Ian Billett

This PR implements an example which defines and generates all of the configuration needed to run a fully-featured Prometheus instance on Kubernetes.

Getting Dir & InternalDir mixed up - is there a better way?

1

Knowing when to use `Dir` & `InternalDir` is confusing and getting them mixed up can lead to file permission issues when your containers start up. For example, when trying to...

Can't reasonably load Querier UI with prom-label-proxy

3

Since https://github.com/prometheus-community/prom-label-proxy/pull/48 was merged, users are required to explicitly state an allowlist of endpoints (`unsafe-passthrough-paths`) that any tenant can retrieve without interference of `prom-label-proxy`. This means that any UI components...

Should we support per-tenant store endpoints?

1

Currently `prom-label-proxy` does not support tenants being able to retrieve stores that present their identifying label set. As this endpoint is not parameterised, implementation would presumably follow the pattern established...

Kubed panics when restricted to specific named secrets and configmaps

3

Hello 👋 Nice project! It is helping me navigate LE's certificate rate limits :) I bumped into a panic when deploying kubed into my cluster... For obvious reasons, I don't...

``` slo.latencyburn({ alertName: 'TelemeterServerMetricsWriteLatencyErrorBudgetBurning', alertMessage: 'Telemeter Server /upload or /receive is burning too much error budget to gurantee latency SLOs', metric: 'http_request_duration_seconds', selectors: ['job="telemeter-server"', 'handler=~"upload|receive"', 'code!~"4.."'], latencyTarget: 5, latencyBudget: 0.9,...