A prometheus.io exporter written in Python for metrics from the IBM Z Hardware Management Console using zhmcclient. Tested with Python 3.4 through 3.7.
$ pip3 install zhmc-prometheus-exporter
Set up your exporter. Edit your credentials YAML file. A sample credentials YAML is provided. Enter the IP address of the HMC, your username, and your password there. Put this file and metrics.yaml
into /etc/zhmc-prometheus-exporter/
. The latter defines the metrics and descriptions, you do not have to edit it. You can then run
$ zhmc_prometheus_exporter
The default port is 9291, you can change it with -p
. If you do not want to put hmccreds.yaml
and metrics.yaml
into /etc/zhmc-prometheus-exporter
, you can also specify them with -c
and -m
respectively.
If you want a quick "three simple metrics" setup with Prometheus and Grafana you can proceed as follows:
Set up a Prometheus server. Get it from Prometheus. A sample configuration YAML is provided. Fill in the IP and port the exporter will run on. If you left it at default, the port will be 9291. You can then run:
$ ./prometheus --config.file=prometheus.yaml
See also Prometheus' guide.
Set up a Grafana server. Get it from Grafana. You can then run:
$ ./bin/grafana-server web
By default it will be on
localhost:3000
. You will have to set IP and port of the Prometheus server. If you didn't change it, it'slocalhost:9090
. See also Prometheus' guide on Grafana.
- Create the dashboard in Grafana. A sample JSON is provided. If you want it to work natively, you will have to name your source
ZHMC_Prometheus
.
The following image illustrates what the setup described above could look like.