Repository navigation
NO-ISSUE: chore(deploy)!: drop the observability stack and NFD from moai-inference-framework - #191
Merged
Conversation
Contributor
There was a problem hiding this comment.
Copilot review overview
🟡 Changes recommended
Public guidance exposes internal repository details, and the generated chart documentation retains an obsolete prerequisite.
Review effort: Balanced
Findings: 3
Open (3)
What changed in this PR
Removes bundled observability and NFD components from the framework Helm chart in favor of separate GitOps management.
Changes:
- Removes observability/NFD dependencies, values, templates, dashboards, and alerts.
- Retains only common, kubernetes-replicator, and ECR token refresher functionality.
- Regenerates chart metadata and contributor guidance.
| File | Description |
|---|---|
deploy/helm/AGENTS.md |
Updates Helm contribution guidance. |
deploy/helm/moai-inference-framework/Chart.yaml |
Removes observability and NFD dependencies. |
deploy/helm/moai-inference-framework/Chart.lock |
Regenerates the dependency lock. |
deploy/helm/moai-inference-framework/values.yaml |
Removes deleted components’ configuration. |
deploy/helm/moai-inference-framework/README.md |
Regenerates dependency and values documentation. |
deploy/helm/moai-inference-framework/templates/_helpers.tpl |
Removes obsolete storage helpers. |
deploy/helm/moai-inference-framework/templates/tempo/credentials.yaml |
Removes Tempo credentials. |
deploy/helm/moai-inference-framework/templates/loki/credentials.yaml |
Removes Loki credentials. |
deploy/helm/moai-inference-framework/templates/minio/init-job.yaml |
Removes MinIO provisioning. |
deploy/helm/moai-inference-framework/templates/nfd/moai-accelerator.yaml |
Removes NFD accelerator rules. |
deploy/helm/moai-inference-framework/templates/grafana/heimdall-slack-configmap.yaml |
Removes Slack alert provisioning. |
deploy/helm/moai-inference-framework/templates/grafana/datasource-tempo.yaml |
Removes the Tempo datasource. |
deploy/helm/moai-inference-framework/templates/grafana/datasource-loki.yaml |
Removes the Loki datasource. |
deploy/helm/moai-inference-framework/templates/grafana/dashboard-configmap.yaml |
Removes dashboard provisioning. |
deploy/helm/moai-inference-framework/templates/grafana/alert-configmap.yaml |
Removes alert provisioning. |
deploy/helm/moai-inference-framework/files/dashboards/mif-aigateway.json |
Removes the AIGateway dashboard. |
deploy/helm/moai-inference-framework/files/dashboards/amd-gpu-usage-monitor.json |
Removes the AMD GPU dashboard. |
deploy/helm/moai-inference-framework/files/alerts/heimdall-templates.yaml |
Removes alert notification templates. |
deploy/helm/moai-inference-framework/files/alerts/heimdall-rules.yaml |
Removes Heimdall alert rules. |
deploy/helm/moai-inference-framework/files/alerts/heimdall-policies.yaml |
Removes alert routing policies. |
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
| - Use official upstream repos: loki `https://grafana.github.io/helm-charts`, vector `https://helm.vector.dev`, minio `https://charts.min.io`, tempo `https://grafana-community.github.io/helm-charts` (Grafana's community-maintained charts repo, which carries current `tempo-distributed`). | ||
| - Keep `prometheus-stack.defaultRules.create: true`. The bundled "Kubernetes / Compute Resources" Grafana dashboards (panels *and* their `workload`/`type` template variables) query kubernetes-mixin recording rules such as `namespace_workload_pod:kube_pod_owner:relabel`, so disabling default rules blanks those dashboards even though raw metrics are still scraped. | ||
| - The chart bundles only `common` and `kubernetes-replicator`. Observability (Prometheus, Loki, Tempo, Vector, MinIO, Grafana dashboards and alerts) and NFD are deployed by GitOps (`moreh-dev/gitops`, `moreh-dev/gitops-customers`), not by this chart. Do not add them back. | ||
| - Add a sub-chart only with a `condition:` entry in `Chart.yaml` and official upstream repos. |
Comment on lines
20
to
21
| | https://helm.mittwald.de | replicator(kubernetes-replicator) | 2.12.2 | | ||
| | https://helm.vector.dev | vector | 0.39.0 | | ||
| | https://prometheus-community.github.io/helm-charts | prometheus-stack(kube-prometheus-stack) | 80.7.0 | | ||
| | oci://registry-1.docker.io/bitnamicharts | common | 2.31.4 | |
…oai-inference-framework Prometheus, Loki, Tempo, Vector, MinIO, Grafana dashboards/alerts and NFD are deployed by GitOps now. Published chart versions are unaffected.
juwonhwang-wq
force-pushed
the
NO-ISSUE-drop-observability-from-mif-chart
branch
from
September 30, 2026 03:41
76dc471 to
5f41934
Compare
hhk7734
approved these changes
Sep 30, 2026
This was referenced Sep 30, 2026
2 tasks done
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
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.
요약
moai-inference-framework차트에서 관측 스택(prometheus-stack, loki, tempo, vector, minio, Grafana 대시보드·알림)과 NFD를 뺍니다. 이 컴포넌트들은 GitOps 로 별도 배포합니다. 차트에는common,kubernetes-replicator,ecrTokenRefresher만 남습니다.영향
확인
helm dependency update,helm lint통과helm template은 ecrTokenRefresher 리소스만 렌더링make helm-docs로 README 재생성후속
website/docs(quickstart, prerequisites, monitoring/traces, latest-release)와 release 스킬은 아직 번들 스택을 전제로 씁니다. 다음 릴리스 전에 별도 PR 로 고칩니다.