console/deployment/grafana-dashboards/README.md

44 lines
1.8 KiB
Markdown

# Grafana Dashboards (Hive Cloud)
This directory includes Grafana dashboard used internally for managing and monitoring the Hive Cloud
service.
> If you wish to use these metrics and dashboards, please contact us for additional information.
## Datasources in use and their names
- `CDN`: Datasource of kind `vertamedia-clickhouse-datasource`
(https://grafana.com/grafana/plugins/vertamedia-clickhouse-datasource/) linked to CloudFlare
Worker Analytics (https://developers.cloudflare.com/analytics/analytics-engine/get-started/)
- `Azure Monitor`: https://grafana.com/docs/grafana/latest/datasources/azure-monitor/
- `grafanacloud-logs`: Loki data source for logs
- `grafanacloud-prom`: Prometheus data source with metrics scraped from Hive services
## Adding a new dashboard
Create a new dashboard in Grafana dashboard, and then click the "Share" icon, pick "Export" tab and
then copy/export the JSON file. Place the file in this directory, and you are done, Pulumi will pick
it up in next deployment.
## Editing an existing dashboard
You may edit dashboards, but note that it will be override by the next deployment. To update a
dashboard, update it first in Grafana, and then click the "Share" icon, pick "Export" tab and then
copy/export the JSON file. Replace the file as-is.
## Parameters
Some aspects of the dashboard might be effected/related to environment-specific variables. To pass
variables, add to the Pulumi stack configuration file, a key under `grafanaDashboards:params` with
the `variable = value`, for example:
```yaml
grafanaDashboards:params:
SOME_VAR: value
```
Then, in the dashboard's JSON model file, you can use the variable name, as is (we do a simple
string replacement).
> The `TABLE_SUFFIX` is injected automatically, with the name of the Pulumi stack ("prod" is
> replaced with "production").