We provide two dashboards for Grafana, leveraging the exporter.
In addition to ha_cluster_exporter, these dashboards require Prometheus node_exporter to be configured on the target nodes.
They also assume that the target nodes in each cluster are grouped via the job label.
This dashboard gives an overview of multiple clusters monitored by the same Prometheus server.
This dashboard shows the details of a single cluster.
On openSUSE and SUSE Linux Enterprise distributions, you can install the package via zypper in your Grafana host:
zypper in grafana-ha-cluster-dashboards
systemctl restart grafana-server
For the latest development version, please refer to the development upstream project in OBS, which is automatically updated everytime we merge changes in this repository.
Copy the provider configuration file in /etc/grafana/provisioning/dashboards and then the JSON files inside /var/lib/grafana/dashboards/sleha.
Once done, restart the Grafana server.
Dashboards will be soon available on grafana.com/dashboards
- Please make sure the
versionfield in the JSON is incremented just once per PR. - Unlike the exporter, OBS Submit Requests are not automated for the dashboard package.
Once PRs are merged, you will have to manually perform a Submit Request, after updating theversionfield in the_servicefile and adding an entry to thegrafana-ha-cluster-dashboards.changesfile.