feat(observability): GrafanaDashboard CR + SLO row, export to grafana-agent#8
Merged
Conversation
…O row, export to grafana-agent
Brings the tenant's observability to its correct prod shape — the org runs
grafana-operator + grafana-agent (Alloy) → Amazon Managed Prometheus/Grafana,
not a kube-prometheus-stack sidecar or a cluster otel-collector.
- Dashboard delivery: the vendored tenant-chart-base helper now emits the
GrafanaDashboard CR (instanceSelector dashboards: external) the grafana-operator
reconciles onto Amazon Managed Grafana — the portable path that works on both EKS
and the local kx cluster. The whole vendored library is synced to the canonical
nanohype skeleton (adds _servicemonitor.tpl; _helpers gains commonLabels merging
for the tagging-governance labels).
- SLO row: the board leads with a crawl-availability SLO row — availability (30d),
error-budget remaining, and burn rate — inline over the real good/bad counter
competitive_intelligence_crawl_sources_total{outcome}, self-contained so it renders
against AMP with no recording-rule ruler.
- Telemetry path: OTLP exports to the grafana-agent OTLP receiver
(grafana-agent.monitoring.svc:4318), which forwards traces → Tempo, metrics → AMP,
logs → Loki; the NetworkPolicy allows egress to the monitoring namespace on 4318.
- Docs: CLAUDE.md / ARCHITECTURE / RUNBOOK / chart README / Dockerfile / metrics.ts
describe the real backends (AMP/Tempo/Loki via grafana-agent), not Grafana Cloud /
Mimir / a cluster otel-collector.
The Loki logs row is a follow-up — the exact Alloy stream-label selector needs
verifying on a live cluster. Closes #7.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Brings the tenant's observability to its correct prod shape (grafana-operator + grafana-agent → AMP/AMG, not a kube-prometheus sidecar or cluster otel-collector).
tenant-chart-basehelper now emits the CR (instanceSelectordashboards: external) the grafana-operator reconciles onto Amazon Managed Grafana. The vendored library is synced to the canonical nanohype skeleton (adds_servicemonitor.tpl;_helpersgainscommonLabelsfor the tagging-governance labels).competitive_intelligence_crawl_sources_total{outcome}counter, self-contained (no ruler needed).grafana-agent.monitoring.svc:4318(forwards traces→Tempo, metrics→AMP, logs→Loki); NetworkPolicy egress opened to themonitoringnamespace.Validated:
helm templateemits the CR (no stale ConfigMap) with OTLP→grafana-agent; dashboard JSON parses; helm lint clean. The Loki logs row is a follow-up (the Alloy stream-label selector needs live-cluster verification).Closes #7.