PrometheusΒΆ
Metric |
Target |
|---|---|
RPO |
n/a (not backed up) |
RTO |
1 hour |
The main Prometheus DB lives in Hetzner. It is configured to federate from multiple other Prometheuses and is the place where we configure alerting.
It is provisioned using Ansible, like the rest of the stack in Hetzner.
Prometheus is not backed up as it does not hold any particularly sensitive data, and has a retention period of a couple of months.
Should Prometheus need to be re-created, that can be done by re-applying this playbook.