Add FastAPI health server and align with camera-proxy deploy pattern

.github/workflows/deploy.yml

@@ -95,18 +95,28 @@ jobs:
   infrastructure:
     runs-on: ubuntu-latest
     needs: changes
+    environment: ${{ needs.changes.outputs.tfvars }}
     if: needs.changes.outputs.infra == 'true'
     steps:
       - name: Checkout
         uses: actions/checkout@v4
 
+      - name: Generate token from GitHub App
+        id: app-token
+        uses: actions/create-github-app-token@v1
+        with:
+          app-id: ${{ secrets.AC_APP_ID }}
+          private-key: ${{ secrets.AC_APP_PRIVATE_KEY }}
+          owner: AccelerationConsortium
+          repositories: infrastructure-modules
+
       - name: Checkout infrastructure-modules
         uses: actions/checkout@v4
         with:
           repository: AccelerationConsortium/infrastructure-modules
           ref: main
           path: .modules/infrastructure-modules
-          token: ${{ secrets.GH_PAT }}
+          token: ${{ steps.app-token.outputs.token }}
 
       - name: Configure AWS Credentials
         uses: aws-actions/configure-aws-credentials@v4
@@ -174,6 +184,7 @@ jobs:
   deploy-app:
     runs-on: ubuntu-latest
     needs: [changes, infrastructure]
+    environment: ${{ needs.changes.outputs.tfvars }}
     if: always() && !failure() && !cancelled() && needs.changes.outputs.app == 'true' && github.event_name == 'push'
     steps:
       - name: Checkout

app/Dockerfile

@@ -17,7 +17,9 @@ RUN pip install --no-cache-dir -r requirements.txt
 COPY . .
 RUN chmod +x entrypoint.sh
 
-HEALTHCHECK --interval=60s --timeout=10s --start-period=60s --retries=3 \
-  CMD pgrep -f "main.py" || exit 1
+EXPOSE 8000
+
+HEALTHCHECK --interval=30s --timeout=5s --start-period=30s --retries=3 \
+  CMD python -c "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')" || exit 1
 
 ENTRYPOINT ["./entrypoint.sh"]

app/entrypoint.sh

@@ -29,4 +29,4 @@ else
 fi
 
 echo "Starting uploader agent..."
-exec python main.py run --config /app/config.yaml
+exec uvicorn main:app --host 0.0.0.0 --port ${PORT:-8000}

app/main.py

@@ -1,6 +1,64 @@
-"""Lab Data Uploader Agent — entry point."""
+"""Lab Data Uploader Agent — FastAPI entry point with background scheduler."""
 
-from agent.cli import app
+from __future__ import annotations
 
-if __name__ == "__main__":
-    app()
+import os
+import threading
+from contextlib import asynccontextmanager
+from pathlib import Path
+
+from fastapi import FastAPI
+
+from agent.config import load_config
+from agent.logging_utils import get_logger, setup_logging
+from agent.scheduler import UploadScheduler
+
+_scheduler: UploadScheduler | None = None
+_scheduler_thread: threading.Thread | None = None
+
+
+def _run_scheduler(scheduler: UploadScheduler) -> None:
+    """Run the upload scheduler loop in a background thread."""
+    scheduler.run_loop()
+
+
+@asynccontextmanager
+async def lifespan(application: FastAPI):  # noqa: ARG001
+    """Start scheduler on startup, stop on shutdown."""
+    global _scheduler, _scheduler_thread
+
+    config_path = Path(os.environ.get("AGENT_CONFIG", "/app/config.yaml"))
+    cfg = load_config(config_path)
+    setup_logging(cfg.storage.log_dir)
+    logger = get_logger("main")
+
+    roots = [r.path for r in cfg.watch.session_roots]
+    logger.info(
+        "agent_startup",
+        machine_id=cfg.agent.machine_id,
+        lab_id=cfg.agent.lab_id,
+        session_roots=roots,
+        scan_interval=cfg.agent.scan_interval_seconds,
+    )
+
+    _scheduler = UploadScheduler(cfg)
+    _scheduler_thread = threading.Thread(target=_run_scheduler, args=(_scheduler,), daemon=True)
+    _scheduler_thread.start()
+
+    logger.info("health_server_ready", port=int(os.environ.get("PORT", "8000")))
+    yield
+
+    logger.info("agent_shutdown", reason="lifespan_shutdown")
+    if _scheduler:
+        _scheduler.stop()
+    if _scheduler_thread:
+        _scheduler_thread.join(timeout=10)
+
+
+app = FastAPI(title="Lab Data Uploader Agent", lifespan=lifespan)
+
+
+@app.get("/health")
+def health():
+    """Health check endpoint for ALB."""
+    return {"status": "healthy"}

