Support JSON format request logging for easier parsing (#15743)

This commit is contained in:
fzyzcjy
2025-12-24 19:52:43 +08:00
committed by GitHub
parent c6a6ba4368
commit e245cac0c5
5 changed files with 140 additions and 13 deletions

View File

@@ -14,7 +14,9 @@
from __future__ import annotations
import dataclasses
import json
import logging
from datetime import datetime
from functools import lru_cache
from typing import TYPE_CHECKING, Any, Optional, Set, Tuple, Union
@@ -25,11 +27,23 @@ if TYPE_CHECKING:
logger = logging.getLogger(__name__)
_json_logger = logging.getLogger(__name__ + ".json")
_json_logger.propagate = False
_json_handler = logging.StreamHandler()
_json_handler.setFormatter(logging.Formatter("%(message)s"))
_json_logger.addHandler(_json_handler)
class RequestLogger:
def __init__(self, log_requests: bool, log_requests_level: int):
def __init__(
self,
log_requests: bool,
log_requests_level: int,
log_requests_format: str,
):
self.log_requests = log_requests
self.log_requests_level = log_requests_level
self.log_requests_format = log_requests_format
self.metadata: Tuple[Optional[int], Optional[Set[str]], Optional[Set[str]]] = (
self._compute_metadata()
)
@@ -38,11 +52,14 @@ class RequestLogger:
self,
log_requests: Optional[bool] = None,
log_requests_level: Optional[int] = None,
log_requests_format: Optional[str] = None,
) -> None:
if log_requests is not None:
self.log_requests = log_requests
if log_requests_level is not None:
self.log_requests_level = log_requests_level
if log_requests_format is not None:
self.log_requests_format = log_requests_format
self.metadata = self._compute_metadata()
def log_received_request(
@@ -52,9 +69,16 @@ class RequestLogger:
return
max_length, skip_names, _ = self.metadata
logger.info(
f"Receive: obj={_dataclass_to_string_truncated(obj, max_length, skip_names=skip_names)}"
)
if self.log_requests_format == "json":
log_data = {
"rid": obj.rid,
"obj": _transform_data_for_logging(obj, max_length, skip_names),
}
_log_json("request.received", log_data)
else:
logger.info(
f"Receive: obj={_dataclass_to_string_truncated(obj, max_length, skip_names=skip_names)}"
)
# FIXME: This is a temporary fix to get the text from the input ids.
# We should remove this once we have a proper way.
@@ -77,11 +101,22 @@ class RequestLogger:
return
max_length, skip_names, out_skip_names = self.metadata
if is_multimodal_gen:
msg = f"Finish: obj={_dataclass_to_string_truncated(obj, max_length, skip_names=skip_names)}"
if self.log_requests_format == "json":
log_data = {
"rid": obj.rid,
"obj": _transform_data_for_logging(obj, max_length, skip_names),
}
if not is_multimodal_gen:
log_data["out"] = _transform_data_for_logging(
out, max_length, out_skip_names
)
_log_json("request.finished", log_data)
else:
msg = f"Finish: obj={_dataclass_to_string_truncated(obj, max_length, skip_names=skip_names)}, out={_dataclass_to_string_truncated(out, max_length, skip_names=out_skip_names)}"
logger.info(msg)
if is_multimodal_gen:
msg = f"Finish: obj={_dataclass_to_string_truncated(obj, max_length, skip_names=skip_names)}"
else:
msg = f"Finish: obj={_dataclass_to_string_truncated(obj, max_length, skip_names=skip_names)}, out={_dataclass_to_string_truncated(out, max_length, skip_names=out_skip_names)}"
logger.info(msg)
def _compute_metadata(
self,
@@ -130,6 +165,17 @@ def disable_request_logging() -> bool:
return get_bool_env_var("SGLANG_DISABLE_REQUEST_LOGGING")
# TODO unify logging, e.g. allow normal logs to be JSON as well
def _log_json(event: str, data: dict) -> None:
log_data = {
"timestamp": datetime.now().isoformat(),
"event": event,
**data,
}
_json_logger.info(json.dumps(log_data, ensure_ascii=False))
# TODO unify this w/ `_transform_data_for_logging` if we find performance enough
def _dataclass_to_string_truncated(
data: Any, max_length: int = 2048, skip_names: Optional[Set[str]] = None
) -> str:
@@ -170,3 +216,37 @@ def _dataclass_to_string_truncated(
)
else:
return str(data)
def _transform_data_for_logging(
data: Any, max_length: int = 2048, skip_names: Optional[Set[str]] = None
) -> Any:
if skip_names is None:
skip_names = set()
if isinstance(data, str):
if len(data) > max_length:
half_length = max_length // 2
return data[:half_length] + "..." + data[-half_length:]
return data
elif isinstance(data, (list, tuple)):
if len(data) > max_length:
half_length = max_length // 2
return list(data[:half_length]) + ["..."] + list(data[-half_length:])
return [_transform_data_for_logging(v, max_length) for v in data]
elif isinstance(data, dict):
return {
k: _transform_data_for_logging(v, max_length)
for k, v in data.items()
if k not in skip_names
}
elif dataclasses.is_dataclass(data):
fields = dataclasses.fields(data)
return {
f.name: _transform_data_for_logging(getattr(data, f.name), max_length)
for f in fields
if f.name not in skip_names
}
elif isinstance(data, (int, float, bool, type(None))):
return data
else:
return str(data)