fix: S06.3 add retry/backoff to Docker API calls (3 retries, exponential backoff 1s/2s/4s)

dashboard/backend/main.py

@@ -176,9 +176,25 @@ async def audit_log(request: Request, call_next):
 
 async def monitor_containers():
     import docker
+    import time
+    import logging
 
     from config import DOCKER_HOST
 
+    _monitor_logger = logging.getLogger(__name__)
+
+    def _docker_retry(fn, max_retries=3, base_delay=1):
+        for attempt in range(max_retries):
+            try:
+                return fn()
+            except Exception as e:
+                delay = base_delay * (2 ** attempt)
+                _monitor_logger.warning("Docker monitor error (attempt %d/%d): %s. Retrying in %ds...", attempt+1, max_retries, e, delay)
+                if attempt < max_retries - 1:
+                    time.sleep(delay)
+        _monitor_logger.error("Docker monitor failed after %d attempts", max_retries)
+        return None
+
     client = docker.DockerClient(base_url=DOCKER_HOST)
 
     # Store initial states
@@ -186,7 +202,9 @@ async def monitor_containers():
 
     while True:
         try:
-            containers = await asyncio.to_thread(client.containers.list, all=True)
+            containers = await asyncio.to_thread(
+                lambda: docker_retry(lambda: client.containers.list(all=True)) or []
+            )
             for c in containers:
                 current_status = c.status
                 if c.id in previous_states: