FastApi in Docker shutting down after some requests

First Check

• I added a very descriptive title to this issue.
• I used the GitHub search to find a similar issue and didn’t find it.
• I searched the FastAPI documentation, with the integrated search.
• I already searched in Google “How to X in FastAPI” and didn’t find any information.
• I already read and followed all the tutorial in the docs and didn’t find an answer.
• I already checked if it is not related to FastAPI but to Pydantic.
• I already checked if it is not related to FastAPI but to Swagger UI.
• I already checked if it is not related to FastAPI but to ReDoc.

Commit to Help

• I commit to help with one of those options 👆

Example Code

# Dockerfile
ENTRYPOINT [ "poetry", "run", "uvicorn", "main:app" ]
CMD [ "--host", "0.0.0.0", "--app-dir", "/app/sales_api/", "--log-config", "/logging.conf", "--debug", "true" ]

# logs 

[INFO][2022-09-1416:04:01,091] [Sales API] [sales _api] L36 : Creating the FastAPI application
1663171441105 [INFO][2022-09-1416:04:01,105] [Sales API] [sales_ api] L45 : Attaching API routes to the main application
1663171441106 [INFO][2022-09-1416:04:01,105] [Sales API] [uvicorn.error] L75 : Started server process [15]
1663171441106 (INFO][2022-09-1416:04:01,106] [Sales API] [uvicorn.error] L47 : Waiting for application startup.
1663171441106 (INFO][2022-09-1416:04:01,106] [Sales API] [uvicorn.error] L61 : Application startup complete.
1663171455907 [INFO1[2022-09-1416:04:15,9071 [Sales API1 [uvicorn.access]L474:172.31.22.229:27194-"GET/api/v1/healthcheckHTTP/1.1"307
1663171455910 [INF0][2022-09-1416:04:15,9091 [Sales API] (uvicorn.access]L474:172.31.2.217:42564-"GET/api/v1/healthcheckHTTP/1.1"307
1663171455910 [INF0][2022-09-1416:04:15,9091 [Sales API] [uvicorn.access] L474 : 172.31.33.74:16698 - "GET/api/v1/healthcheck HTTP/1.1" 307
1663171479324 [INFO1[2022-09-1416:04:39.3241 [Sales API1 [uvicorn.access]L474:172.31.33.74:42612-"GET/docsHTTP/1.1"200
1663171479379 [INFO1(2022-09-1416:04:39,3781 [Sales API] (uvicorn.access]L474:172.31.33.74:42612-"GET/api/v1/openapi.jsonHTTP/1.1"200
1663171485937 [INFO][2022-09-1416:04:45,937] [Sales API] [uvicorn.access] L474 : 172.31.22.229:5556 - "GET /api/v1/healthcheck HTTP/1.1" 307
1663171485939 [INFO][2022-09-1416:04:45,939] [Sales API] [uvicorn.access] L474 : 172.31.2.217:34220 - "GET/api/v1/healthcheck HTTP/1.1" 307
1663171485940 [INFO1(2022-09-1416:04:45.9401 [Sales API1 [uvicorn.access1 L474 : 172.31.33.74:10670 - "GET/api/v1/healthcheck HTTP/1.1" 307
1663171501825 [INFO1(2022-09-1416:05:01.8251 [Sales API1 [uvicorn.error] L253: Shuttina down
1663171501925 [INFO1(2022-09-1416:05:01,9251 [Sales API] [uvicorn.error] L66: Waiting for application shutdown.
1663171501925 (INFO1(2022-09-1416:05:01,9251 [Sales API] [uvicorn.error] L77 : Application shutdown complete.
1663171501926 [INFO1[2022-09-1416:05:01,9251 [Sales API] [uvicorn.error] L85 : Finished server process [14]
1663171502985 [INFO1[2022-09-1416:05:02,9851 [Sales API1 [uvicorn.error] L253: Shutting down
1663171503085 [INFO1(2022-09-14 16:05:03.0851 (Sales API] [uvicorn.error] L66 : Waiting for application shutdown.
1663171503086 [INFO1(2022-09-14 16:05:03.0851 Sales API1 [uvicorn.error] L77: Application shutdown complete.
1663171503086 [INFO1(2022-09-14 16:05:03.0861 [Sales API1 [uvicorn.error1 L85 : Finished server process [151]

# Package versions
Poetry == 1.2.0
FastAPI == 0.18.0
uvicorn == 0.18.3
gunicorn == 20.1.0

Description

App working correctly locally on macOS on :

• Normal start.
• Docker start.

When deploying under AWS app keep on shutting down on some requests as you can see it in the given logs.

Note that I got correct responses on all routes within the 10-15 first seconds and then a shutdown is forced.

Operating System

Linux

Operating System Details

Under AWS, will give some further details if required. (I do have another FastAPI within the same context and it is working properly)

FastAPI Version

0.80.0

Python Version

3.8.6

Additional Context

No response

[JarroVGIT]

It seems the AWS service is calling the healthcheck endpoint but is redirected every time (that is the 307). I would guess the AWS service is then stopping your application.

[kaieslamiri]

👌