mirror of
https://github.com/BerriAI/litellm.git
synced 2026-09-14 23:21:35 +00:00
fix(prometheus): add _safe_labels helper to propagate exclude_labels to observation sites
When prometheus_exclude_labels strips labels at registration time, observation call sites that still pass the full positional/keyword label set raise ValueError. Add _safe_labels(metric, **kwargs) that filters kwargs to the metric's registered _labelnames. Apply it to: - guardrail observation calls (litellm_guardrail_latency/requests/errors) - litellm_remaining_api_key_requests/tokens_for_model (convert positional → keyword) - litellm_deployment_cooled_down (convert positional → keyword) - litellm_llm_api_failed_requests_metric (convert positional → keyword)
This commit is contained in:
Ishaan Jaffer
parent
7d3cad1e18
commit
56dec8d040
1 changed files with 68 additions and 33 deletions
|
|
@ -923,6 +923,20 @@ class PrometheusLogger(CustomLogger):
|
|||
|
||||
return labels
|
||||
|
||||
def _safe_labels(self, metric, **kwargs) -> dict:
|
||||
"""Return only the label kwargs that are registered for the metric.
|
||||
|
||||
When prometheus_exclude_labels strips labels at registration time, observation
|
||||
call sites must supply only the registered subset or prometheus_client raises
|
||||
ValueError (label count mismatch).
|
||||
"""
|
||||
if isinstance(metric, NoOpMetric):
|
||||
return kwargs
|
||||
registered = getattr(metric, "_labelnames", None)
|
||||
if registered is None:
|
||||
return kwargs
|
||||
return {k: v for k, v in kwargs.items() if k in registered}
|
||||
|
||||
async def async_log_success_event(self, kwargs, response_obj, start_time, end_time):
|
||||
# Define prometheus client
|
||||
from litellm.types.utils import StandardLoggingPayload
|
||||
|
|
@ -1326,17 +1340,23 @@ class PrometheusLogger(CustomLogger):
|
|||
)
|
||||
|
||||
self.litellm_remaining_api_key_requests_for_model.labels(
|
||||
_sanitize_prometheus_label_value(user_api_key),
|
||||
_sanitize_prometheus_label_value(user_api_key_alias),
|
||||
_sanitize_prometheus_label_value(model_group),
|
||||
_sanitize_prometheus_label_value(model_id),
|
||||
**self._safe_labels(
|
||||
self.litellm_remaining_api_key_requests_for_model,
|
||||
hashed_api_key=_sanitize_prometheus_label_value(user_api_key),
|
||||
api_key_alias=_sanitize_prometheus_label_value(user_api_key_alias),
|
||||
model=_sanitize_prometheus_label_value(model_group),
|
||||
model_id=_sanitize_prometheus_label_value(model_id),
|
||||
)
|
||||
).set(remaining_requests)
|
||||
|
||||
self.litellm_remaining_api_key_tokens_for_model.labels(
|
||||
_sanitize_prometheus_label_value(user_api_key),
|
||||
_sanitize_prometheus_label_value(user_api_key_alias),
|
||||
_sanitize_prometheus_label_value(model_group),
|
||||
_sanitize_prometheus_label_value(model_id),
|
||||
**self._safe_labels(
|
||||
self.litellm_remaining_api_key_tokens_for_model,
|
||||
hashed_api_key=_sanitize_prometheus_label_value(user_api_key),
|
||||
api_key_alias=_sanitize_prometheus_label_value(user_api_key_alias),
|
||||
model=_sanitize_prometheus_label_value(model_group),
|
||||
model_id=_sanitize_prometheus_label_value(model_id),
|
||||
)
|
||||
).set(remaining_tokens)
|
||||
|
||||
def _set_latency_metrics(
|
||||
|
|
@ -1457,16 +1477,19 @@ class PrometheusLogger(CustomLogger):
|
|||
|
||||
try:
|
||||
self.litellm_llm_api_failed_requests_metric.labels(
|
||||
_sanitize_prometheus_label_value(end_user_id),
|
||||
_sanitize_prometheus_label_value(user_api_key),
|
||||
_sanitize_prometheus_label_value(user_api_key_alias),
|
