bytedance/deer-flow · error · HTTPException

MCP task repo not available

Error message

MCP task repo not available

What it means

HTTP 503 from get_mcp_task_repo() when app.state.mcp_task_repo is None. The MCP task repository stores MCP-server task records; it is attached during Gateway lifespan as part of persistence bootstrap.

Source

Thrown at backend/app/gateway/deps.py:668

def get_scheduled_task_run_repo(request: Request):
    val = getattr(request.app.state, "scheduled_task_run_repo", None)
    if val is None:
        raise HTTPException(status_code=503, detail="Scheduled task run repo not available")
    return val


def get_scheduled_task_service(request: Request):
    val = getattr(request.app.state, "scheduled_task_service", None)
    if val is None:
        raise HTTPException(status_code=503, detail="Scheduled task service not available")
    return val


def get_mcp_task_repo(request: Request):
    val = getattr(request.app.state, "mcp_task_repo", None)
    if val is None:
        raise HTTPException(status_code=503, detail="MCP task repo not available")
    return val


def get_mcp_task_service(request: Request):
    val = getattr(request.app.state, "mcp_task_service", None)
    if val is None:
        raise HTTPException(status_code=503, detail="MCP task service not available")
    return val


def get_run_context(request: Request) -> RunContext:
    """Build a :class:`RunContext` from ``app.state`` singletons.

    Returns a *base* context with infrastructure dependencies. The
    ``app_config`` field is resolved live so per-run fields (e.g.
    ``models[*].max_tokens``) follow ``config.yaml`` edits; the
    ``event_store`` / ``run_events_config`` pair stays frozen to the snapshot
    captured in :func:`langgraph_runtime` so callers never see a store bound

View on GitHub (pinned to 1dd6ba1acb)

Solutions

1. Check Gateway startup logs for the MCP task repo init exception
2. Verify the persistence database is reachable and the relevant tables/migrations exist
3. Restart the Gateway after fixing persistence so lifespan re-attaches mcp_task_repo

null

null

try:
    tasks = await api.list_mcp_tasks()
except HTTP503 as e:
    if 'MCP task repo' in e.detail:
        tasks = []
    else:
        raise

When it happens

Trigger: Calling MCP task routes (list/create/delete MCP task records) on a Gateway whose lifespan did not attach mcp_task_repo — persistence init failure or running against an app instance without the Gateway lifespan.

Common situations: Database unavailable at startup so the repo never constructed; version mismatch between frontend expecting MCP task endpoints and a backend build without them; routers exercised in tests without full lifespan.

Related errors

• Thread metadata store not available
• MCP task service not available
• {label} not available
• Scheduled task run repo not available
• Configuration not available

AI-assisted analysis of bytedance/deer-flow@1dd6ba1acb (2026-08-14). Data as JSON: /api/errors/eec9985e87060e82. Report an issue: GitHub.