 \
--arangodb.jwtsecret=<your-jwt-secret> \
--ssl.keyfile=<your-optional-ssl-keyfile>
This results in an ArangoDB Exporter exposing all statistics of
the ArangoDB server (running at http://<your-database-host>:8529
)
at http://<your-host-ip>:9101/metrics
.
To run the ArangoDB Exporter in docker, use an image such as
arangodb/arangodb-exporter:0.1.3
.
There are several ways to configure Prometheus to fetch metrics from the ArangoDB Exporter.
Below you're find a sample Prometheus configuration file that can be used to fetch metrics from an ArangoDB exporter listening on localhost port 9101 (without TLS).
global:
scrape_interval: 15s
scrape_configs:
- job_name: arangodb
static_configs:
- targets: ['localhost:9101']
For more info on configuring Prometheus go to its configuration documentation.
If you're using the Prometheus Operator
in Kubernetes, you need to create an additional Service
and a ServiceMonitor
resource
like this:
kind: Service
apiVersion: v1
metadata:
name: arangodb-exporters-service
labels:
app: arangodb-exporter
spec:
selector:
app: arangodb-exporter
ports:
- name: metrics
port: 9101
---
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
name: arangodb-exporter
namespace: monitoring
labels:
team: frontend
prometheus: kube-prometheus
spec:
namespaceSelector:
matchNames:
- default
selector:
matchLabels:
app: arangodb-exporter
endpoints:
- port: metrics
scheme: https
tlsConfig:
insecureSkipVerify: true
Note 1: that the typical deployment on the Prometheus operator is done in
a namespace called monitoring
. Make sure to match the namespace
of the ServiceMonitor
to match that.
Note 2: that the Prometheus
custom resource has a field called serviceMonitorSelector
.
Make sure that the matchLabels
selector in there matches the labels of
your ServiceMonitor
.
To build this project, you need Go 1.8 or higher and Docker installed. Then run:
DOCKERNAMESPACE=<your docker hub account name> make