app/requirements.txt

@@ -5,4 +5,5 @@ tenacity>=8.0,<10
 structlog>=24.0,<26
 typer>=0.12,<1
 boto3>=1.34,<2
-psycopg2-binary>=2.9,<3
+fastapi>=0.115,<1
+uvicorn>=0.32,<1

platform/vars/dev.tfvars

@@ -2,17 +2,17 @@ app_name    = "lab-data-uploader"
 environment = "dev"
 
 # Container configuration
-container_port = 8080
+container_port = 8000
 cpu            = 512
 memory         = 1024
 desired_count  = 1
 
 # Health check
 health_check_path     = "/health"
-health_check_interval = 60
-health_check_timeout  = 10
+health_check_interval = 30
+health_check_timeout  = 5
 
-# No public access needed
+# No public access needed — internal service
 enable_cloudfront       = false
 allow_cloudfront_access = false
 enable_waf              = false
@@ -22,7 +22,7 @@ enable_cognito_auth     = false
 # Environment variables
 environment_variables = {
   ENVIRONMENT = "dev"
-  NFS_MOUNTS  = "100.x.x.x:/labdata:/mnt/lab1"
+  NFS_MOUNTS  = "100.115.219.51:/labdata:/mnt/lab1"
 }
 
 # Secrets (stored in AWS Secrets Manager)

platform/vars/prod.tfvars

@@ -2,17 +2,17 @@ app_name    = "lab-data-uploader"
 environment = "prod"
 
 # Container configuration
-container_port = 8080
+container_port = 8000
 cpu            = 512
 memory         = 1024
 desired_count  = 1
 
 # Health check
 health_check_path     = "/health"
-health_check_interval = 60
-health_check_timeout  = 10
+health_check_interval = 30
+health_check_timeout  = 5
 
-# No public access needed
+# No public access needed — internal service
 enable_cloudfront       = false
 allow_cloudfront_access = false
 enable_waf              = false
@@ -22,7 +22,7 @@ enable_cognito_auth     = false
 # Environment variables
 environment_variables = {
   ENVIRONMENT = "prod"
-  NFS_MOUNTS  = "100.x.x.x:/labdata:/mnt/lab1"
+  NFS_MOUNTS  = "100.115.219.51:/labdata:/mnt/lab1"
 }
 
 # Secrets (stored in AWS Secrets Manager)

platform/vars/test.tfvars

@@ -2,17 +2,17 @@ app_name    = "lab-data-uploader"
 environment = "test"
 
 # Container configuration
-container_port = 8080
+container_port = 8000
 cpu            = 512
 memory         = 1024
 desired_count  = 1
 
 # Health check
 health_check_path     = "/health"
-health_check_interval = 60
-health_check_timeout  = 10
+health_check_interval = 30
+health_check_timeout  = 5
 
-# No public access needed
+# No public access needed — internal service
 enable_cloudfront       = false
 allow_cloudfront_access = false
 enable_waf              = false
@@ -22,7 +22,7 @@ enable_cognito_auth     = false
 # Environment variables
 environment_variables = {
   ENVIRONMENT = "test"
-  NFS_MOUNTS  = "100.x.x.x:/labdata:/mnt/lab1"
+  NFS_MOUNTS  = "100.115.219.51:/labdata:/mnt/lab1"
 }
 
 # Secrets (stored in AWS Secrets Manager)