BerriAI/litellm · error · Exception

Not connected to DB!

Error message

Not connected to DB!

What it means

POST /customer/update raises a bare Exception('Not connected to DB!') when prisma_client is None; the surrounding handler passes it through handle_exception_on_proxy, which surfaces it to the caller as an HTTP 500. Same root cause as the standardized db_not_connected_error used elsewhere, but via a plain exception on this route.

Source

Thrown at litellm/proxy/management_endpoints/customer_endpoints.py:624

    --data '{
        "user_id": "user_1",
        "object_permission": {
          "mcp_servers": ["server_3"],
          "vector_stores": ["vector_store_2", "vector_store_3"]
        }
      }'

    See below for all params
    ```
    """

    from litellm.proxy.proxy_server import litellm_proxy_admin_name, prisma_client

    try:
        data_json: Final = _STR_OBJECT_DICT.validate_python(data.json())
        # get the row from db
        if prisma_client is None:
            raise Exception("Not connected to DB!")

        # get non default values for key
        non_default_values: Final = dict[str, object]()
        for k, v in data_json.items():
            if v is not None and v not in (
                [],
                {},
                0,
            ):  # models default to [], spend defaults to 0, we should not reset these values
                non_default_values[k] = v

        ## Get end user table data ##
        end_user_table_data: Final = await _typed_table(EndUserRepository(prisma_client)).find_first(
            where={"user_id": data.user_id}, include={"litellm_budget_table": True}
        )

        if end_user_table_data is None:
            raise ProxyException(

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Set DATABASE_URL to a postgresql:// connection string and restart the proxy
2. Verify Prisma connectivity in startup logs and via another DB-backed endpoint
3. Retry the update once the DB-backed health check passes
4. Skip update calls on deployments intentionally running without a database

import os


def update_endpoint_ready() -> bool:
    return os.getenv("DATABASE_URL", "").startswith(("postgresql://", "postgres://"))

try:
    r = httpx.post(f"{base}/customer/update", json=payload, headers=headers)
except httpx.HTTPStatusError as e:
    if e.response.status_code == 500 and "Not connected to DB" in e.response.text:
        raise RuntimeError("proxy has no DATABASE_URL; /customer/update needs Postgres") from e
    raise

When it happens

Trigger: POST /customer/update with any payload on a proxy started without DATABASE_URL or whose Prisma connection never came up.

Common situations: Attempting customer updates on routing-only deployments; DATABASE_URL unset in one environment (staging) while tooling assumes parity with another (prod); DB outages after boot.

Related errors

• Postgres DB Not connected
• DB not connected. This endpoint needs a database; set DATABA
• {e}
• Failed updating customer data. User ID does not exist passed
• user_id is required, passed user_id = {data.user_id}

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/7de2f7e639525998. Report an issue: GitHub.