[WIP] Copilot Request

[Copilot]

Thanks for asking me to work on this. I will get started on it and keep this PR's description up to date as I form a plan and make progress.

Apply a fix for failing GitHub Actions job 81921774491 in repository lantanagroup/link-cloud (related to PR #1673).

Observed failure:

• Docker compose startup fails with: "dependency failed to start: container kafka-rest-proxy is unhealthy"
• Failing workflow: .github/workflows/tests.yaml
• Context shows service rest-proxy marked unhealthy shortly after startup.

Task:

1. Update docker-compose configuration for the Kafka REST Proxy service to make startup/health robust in CI.

   • Ensure bootstrap/listener environment variables are explicit and compatible with internal compose networking.
   • Ensure depends_on waits for kafka broker health where supported.
   • Improve REST proxy healthcheck command and timings (interval/retries/start_period) to avoid premature unhealthy state.
   • Remove obsolete compose version key if present.

2. Update .github/workflows/tests.yaml (or relevant script section) to wait explicitly for rest-proxy health before running dependent steps/tests.

   • Add a bounded retry loop and clear diagnostics (service health + logs on failure).
   • Keep behavior deterministic and fail fast with useful output.

3. Keep changes minimal and focused only on CI startup reliability for rest-proxy.

4. Open/update PR with these changes and include a concise summary in the PR body/commit message referencing the failing job id 81921774491.

The following is the prior conversation context from the user's chat exploration (may be truncated):

User: Please find a solution for failing job 81921774491. Use the logs, job definition, and any referenced files where the failure occurred. Keep your response focused on the solution and include code suggestions when appropriate.

Assistant: The failure is not from .NET warnings (those builds show 0 Error(s)), it’s from Docker Compose integration startup:

• dependency failed to start: container kafka-rest-proxy is unhealthy
• failing service detected as rest-proxy
• workflow: .github/workflows/tests.yaml (ref 7ae05370d23671364c926373d9cd93e8ea6dfbf0)

Root cause

kafka-rest-proxy healthcheck fails during startup. The logs show REST Proxy boots, then becomes unhealthy before dependencies are ready (most likely broker/listener mismatch or too-strict health timing). This is the gate that kills the job.

---

Solution

1) Fix rest-proxy container config in docker-compose.yml

Use explicit bootstrap server and remove schema-registry dependency unless truly required.

services:
  kafka-rest-proxy:
    image: confluentinc/cp-kafka-rest:7.6.1
    depends_on:
      kafka-broker:
        condition: service_healthy
    environment:
      KAFKA_REST_BOOTSTRAP_SERVERS: "PLAINTEXT://kafka-broker:9092"
      KAFKA_REST_LISTENERS: "http://0.0.0.0:8082"
      KAFKA_REST_HOST_NAME: "kafka-rest-proxy"
      KAFKA_REST_CLIENT_SECURITY_PROTOCOL: "PLAINTEXT"
      # only if schema registry exists in compose:
      # KAFKA_REST_SCHEMA_REGISTRY_URL: "http://schema-registry:8081"
    healthcheck:
      test: ["CMD-SHELL", "curl -fsS http://localhost:8082/ || exit 1"]
      interval: 10s
      timeout: 5s
      retries: 12
      start_period: 40s

2) Make CI less brittle in .github/workflows/tests.yaml

Before running tests, explicitly wait for REST Proxy healthy (instead of relying only on compose up timing).

for i in {1..24}; do
  status=$(docker inspect --format='{{if .State.Health}}{{.State.Health.Status}}{{else}}none{{end}}' kafka-rest-proxy || true)
  echo "kafka-rest-proxy health=$status"
  [[ "$status" == "healthy" ]] && break
  sleep 5
done

[[ "$status" == "healthy" ]] || {
  docker compose logs rest-proxy
  exit 1
}

3) Optional cleanup

Remove deprecated version: from docker-compose.yml (warning seen in logs). Not the failure, but good hygiene.

---

If you want, I can draft an exact patch for both .github/workflows/tests.yaml and docker-compose.yml targeted at ref 7ae05370d23671364c926373d9cd93e8ea6dfbf0.

User: Yes

Assistant: Great — I can prepare the patch, but I need one clarification first:

Do you want me to apply the fix directly to PR #1673 (by creating commits via the coding workflow), or just provide the exact diff text here for you to apply manually?

User: Directly.

Assistant: Understood — I’ll apply the fix directly to PR #1673 by updating the compose/workflow configuration to stabilize kafka-rest-proxy startup and health checks.

User: Okay.. so ... uh.. gonna do it, or what?

Assistant: I am — I’ve already kicked off the direct-fix action, and we’re currently waiting on your approval in the confirmation dialog before it can execute.