Conversation
… test-compile.yml
❌ Deploy Preview for autodiscoverylegal failed.
|
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 7216905d3b
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| "workspaces": [ | ||
| "autodiscovery-cli", | ||
| "autodiscovery-contract", | ||
| "frontend-demoland-vite-react", | ||
| "frontend-realdeal" | ||
| "frontend-demoland-vite-react" | ||
| ], |
There was a problem hiding this comment.
The repo still contains autodiscovery-cli and frontend-realdeal and existing production Docker/compose paths still build them, but this workspace list no longer exposes either package to root npm commands. After this change, npm pkg get name --workspace=frontend-realdeal and npm run ... --workspace=autodiscovery-cli report No workspaces found, so root installs/builds only the contract and Demoland packages and the existing CLI/RealDeal workflows become unusable. Keep these packages in workspaces or update the deployment paths that still rely on them.
Useful? React with 👍 / 👎.
| services: | ||
| # Frontend Demoland - React application with nginx (v6.4.1 + React 19 + Tailwind 4) | ||
| frontend: | ||
| image: autodiscovery-preview:latest |
There was a problem hiding this comment.
The deploy job pulls ghcr.io/${{ github.repository }}/autodiscovery-preview:main/:latest before running docker compose, but this Compose file starts the unqualified local tag autodiscovery-preview:latest instead. On a fresh server Compose will try to run or pull that different tag, and on a reused server it can restart a stale local image, so the image built by CI is not what gets deployed. Use the same GHCR image/tag here or parameterize it from the deploy script.
Useful? React with 👍 / 👎.
| if: github.ref == 'refs/heads/main' && github.event.inputs.environment == 'production' | ||
| environment: production | ||
| needs: deploy-staging |
There was a problem hiding this comment.
Production currently has no path to the SSH deploy step. For the workflow_run trigger on main, github.event.inputs.environment is unset, so this if is false; for a manual run with environment=production, deploy-staging is skipped because its condition only accepts develop/staging, and this job is then skipped because it needs that skipped job. Split the workflow_run/manual conditions or make the dependency run/continue for production dispatches.
Useful? React with 👍 / 👎.
1 participant
No description provided.