||||
_sanitize_prometheus_label_value(model),
|
||||
_sanitize_prometheus_label_value(user_api_team),
|
||||
_sanitize_prometheus_label_value(user_api_team_alias),
|
||||
_sanitize_prometheus_label_value(user_id),
|
||||
_sanitize_prometheus_label_value(
|
||||
standard_logging_payload.get("model_id", "")
|
||||
),
|
||||
**self._safe_labels(
|
||||
self.litellm_llm_api_failed_requests_metric,
|
||||
end_user=_sanitize_prometheus_label_value(end_user_id),
|
||||
hashed_api_key=_sanitize_prometheus_label_value(user_api_key),
|
||||
api_key_alias=_sanitize_prometheus_label_value(user_api_key_alias),
|
||||
model=_sanitize_prometheus_label_value(model),
|
||||
team=_sanitize_prometheus_label_value(user_api_team),
|
||||
team_alias=_sanitize_prometheus_label_value(user_api_team_alias),
|
||||
user=_sanitize_prometheus_label_value(user_id),
|
||||
model_id=_sanitize_prometheus_label_value(
|
||||
standard_logging_payload.get("model_id", "")
|
||||
),
|
||||
)
|
||||
).inc()
|
||||
self.set_llm_deployment_failure_metrics(kwargs)
|
||||
except Exception as e:
|
||||
|
|
@ -2180,25 +2203,34 @@ class PrometheusLogger(CustomLogger):
|
|||
try:
|
||||
# Record latency
|
||||
self.litellm_guardrail_latency_metric.labels(
|
||||
guardrail_name=guardrail_name,
|
||||
status=status,
|
||||
error_type=error_type or "none",
|
||||
hook_type=hook_type,
|
||||
**self._safe_labels(
|
||||
self.litellm_guardrail_latency_metric,
|
||||
guardrail_name=guardrail_name,
|
||||
status=status,
|
||||
error_type=error_type or "none",
|
||||
hook_type=hook_type,
|
||||
)
|
||||
).observe(latency_seconds)
|
||||
|
||||
# Record request count
|
||||
self.litellm_guardrail_requests_total.labels(
|
||||
guardrail_name=guardrail_name,
|
||||
status=status,
|
||||
hook_type=hook_type,
|
||||
**self._safe_labels(
|
||||
self.litellm_guardrail_requests_total,
|
||||
guardrail_name=guardrail_name,
|
||||
status=status,
|
||||
hook_type=hook_type,
|
||||
)
|
||||
).inc()
|
||||
|
||||
# Record error count if there was an error
|
||||
if status == "error" and error_type:
|
||||
self.litellm_guardrail_errors_total.labels(
|
||||
guardrail_name=guardrail_name,
|
||||
error_type=error_type,
|
||||
hook_type=hook_type,
|
||||
**self._safe_labels(
|
||||
self.litellm_guardrail_errors_total,
|
||||
guardrail_name=guardrail_name,
|
||||
error_type=error_type,
|
||||
hook_type=hook_type,
|
||||
)
|
||||
).inc()
|
||||
except Exception as e:
|
||||
verbose_logger.debug(f"Error recording guardrail metrics: {str(e)}")
|
||||
|
|
@ -2382,11 +2414,14 @@ class PrometheusLogger(CustomLogger):
|
|||
increment metric when litellm.Router / load balancing logic places a deployment in cool down
|
||||
"""
|
||||
self.litellm_deployment_cooled_down.labels(
|
||||
_sanitize_prometheus_label_value(litellm_model_name),
|
||||
_sanitize_prometheus_label_value(model_id),
|
||||
_sanitize_prometheus_label_value(api_base),
|
||||
_sanitize_prometheus_label_value(api_provider),
|
||||
_sanitize_prometheus_label_value(exception_status),
|
||||
**self._safe_labels(
|
||||
self.litellm_deployment_cooled_down,
|
||||
litellm_model_name=_sanitize_prometheus_label_value(litellm_model_name),
|
||||
model_id=_sanitize_prometheus_label_value(model_id),
|
||||
api_base=_sanitize_prometheus_label_value(api_base),
|
||||
api_provider=_sanitize_prometheus_label_value(api_provider),
|
||||
exception_status=_sanitize_prometheus_label_value(exception_status),
|
||||
)
|
||||
).inc()
|
||||
|
||||
def increment_callback_logging_failure(
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue