This project was generated with fastapi-mvc
If You want to go easy way and use provided virtualized environment You'll need to have installed:
- rsync
- Vagrant How to install vagrant
- (Optional) Enabled virtualization in BIOS
Otherwise, for local complete project environment with k8s infrastructure bootstrapping You'll need:
For application:
- Python 3.7 or later installed How to install python
- Poetry How to install poetry
For infrastructure:
- make, gcc, golang
- minikube version 1.22.0 How_to_install_minikube
- helm version 3.0.0 or higher How_to_install_helm
- kubectl version 1.16 up to 1.20.8 How_to_install_kubectl
- Container runtime interface. NOTE! dev-env script uses docker for minikube, for other CRI you'll need to modify this line in dev-env.sh
MINIKUBE_IN_STYLE=0 minikube start --driver=docker 2>/dev/null
First run vagrant up
in project root directory and enter virtualized environment using vagrant ssh
Then run following commands to bootstrap local development cluster exposing fastapi-mvc
application.
$ cd /syncd
$ make dev-env
Note: this process may take a while on first run.
Once development cluster is up and running you should see summary listing application address:
Kubernetes cluster ready
fastapi-mvc available under: http://das_sankhya.192.168.49.2.nip.io/
You can delete dev-env by issuing: minikube delete
Note: above address may be different for your installation.
Note: provided virtualized env doesn't have port forwarding configured which means, that bootstrapped application stack in k8s won't be accessible on Host OS.
Deployed application stack in Kubernetes:
vagrant@ubuntu-focal:/syncd$ make dev-env
...
...
...
Kubernetes cluster ready
FastAPI available under: http://das_sankhya.192.168.49.2.nip.io/
You can delete dev-env by issuing: make clean
vagrant@ubuntu-focal:/syncd$ kubectl get all -n das_sankhya
NAME READY STATUS RESTARTS AGE
pod/das_sankhya-7f4dd8dc7f-p2kr7 1/1 Running 0 55s
pod/rfr-redisfailover-persistent-keep-0 1/1 Running 0 3m39s
pod/rfr-redisfailover-persistent-keep-1 1/1 Running 0 3m39s
pod/rfr-redisfailover-persistent-keep-2 1/1 Running 0 3m39s
pod/rfs-redisfailover-persistent-keep-5d46b5bcf8-2r7th 1/1 Running 0 3m39s
pod/rfs-redisfailover-persistent-keep-5d46b5bcf8-6kqv5 1/1 Running 0 3m39s
pod/rfs-redisfailover-persistent-keep-5d46b5bcf8-sgtvv 1/1 Running 0 3m39s
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
service/das_sankhya ClusterIP 10.110.42.252 <none> 8000/TCP 56s
service/rfs-redisfailover-persistent-keep ClusterIP 10.110.4.24 <none> 26379/TCP 3m39s
NAME READY UP-TO-DATE AVAILABLE AGE
deployment.apps/das_sankhya 1/1 1 1 55s
deployment.apps/rfs-redisfailover-persistent-keep 3/3 3 3 3m39s
NAME DESIRED CURRENT READY AGE
replicaset.apps/das_sankhya-7f4dd8dc7f 1 1 1 55s
replicaset.apps/rfs-redisfailover-persistent-keep-5d46b5bcf8 3 3 3 3m39s
NAME READY AGE
statefulset.apps/rfr-redisfailover-persistent-keep 3/3 3m39s
NAME AGE
redisfailover.databases.spotahome.com/redisfailover-persistent-keep 3m39s
vagrant@ubuntu-focal:/syncd$ curl http://das_sankhya.192.168.49.2.nip.io/api/ready
{"status":"ok"}
With make:
make install
You can customize poetry installation with environment variables
export POETRY_HOME=/custom/poetry/path
export POETRY_CACHE_DIR=/custom/poetry/path/cache
export POETRY_VIRTUALENVS_IN_PROJECT=true
make install
Or using poetry directly:
poetry install
To bootstrap local minikube Kubernetes cluster exposing das_sankhya
application:
make dev-env
This package exposes simple CLI for easier interaction:
$ das_sankhya --help
Usage: das_sankhya [OPTIONS] COMMAND [ARGS]...
das_sankhya CLI root.
Options:
-v, --verbose Enable verbose logging.
--help Show this message and exit.
Commands:
serve das_sankhya CLI serve command.
$ das_sankhya serve --help
Usage: das_sankhya serve [OPTIONS]
das_sankhya CLI serve command.
Options:
--host TEXT Host to bind. [default: localhost]
-p, --port INTEGER Port to bind. [default: 8000]
-w, --workers INTEGER RANGE The number of worker processes for handling
requests. [default: 2;1<=x<=8]
--help Show this message and exit.
NOTE: Maximum number of workers may be different in your case, it's limited to multiprocessing.cpu_count()
To serve application simply run:
$ das_sankhya serve
To confirm it's working:
$ curl localhost:8000/api/ready
{"status":"ok"}
This repository provides Dockerfile for virtualized environment.
NOTE: Replace podman with docker if it's yours containerization engine.
$ make image
$ podman run -dit --name das_sankhya -p 8000:8000 das_sankhya:$(cat TAG)
f41e5fa7ffd512aea8f1aad1c12157bf1e66f961aeb707f51993e9ac343f7a4b
$ podman ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
f41e5fa7ffd5 localhost/das_sankhya:0.1.0 /usr/bin/fastapi ... 2 seconds ago Up 3 seconds ago 0.0.0.0:8000->8000/tcp das_sankhya
$ curl localhost:8000/api/ready
{"status":"ok"}
This application provides flexibility of configuration. All significant settings are defined by the environment variables, each with the default value. Moreover, package CLI allows overriding core ones: host, port, workers. You can modify all other available configuration settings in the gunicorn.conf.py file.
Priority of overriding configuration:
- cli
- environment variables
- gunicorn.conf.py
All application configuration is available in das_sankhya.config
submodule.
Key | Default | Description |
---|---|---|
FASTAPI_HOST | "127.0.0.1" |
FastAPI host to bind. |
FASTAPI_PORT | "8000" |
FastAPI port to bind. |
FASTAPI_WORKERS | "2" |
Number of gunicorn workers (uvicorn.workers.UvicornWorker) |
FASTAPI_DEBUG | "True" |
FastAPI logging level. You should disable this for production. |
FASTAPI_PROJECT_NAME | "das_sankhya" |
FastAPI project name. |
FASTAPI_VERSION | "0.4.0" |
Application version. |
FASTAPI_DOCS_URL | "/" |
Path where swagger ui will be served at. |
FASTAPI_USE_REDIS | "False" |
Whether or not to use Redis. |
FASTAPI_GUNICORN_LOG_LEVEL | "info" |
The granularity of gunicorn log output |
FASTAPI_GUNICORN_LOG_FORMAT | '%(h)s %(l)s %(u)s %(t)s "%(r)s" %(s)s %(b)s "%(f)s" "%(a)s"' |
Gunicorn log format |
Key | Default | Description |
---|---|---|
FASTAPI_REDIS_HOTS | "127.0.0.1" |
Redis host. |
FASTAPI_REDIS_PORT | "6379" |
Redis port. |
FASTAPI_REDIS_USERNAME | "" |
Redis username. |
FASTAPI_REDIS_PASSWORD | "" |
Redis password. |
FASTAPI_REDIS_USE_SENTINEL | "False" |
If provided Redis config is for Sentinel. |
- Source:
das_sankhya.config/gunicorn.conf.py
- Gunicorn configuration file documentation
Endpoints are defined in das_sankhya.config.router
. Just simply import your controller and include it to FastAPI router:
from fastapi import APIRouter
from das_sankhya.app.controllers.api.v1 import ready
router = APIRouter(
prefix="/api"
)
router.include_router(ready.router, tags=["ready"])
You can implement your own web routes logic straight away in das_sankhya.app.controllers.api.v1
submodule. For more information please see FastAPI documentation.
For your discretion, I've provided some basic utilities:
- RedisClient
das_sankhya.app.utils.redis
- AiohttpClient
das_sankhya.app.utils.aiohttp_client
They're initialized in asgi.py
on FastAPI startup event handler:
async def on_startup():
"""Fastapi startup event handler.
Creates RedisClient and AiohttpClient session.
"""
log.debug("Execute FastAPI startup event handler.")
# Initialize utilities for whole FastAPI application without passing object
# instances within the logic. Feel free to disable it if you don't need it.
if settings.USE_REDIS:
await RedisClient.open_redis_client()
AiohttpClient.get_aiohttp_client()
async def on_shutdown():
"""Fastapi shutdown event handler.
Destroys RedisClient and AiohttpClient session.
"""
log.debug("Execute FastAPI shutdown event handler.")
# Gracefully close utilities.
if settings.USE_REDIS:
await RedisClient.close_redis_client()
await AiohttpClient.close_aiohttp_client()
and are available for whole application scope without passing object instances. In order to utilize it just execute classmethods directly.
Example:
from das_sankhya.app.utils import RedisClient
response = RedisClient.get("Key")
from das_sankhya.app.utils import AiohttpClient
response = AiohttpClient.get("http://foo.bar")
Source: das_sankhya.app.exceptions.http.py
This exception combined with http_exception_handler
method allows you to use it the same manner as you'd use FastAPI.HTTPException
with one difference.
You have freedom to define returned response body, whereas in FastAPI.HTTPException
content is returned under "detail" JSON key.
In this application custom handler is added in asgi.py
while initializing FastAPI application.
This is needed in order to handle it globally.
All routes documentation is available on:
/
with Swagger/redoc
or ReDoc.
This project is licensed under the terms of the MIT license.
With the help of some information obtained on the internet I was able to customize the fastapi-mvc-example to do the following:
-
Use loguru instead of the standard Python logging module to manage system logs. Apparently loguru works better with threads and multiprocessing.
-
Capture all logs, even Gunicorn startup logs with the custom log format I intend to use in my application.
-
Put the following information in the log lines: PID of the process (gunicorn worker) that generated the log, x-correlation-id and x-request-id (a necessity of my application and which are generated/obtained by a fastapi middleware )
-
Gunicorn's master and workers appear with a friendly name when I run a ps axu
By default I generate the logs in stdout, but I also tested saving the logs in a file and it worked perfectly, even after rotating the file. Each worker continued writing to the same file after loguru's file rotation.
My version of wsgi.py is hardcoded to configure Gunicorn as per this post: https://medium.com/building-the-system/gunicorn-3-means-of-concurrency-efbb547674b7 - I don't know if it's the most efficient way , but in my tests it worked fine.
Although not complete, I added a new command to start a uvicorn server only, this is because I like to use its reload feature when I'm developing my code and I couldn't find a way to do this with Gunicorn - apparently it's impossible. The command is:
das_sankhya devserve
I used this information to prepare this demo:
- https://github.com/pahntanapat/Unified-FastAPI-Gunicorn-Log
- Delgan/loguru#365
- https://loguru.readthedocs.io/en/stable/api/logger.html#sink
And the main changes I made are in these files: