phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

fix: respect log_level in uvicorn and third party libs (#1524)

Browse source 
# What does this PR do?

uvicorn has a `log_level` arg in uvicorn.run, pass in the effective
level set by the logger.

Additionally, third party libraries like httpx are using our logging
format, but not honoring our log level.

This seems unintended, so loop through all items in the loggerDict and
apply the same log level as what we have set.


## Test Plan

before:

```
llama stack run --image-type venv ~/.llama/distributions/ollama/ollama-run.yaml
Environment variable LLAMA_STACK_LOGGING found: all=warn
Using virtual environment: /Users/charliedoern/projects/Documents/llama-stack/venv
+ python -m llama_stack.distribution.server.server --yaml-config /Users/charliedoern/.llama/distributions/ollama/ollama-run.yaml --port 8321
Environment variable LLAMA_STACK_LOGGING found: all=warn
WARNING  2025-03-10 16:05:49,706 root:71 uncategorized: Warning: `bwrap` is not available. Code interpreter tool will
         not work correctly.
INFO     2025-03-10 16:05:49,916 datasets:54 uncategorized: PyTorch version 2.5.1 available.
INFO     2025-03-10 16:05:50,010 httpx:1740 uncategorized: HTTP Request: GET http://localhost:11434/api/ps "HTTP/1.1 200
         OK"
INFO     2025-03-10 16:05:50,297 httpx:1740 uncategorized: HTTP Request: POST http://localhost:11434/api/pull "HTTP/1.1
         200 OK"
INFO     2025-03-10 16:05:50,314 httpx:1740 uncategorized: HTTP Request: GET http://localhost:11434/api/tags "HTTP/1.1
         200 OK"
INFO:     Started server process [89663]
INFO:     Waiting for application startup.
INFO:     ASGI 'lifespan' protocol appears unsupported.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://['::', '0.0.0.0']:8321 (Press CTRL+C to quit)
```

after:

```
llama stack run --image-type venv ~/.llama/distributions/ollama/ollama-run.yaml
Environment variable LLAMA_STACK_LOGGING found: all=warn
Using virtual environment: /Users/charliedoern/projects/Documents/llama-stack/venv
+ python -m llama_stack.distribution.server.server --yaml-config /Users/charliedoern/.llama/distributions/ollama/ollama-run.yaml --port 8321
Environment variable LLAMA_STACK_LOGGING found: all=warn
WARNING  2025-03-10 16:05:20,429 root:71 uncategorized: Warning: `bwrap` is not available. Code interpreter tool will
         not work correctly.
INFO     2025-03-10 16:05:20,639 datasets:54 uncategorized: PyTorch version 2.5.1 available.
```

Signed-off-by: Charlie Doern <cdoern@redhat.com>
This commit is contained in:
Charlie Doern 2025-03-12 14:07:28 -04:00 • committed by GitHub
parent 00da911167
commit 4eee349acd
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 6 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

5
llama_stack/log.py
View file

@ -170,6 +170,11 @@ def setup_logging(category_levels: Dict[str, int], log_file: str | None) -> None
}
dictConfig(logging_config)
# Ensure third-party libraries follow the root log level
for _, logger in logging.root.manager.loggerDict.items():
if isinstance(logger, logging.Logger):
logger.setLevel(root_level)
def get_logger(name: str, category: str = "uncategorized") -> logging.LoggerAdapter:
"""