Dockerfile: migrated to dumb-init for fast container signals #20
Conversation
WalkthroughSwitch Docker base image to Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Docker as Docker daemon
participant Init as /usr/bin/dumb-init
participant App as node (dist/index.js)
Docker->>Init: start container (ENTRYPOINT)
Init-->>Docker: reap signals / forward signals
Init->>App: exec node dist/index.js (CMD)
App-->>Init: respond / exit
Init-->>Docker: exit status forwarded
note right of Init #CFE9F8: dumb-init added to manage\nsignal handling and reaping
sequenceDiagram
autonumber
participant GHAction as GitHub Actions
participant GHCR as ghcr.io
GHAction->>GHCR: authenticate using GITHUB_TOKEN (repository_owner)
GHAction->>GHCR: build and push image f213/mock-openai-api
GHCR-->>GHAction: push result / image digest
note right of GHCR #F0F4C3: Registry switched from Docker Hub to GHCR
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Poem
✨ Finishing Touches🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
CodeRabbit Configuration File (
|
|
Related Documentation Checked 1 published document(s). No updates required. How did I do? Any feedback? |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (3)
Dockerfile (3)
2-2: Base image bump looks good; consider digest pin and clarifying stage.
- Nice move to
node:22-alpine3.21for musl-based small images.- For reproducibility, consider pinning by digest.
- The header says “Build stage” but there’s only one stage. Either rename the comment or convert to a true multi-stage build to drop dev deps from the runtime image.
Apply if you just want to clarify intent:
-# Build stage +# Base image FROM node:22-alpine3.21If you prefer multi-stage (optional skeleton):
-FROM node:22-alpine3.21 +FROM node:22-alpine3.21 AS builder ... -RUN npm i +RUN npm ci ... RUN npm run build ... -EXPOSE 3000 -ENV NODE_ENV=production -ENV PORT=3000 -ENV HOST=0.0.0.0 -ENV VERBOSE=true -ENTRYPOINT ["/usr/bin/dumb-init", "--"] -# Start the application with verbose logging enabled -CMD ["node", "dist/index.js"] +FROM node:22-alpine3.21 AS runner +WORKDIR /app +# Only runtime deps +COPY package*.json ./ +RUN npm ci --omit=dev --ignore-scripts +COPY --from=builder /app/dist ./dist +EXPOSE 3000 +ENV NODE_ENV=production +ENV PORT=3000 +ENV HOST=0.0.0.0 +ENV VERBOSE=true +ENTRYPOINT ["/usr/bin/dumb-init", "--"] +CMD ["node", "dist/index.js"]
5-6: Fix comment typo and validate package pin.
- “Install pages” → “Install dumb-init” (typo).
- Version pin
dumb-init==1.2.5-r3is good for repeatability but can break if the repo rotates; ensure it exists for alpine3.21.Apply the typo fix:
-# Install pages +# Install dumb-init (for proper signal handling as PID 1) RUN apk add --no-cache dumb-init==1.2.5-r3Please verify the pin matches the base repo by running locally:
docker run --rm node:22-alpine3.21 sh -lc 'apk add --no-cache dumb-init==1.2.5-r3 && dumb-init --version'
32-35: ENTRYPOINT via dumb-init: good; add STOPSIGNAL and note behavior change.
- Good use of exec-form with dumb-init to forward signals and reap zombies.
- Recommend declaring STOPSIGNAL to align with graceful shutdown.
- ENTRYPOINT changes container UX (overriding CMD no longer bypasses dumb-init). Document in README or release notes.
Apply:
+STOPSIGNAL SIGTERM ENTRYPOINT ["/usr/bin/dumb-init", "--"]If you prefer not to bake init into the image, an alternative is advising
docker run --init …or Composeinit: truein ops docs.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
Dockerfile(2 hunks)
There was a problem hiding this comment.
Actionable comments posted: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
.github/workflows/docker-publish.yml (1)
65-73: Remove or disable stale Docker Hub description step
TheUpdate Docker Hub descriptionstep still depends on Docker Hub credentials and uses${{ env.IMAGE_NAME }}(now pointing to GHCR), causing failures post-migration. Choose one:
Option A — Guard and point to an explicit Docker Hub repo:
- name: Update Docker Hub description - if: github.event_name != 'pull_request' && github.ref == 'refs/heads/main' + name: Update Docker Hub description + if: github.event_name != 'pull_request' && github.ref == 'refs/heads/main' && env.DOCKERHUB_REPOSITORY != '' uses: peter-evans/dockerhub-description@v4 with: username: ${{ secrets.DOCKERHUB_USERNAME }} password: ${{ secrets.DOCKERHUB_TOKEN }} - repository: ${{ env.IMAGE_NAME }} + repository: ${{ env.DOCKERHUB_REPOSITORY }} readme-filepath: ./README.mdAnd add at top-level
env::DOCKERHUB_REPOSITORY: "" # e.g., "username/repo" if you still maintain Docker HubOption B — Remove this step entirely if Docker Hub is no longer used.
🧹 Nitpick comments (2)
.github/workflows/docker-publish.yml (2)
16-17: Parameterize image name to avoid hardcoding owner/repo.Hardcoding
f213/mock-openai-apimakes forks/org renames brittle. Prefer deriving from the repository.Apply:
- REGISTRY: ghcr.io - IMAGE_NAME: f213/mock-openai-api + REGISTRY: ghcr.io + IMAGE_NAME: ${{ github.repository }}
33-39: Step name and auth now target GHCR, not Docker Hub.Rename for clarity; config is correct for GHCR with
GITHUB_TOKENandpackages: write.- - name: Log in to Docker Hub + - name: Log in to GHCR if: github.event_name != 'pull_request' uses: docker/login-action@v3 with: registry: ${{ env.REGISTRY }} username: ${{ github.repository_owner }} password: ${{ secrets.GITHUB_TOKEN }}
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
.github/workflows/docker-publish.yml(3 hunks)
1 participant
Current container implemntation takes 10 secs to stop after pressing Ctrl-C. This happens because of node.js not handling docker signals.
Summary by CodeRabbit