Dockerfile (1)

177-178: Consider adding a startup grace period to reduce cold-start false negatives.

A short --start-period typically improves reliability when services take time to bootstrap.

Proposed tweak

-HEALTHCHECK --interval=30s --timeout=5s --retries=3 \
-    CMD curl -sf http://127.0.0.1:18789/health || exit 1
+HEALTHCHECK --interval=30s --timeout=5s --start-period=45s --retries=3 \
+    CMD curl -fsS http://127.0.0.1:18789/health || exit 1

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@Dockerfile` around lines 177 - 178, The Dockerfile's HEALTHCHECK currently
lacks a startup grace period, causing cold-starts to be reported as failures;
update the HEALTHCHECK directive that uses "HEALTHCHECK --interval=30s
--timeout=5s --retries=3 CMD curl -sf http://127.0.0.1:18789/health || exit 1"
to include a suitable --start-period (for example --start-period=60s or another
value appropriate for your service) so Docker will wait before beginning health
probes, keeping the rest of the flags intact.

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

Dockerfile.base (1)

128-129: Derived test image inherits insufficient base HEALTHCHECK

The base HEALTHCHECK (lines 128-129) only validates that node can execute, not that the actual service is running. The verification confirms that test/Dockerfile.sandbox (line 8 of provided context) inherits from ${BASE_IMAGE} but does not define its own HEALTHCHECK override, leaving it vulnerable to false-positive health reports.

To mitigate this, either require all child images to define a service-level HEALTHCHECK probe, or parameterize the base command so inheritors can inject a meaningful healthcheck (e.g., a startup script or port check).

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@Dockerfile.base` around lines 128 - 129, The base HEALTHCHECK is too
superficial (only running node) and causes child images like
test/Dockerfile.sandbox to inherit a false-positive probe; add a build-time
argument (e.g., ARG HEALTHCHECK_CMD) and use it in the HEALTHCHECK instruction
so inheritors can supply a meaningful check, or alternatively remove the base
HEALTHCHECK and document that every child image must define its own HEALTHCHECK;
update the HEALTHCHECK line to reference HEALTHCHECK_CMD and ensure
test/Dockerfile.sandbox sets HEALTHCHECK_CMD (or defines its own HEALTHCHECK) so
the probe actually validates the service.

test/Dockerfile.sandbox (1)

89-92: Remove the redundant || exit 1 suffix.

The node -e "process.exit(0)" command already returns non-zero on failure, which Docker HEALTHCHECK treats as unhealthy. The || exit 1 adds no functional value.

♻️ Suggested simplification

 # Test image: no long-running service, so just verify the runtime works.
 # Overrides any inherited HEALTHCHECK to avoid false-positive probes.
 HEALTHCHECK --interval=30s --timeout=5s --start-period=10s --retries=3 \
-    CMD node -e "process.exit(0)" || exit 1
+    CMD node -e "process.exit(0)"

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@test/Dockerfile.sandbox` around lines 89 - 92, The HEALTHCHECK line includes
a redundant shell suffix "|| exit 1" after the command `node -e
"process.exit(0)"`; remove the `|| exit 1` so the HEALTHCHECK simply runs `node
-e "process.exit(0)"` (Docker already treats a non-zero exit from that command
as unhealthy). Update the HEALTHCHECK invocation (the line containing
HEALTHCHECK --interval=30s ... CMD node -e "process.exit(0)" || exit 1) to drop
the `|| exit 1` fragment.

Dockerfile.base (1)

131-138: Well-documented baseline health check — minor simplification possible.

The approach is sound: a minimal runtime validation for the base image, with clear guidance that child images override with service-specific probes. The parameters are reasonable.

The || exit 1 suffix is redundant—if node fails to execute or exits non-zero, Docker already treats it as unhealthy. You can simplify to just the command.

🔧 Optional simplification

 HEALTHCHECK --interval=30s --timeout=5s --start-period=45s --retries=3 \
-    CMD node -e "process.exit(0)" || exit 1
+    CMD node -e "process.exit(0)"

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@Dockerfile.base` around lines 131 - 138, Remove the redundant shell fallback
at the end of the HEALTHCHECK; update the instruction that currently uses
'HEALTHCHECK ... CMD node -e "process.exit(0)" || exit 1' to simply run the node
command directly (i.e., keep the HEALTHCHECK options and use CMD node -e
"process.exit(0)" without the '|| exit 1' suffix) so Docker can interpret the
exit status of Node itself.