fix(chat): resolve Auto-route failover event-loop freeze (#812)

Root cause: ensure_structured_column() executed a synchronous blocking
engine.connect().execute(SELECT information_schema.columns ...) inside the
async request path (_build_chat_history / _maybe_compress_history), freezing
the uvicorn event loop and hanging all endpoints (~200s, recovered only by
docker restart).

Fix:
- chat_context.py: remove per-request ensure_structured_column() calls; run
  once idempotently at app startup (main.py lifespan). Use CURRENT_DATABASE()
  for PostgreSQL compatibility.
- main.py: add startup-only ensure_structured_column() hook (try/except).
- llm_client.py: trust_env=False + bounded httpx.Timeout/wait_for so broken
  endpoints fail fast instead of hanging the event loop.
- chat_service.py: run media generation in a background daemon thread with its
  own event loop (_launch_media_task_bg) so SSE end does not cancel image/video
  tasks (#811); remove TEMP faulthandler diagnostic.

Safe-align: persist accumulated live container changes (admin/*, knowledge,
market, models, config, observability, mcp_client, platform_knowledge,
im_bot_platform, chat_media, capability_registry, requirements) and add
workflow_templates.

Verified: live failover test (model-006 -> conn-refused) returns in ~1.9s with
concurrent login 0.83s/200; event-loop responsiveness restored (sub-second).
This commit is contained in:
QihuAI Design committed 2026-07-26 00:35:35 +08:00
1 parent 394a7786e0
commit a2504bce34
27 files changed
+253 -59

No files matched your search

+6 -6
View File
@@ -169,7 +169,7 @@ async def get_im_bot_stats(user=Depends(get_current_user)):
today_stats = _row_to_dict(rows[0]) if rows else {"msg_in": 0, "msg_out": 0, "users": 0, "errors": 0}
# Total routing rules
routes_count = _execute_raw(db, text("SELECT COUNT(*) as cnt FROM im_bot_routes WHERE is_active=1"))[0]._mapping['cnt']
routes_count = _execute_raw(db, text("SELECT COUNT(*) as cnt FROM im_bot_routes WHERE is_active=true"))[0]._mapping['cnt']
return {
"success": True,
@@ -202,7 +202,7 @@ async def get_dashboard(user=Depends(get_current_user)):
SUM(messages_out) as msg_out,
SUM(unique_users) as users
FROM im_bot_daily_stats
WHERE stat_date >= DATE_SUB(CURDATE(), INTERVAL 7 DAY)
WHERE stat_date >= CURRENT_DATE - INTERVAL '7 DAY'
GROUP BY stat_date
ORDER BY stat_date
"""))
@@ -217,7 +217,7 @@ async def get_dashboard(user=Depends(get_current_user)):
LEFT JOIN (
SELECT channel_id, SUM(messages_in + messages_out) as msg_count
FROM im_bot_daily_stats
WHERE stat_date >= DATE_SUB(CURDATE(), INTERVAL 30 DAY)
WHERE stat_date >= CURRENT_DATE - INTERVAL '30 DAY'
GROUP BY channel_id
) s ON c.id = s.channel_id
ORDER BY c.created_at DESC
@@ -357,7 +357,7 @@ async def get_channel_detail(channel_id: str, user=Depends(get_current_user)):
SELECT stat_date, messages_in, messages_out, unique_users, error_count
FROM im_bot_daily_stats
WHERE channel_id = :cid
AND stat_date >= DATE_SUB(CURDATE(), INTERVAL 7 DAY)
AND stat_date >= CURRENT_DATE - INTERVAL '7 DAY'
ORDER BY stat_date
"""), {"cid": channel_id})
channel['stats_7d'] = _rows_to_dict(stats)
@@ -428,7 +428,7 @@ async def activate_channel(channel_id: str, user=Depends(get_current_user)):
db = SessionLocal()
try:
_execute_raw(db, text(
"UPDATE channels SET is_active=1, status='connected' WHERE id=:cid"
"UPDATE channels SET is_active=true, status='connected' WHERE id=:cid"
), {"cid": channel_id})
return {"success": True, "message": "Channel activated"}
except Exception as e:
@@ -443,7 +443,7 @@ async def deactivate_channel(channel_id: str, user=Depends(get_current_user)):
db = SessionLocal()
try:
_execute_raw(db, text(
"UPDATE channels SET is_active=0, status='disconnected' WHERE id=:cid"
"UPDATE channels SET is_active=false, status='disconnected' WHERE id=:cid"
), {"cid": channel_id})
return {"success": True, "message": "Channel deactivated"}
except Exception as e: