Exit when lightningd fails before RPC startup

[satwise]

Summary

• replace the indefinite inotifywait --monitor startup block with a bounded wait loop
• detect when lightningd exits before creating lightning-rpc and propagate its exit status
• add regression coverage for early daemon failure and normal RPC startup
• run the lifecycle test in the existing image CI workflow

This allows the Core Lightning app's existing restart: on-failure policy to recover from transient Bitcoin RPC startup failures instead of leaving a wrapper-only container reported as running.

Fixes #1

Testing

bash -n docker-entrypoint.sh tests/docker-entrypoint-test.sh
bash tests/docker-entrypoint-test.sh
git diff --check

[Copilot]

Copilot review overview

🟡 Changes recommended

Moderate findings remain for watcher failure handling and regression-test coverage.

Get a fresh assessment by requesting another Copilot review.

Review effort: Lite
Findings: 1

Open (1)

• Make monitor stub block to catch the previous entrypoint hang · New

What changed in this PR

Updates Docker startup handling so early lightningd failures propagate and containers can restart automatically.

Changes:

• Replaces indefinite RPC monitoring with bounded polling.
• Adds lifecycle regression tests.
• Runs the tests in Docker CI.

File	Summary
tests/​docker-entrypoint-test.sh	Tests early failure and normal startup lifecycles.
docker-entrypoint.sh	Detects early daemon exits and propagates status.
.github/​workflows/​docker.yml	Runs entrypoint lifecycle tests in CI.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.