main - Fix SSE reconnect loop: catch errors, set retry 60s, return HTTP status on failure

main.py

@@ -158,16 +158,23 @@ async def logs_stream():
     token = os.environ.get("HF_TOKEN", "")
 
     async def _generate():
+        retry = "retry: 60000\n"
         if not token:
-            yield "data: {\"type\":\"stream\",\"data\":\"HF_TOKEN not configured\\n\"}\n\n"
+            yield f"{retry}data: {{\"data\":\"HF_TOKEN not configured\\n\"}}\n\n"
             return
-        async with httpx.AsyncClient(timeout=None) as client:
-            async with client.stream(
-                "GET", _HF_LOGS_URL,
-                headers={"Authorization": f"Bearer {token}"},
-            ) as resp:
-                async for chunk in resp.aiter_bytes():
-                    yield chunk
+        try:
+            async with httpx.AsyncClient(timeout=None) as client:
+                async with client.stream(
+                    "GET", _HF_LOGS_URL,
+                    headers={"Authorization": f"Bearer {token}"},
+                ) as resp:
+                    if resp.status_code != 200:
+                        yield f"{retry}data: {{\"data\":\"HF API returned {resp.status_code}\\n\"}}\n\n"
+                        return
+                    async for chunk in resp.aiter_bytes():
+                        yield chunk
+        except Exception as exc:
+            yield f"{retry}data: {{\"data\":\"Stream error: {exc}\\n\"}}\n\n"
 
     return StreamingResponse(
         _generate(),