BerriAI/litellm · error · HTTPException

Failed to update cost discount config: {e}

Error message

Failed to update cost discount config: {e}

What it means

Catch-all HTTP 500 from PATCH /config/cost_discount_config: after all validation passed, the endpoint loads the config via proxy_config.get_config(), mutates litellm_settings, saves back to the DB, and updates litellm.cost_discount_config in memory; if any of those steps raises, the exception is logged ('Error updating cost discount config') and re-raised wrapped in this message with the original error text.

Source

Thrown at litellm/proxy/management_endpoints/cost_tracking_settings.py:262

        # Update cost_discount_config
        config["litellm_settings"]["cost_discount_config"] = cost_discount_config

        # Save the updated config to DB
        await proxy_config.save_config(new_config=config)

        # Update in-memory litellm.cost_discount_config
        litellm.cost_discount_config = cost_discount_config

        verbose_proxy_logger.info("Updated cost_discount_config: %s", cost_discount_config)

        return {
            "message": "Cost discount configuration updated successfully",
            "status": "success",
            "values": cost_discount_config,
        }
    except Exception as e:
        verbose_proxy_logger.error("Error updating cost discount config: %s", e)
        raise HTTPException(
            status_code=500,
            detail={"error": f"Failed to update cost discount config: {e}"},
        )


@router.get(
    "/config/cost_margin_config",
    tags=["Cost Tracking"],
    dependencies=[Depends(user_api_key_auth)],
)
async def get_cost_margin_config(
    user_api_key_dict: UserAPIKeyAuth = Depends(user_api_key_auth),
):
    """
    Get current cost margin configuration.

    Returns the cost_margin_config from litellm_settings.
    """

View on GitHub (pinned to 77b7c6c40c)

Solutions

1. Read the appended original exception ({e}) in the response and the verbose_proxy_logger line — it names the real cause.
2. If it is a DB error, verify PostgreSQL health/connectivity and retry the PATCH.
3. Inspect the stored config row (config table / GET /config/list) for corruption and repair it before re-applying.
4. Retry once after the transient condition clears; the update is not applied if the exception occurred before save.

try:
    r = requests.patch(f"{PROXY_URL}/config/cost_discount_config", json=payload, headers=HDRS, timeout=15)
    r.raise_for_status()
except requests.HTTPError as e:
    detail = e.response.json().get("detail", {}).get("error", "")
    if e.response.status_code == 500 and "Failed to update cost discount config" in detail:
        log.error("proxy-side failure: %s", detail)  # detail embeds the root cause
        verify_state_with_get()  # confirm old config still active, then decide retry
    else:
        raise

When it happens

Trigger: Prisma errors while writing the config row (DB restarted mid-request, connection dropped, permissions); malformed existing config in DB causing serialization/merge failures; transient failures in proxy_config.save_config.

Common situations: Database connectivity blips during the write; a previously corrupted config JSON in the config table; concurrent config mutations racing each other.

Related errors

• Failed to update cost margin config: {e}
• Error fetching cache settings: {e}
• Failed to fetch analytics: {e}
• internal_server_error
• DB not connected. This endpoint needs a database; set DATABA

AI-assisted analysis of BerriAI/litellm@77b7c6c40c (2026-08-18). Data as JSON: /api/errors/121bd5c0f13fa85a. Report an issue: GitHub.