Support disabling logs

[tanguyMichardiere]

First Check

• I added a very descriptive title here.
• I used the GitHub search to find a similar question and didn't find it.
• I searched in the documentation/README.
• I already searched in Google "How to do X" and didn't find any information.
• I already read and followed all the tutorial in the docs/README and didn't find an answer.

Commit to Help

• I commit to help with one of those options 👆

Example Code

`fastapi run`

Description

Hello,

I would like to be able to disable the CLI's logs, mostly the panels when it starts.
The cherry-on-top would be to be able to disable uvicorn's logs too ("Uvicorn running on ...").
Ideally it could be configured with environment variables.

Thank you

Operating System

macOS

Operating System Details

14.4.1 on M3 Max

Python Version

3.12.3

Additional Context

No response

[quswadress]

This would be funny if it weren't so sad. FastAPI positions itself as a high-performance framework, but its nice ‘rich’ output cannot be disabled or redirected to a file (or null) in Windows.

C:\>fastapi --help > lol.txt
Traceback (most recent call last)
│ D:\PycharmProjects\web_bench\.venv\lib\site-packages\rich\_win32_console.py │
│ :403 in write_text                                                          │
│                                                                             │
│   400 │   │   Args:                                                         │
│   401 │   │   │   text (str): The text to write to the console              │
│   402 │   │   """                                                           │
│ > 403 │   │   self.write(text)                                              │
│   404 │   │   self.flush()                                                  │
│   405 │                                                                     │
│   406 │   def write_styled(self, text: str, style: Style) -> None:          │
│                                                                             │
│ ┌──────────────────────────────── locals ─────────────────────────────────┐ │
│ │ self = <rich._win32_console.LegacyWindowsTerm object at                 │ │
│ │        0x00000237A2993A30>                                              │ │
│ │ text = ' command line app. 😎'                                          │ │
│ └─────────────────────────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────────────────────┘
UnicodeEncodeError: 'charmap' codec can't encode character '\U0001f60e' in
position 19: character maps to <undefined>

With one line if False: in rich\_win32_console.py I increased the number of requests per second by 2-3 times:

PS D:\PycharmProjects\web_bench> echo "FastApi in different cmd"
PS D:\web_bench> .\bombardier-windows-amd64.exe -c 4 -d 60s http://127.0.0.1:8000/
Bombarding http://127.0.0.1:8000/ for 1m0s using 4 connection(s)
Statistics        Avg      Stdev        Max
  Reqs/sec      4246.71      81.69    4495.60
  Latency        0.94ms    33.40us     3.38ms
  HTTP codes:
    1xx - 0, 2xx - 254798, 3xx - 0, 4xx - 0, 5xx - 0
    others - 0
  Throughput:   850.15KB/s
PS D:\web_bench> .\bombardier-windows-amd64.exe -c 16 -d 60s http://127.0.0.1:8000/
Bombarding http://127.0.0.1:8000/ for 1m0s using 16 connection(s)
Statistics        Avg      Stdev        Max
  Reqs/sec      8153.03    1055.87    9373.41
  Latency        1.96ms   701.53us    45.47ms
  HTTP codes:
    1xx - 0, 2xx - 489071, 3xx - 0, 4xx - 0, 5xx - 0
    others - 0
  Throughput:     1.59MB/s
PS D:\web_bench> echo "FastApi with my patch and 2>&1>$null"
PS D:\web_bench> .\bombardier-windows-amd64.exe -c 4 -d 60s http://127.0.0.1:8000/
Bombarding http://127.0.0.1:8000/ for 1m0s using 4 connection(s)
Statistics        Avg      Stdev        Max
  Reqs/sec      4110.98     125.68    7404.66
  Latency        0.97ms    39.81us     4.47ms
  HTTP codes:
    1xx - 0, 2xx - 246598, 3xx - 0, 4xx - 0, 5xx - 0
    others - 0
  Throughput:   822.78KB/s
PS D:\web_bench> .\bombardier-windows-amd64.exe -c 16 -d 60s http://127.0.0.1:8000/
Bombarding http://127.0.0.1:8000/ for 1m0s using 16 connection(s)
Statistics        Avg      Stdev        Max
  Reqs/sec     27319.84     609.04   28754.17
  Latency      582.79us    12.20us     4.00ms
  HTTP codes:
    1xx - 0, 2xx - 1639203, 3xx - 0, 4xx - 0, 5xx - 0
    others - 0
  Throughput:     5.34MB/s

[williamweckl]

I am having some trouble with this subject too.

I have an implementation that is based on this gist, and after changing to use fastapi run instead of starting directly the uvicorn I see double logs for requests because I can't disable the ones from fastapi cli.

What I want to do is basically the same that I was doing with unicorn using the lines bellow:

logging.getLogger("uvicorn.access").handlers.clear()
logging.getLogger("uvicorn.access").propagate = False

Does anyone know if this is possible?

[williamweckl]

Looks like setting up the logger changes insite lifespan fixes the issue, at least the one that I was dealing with.

@asynccontextmanager
async def lifespan(_app: FastAPI):
    setup_logging(log_level=LOG_LEVEL)
    yield

app = FastAPI(lifespan=lifespan)