Site alerts should check for multiple requests #243
Labels
component: monitoring
An issue relating to a monitoring component (e.g. Prometheus, Grafana)
component: networking
An issue relating to a host networking (e.g. DNS, WireGuard, SSH)
group: kubernetes
Issues and pull requests related to the Kubernetes setup
We occasionally receive alerts for endpoints that were only hit one or a few
times, in which cases we do not care about e.g. 4xx errors or high latency.
Update our Prometheus alerting rules to only alert for these when multiple
requests were affected by it.
The text was updated successfully, but these errors were encountered: