BerriAI/litellm · error · Exception

Missing keys={missing_keys} in environment.

Error message

Missing keys={missing_keys} in environment.

What it means

Exception raised by the Lago billing integration's environment validation: it requires LAGO_API_KEY, LAGO_API_BASE, and LAGO_API_EVENT_CODE to all be present as environment variables before it will log billing events to Lago. The message lists exactly which keys are missing.

Source

Thrown at litellm/integrations/lago.py:61

        LAGO_API_EVENT_CODE,

        Optional:
        LAGO_API_CHARGE_BY

        in the environment
        """
        missing_keys: Final = []
        if os.getenv("LAGO_API_KEY", None) is None:
            missing_keys.append("LAGO_API_KEY")

        if os.getenv("LAGO_API_BASE", None) is None:
            missing_keys.append("LAGO_API_BASE")

        if os.getenv("LAGO_API_EVENT_CODE", None) is None:
            missing_keys.append("LAGO_API_EVENT_CODE")

        if len(missing_keys) > 0:
            raise Exception(f"Missing keys={missing_keys} in environment.")

    def _common_logic(self, kwargs: dict, response_obj) -> dict:
        response_obj.get("id", kwargs.get("litellm_call_id"))
        get_utc_datetime().isoformat()
        cost: Final = kwargs.get("response_cost", None)
        model: Final = kwargs.get("model")
        usage = {}

        if (
            isinstance(response_obj, litellm.ModelResponse) or isinstance(response_obj, litellm.EmbeddingResponse)
        ) and hasattr(response_obj, "usage"):
            usage = {
                "prompt_tokens": response_obj["usage"].get("prompt_tokens", 0),
                "completion_tokens": response_obj["usage"].get("completion_tokens", 0),
                "total_tokens": response_obj["usage"].get("total_tokens"),
            }

        litellm_params: Final = kwargs.get("litellm_params", {}) or {}

View on GitHub (pinned to 6c2dcb801b)

Solutions

1. Set all three vars: LAGO_API_KEY, LAGO_API_BASE, LAGO_API_EVENT_CODE
2. For Docker/k8s, ensure the secret/env mapping injects all three into the proxy/SDK process
3. If Lago billing is not intended, remove 'lago' from litellm.callbacks

Example fix

# before
litellm.callbacks = ['lago']  # env vars unset

# after
export LAGO_API_KEY=...
export LAGO_API_BASE=https://api.getlago.com
export LAGO_API_EVENT_CODE=llm-usage
litellm.callbacks = ['lago']

import os

REQUIRED = ("LAGO_API_KEY", "LAGO_API_BASE", "LAGO_API_EVENT_CODE")
missing = [k for k in REQUIRED if not os.getenv(k)]
if missing:
    raise RuntimeError(f"Set required env vars: {missing}")

When it happens

Trigger: Enabling the Lago callback (litellm.callbacks = ['lago']) in a process where one or more of the three LAGO_* env vars are unset. Validation runs at logging time, so the first completion after enabling the callback triggers it.

Common situations: Deploying with the callback enabled but secrets not mounted in the container; .env file not loaded in the runtime; CI/local machine missing the vars while production has them (or vice versa); typo in var names.

Related errors

• invalid LAGO_API_CHARGE_BY set
• External Customer ID is not set. Charge_by={charge_by}. User
• Error: {response.status_code} - {response.text}
• Missing Authorization header
• Invalid bearer token

AI-assisted analysis of BerriAI/litellm@6c2dcb801b (2026-08-15). Data as JSON: /api/errors/634fc79e6956df8e. Report an issue: GitHub.