1

我刚刚开始学习 Sanic 框架,因为它的基准测试速度很快。我制作了一个简单的 hello world API,然后将它与 Gunicorn 连接起来。性能相当不错,但是当我将它与 Nginx 结合使用时,它变得非常糟糕。我发现带有 Nginx 的 Gunicorn 进程限制为每个进程的 1% - 4% CPU 资源。如果没有 Nginx,Gunicorn 可以达到每个进程的 10%。我以为是因为 Nginx 配置错误。谁能给我一些建议?

服务器信息:

OS: Ubuntu 18.04    
Python version: 3.7.2    
Sanic version: 18.12.0    
Processor: i3-4130

Sanic + Gunicorn 性能:

wrk -t8 -c1000 -d60s --timeout 2s http://127.0.0.1:8080/
Running 1m test @ http://127.0.0.1:8080/
  8 threads and 1000 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency    29.54ms   15.13ms 175.77ms   71.23%
    Req/Sec     4.32k     1.29k   19.46k    64.77%
  2060010 requests in 1.00m, 249.50MB read
Requests/sec:  34281.64
Transfer/sec:      4.15MB

Sanic + Gunicorn + Nginx 性能:

wrk -t8 -c1000 -d60s --timeout 2s http://127.0.0.1:8081/
Running 1m test @ http://127.0.0.1:8081/
  8 threads and 1000 connections
  Thread Stats   Avg      Stdev     Max   +/- Stdev
    Latency   364.78ms  271.20ms   1.39s    67.53%
    Req/Sec   370.88    251.66     3.52k    87.12%
  177223 requests in 1.00m, 30.42MB read
Requests/sec:   2948.79
Transfer/sec:    518.25KB

Sanic 应用程序:

from sanic import Sanic
from sanic.response import json


app = Sanic()
app.config.ACCESS_LOG = False

@app.route("/")
async def test(request):
    return json({"hello": "world"})

独角兽命令:

gunicorn --bind 127.0.0.1:8080 --workers 8 --threads 4 app:app --worker-class sanic.worker.GunicornWorker --name SanicHelloWorld

全局 Nginx 配置:

worker_processes 8;
worker_rlimit_nofile 400000;
thread_pool sanic_thread_pool threads=32 max_queue=65536;
pid /run/nginx.pid;
include /etc/nginx/modules-enabled/*.conf;

events {
    multi_accept on;
    worker_connections 25000;
    use epoll;
    accept_mutex off;
}

http {
    access_log off;
    sendfile on;
    sendfile_max_chunk 512k;
    tcp_nopush on;
    tcp_nodelay on;
    server_names_hash_bucket_size 64;
    proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
    proxy_set_header X-Forwarded-Proto $scheme;
    proxy_set_header Host $http_host;
    proxy_set_header X-Real-IP $remote_addr;
    proxy_redirect off;

    include /etc/nginx/mime.types;
    default_type application/octet-stream;

    include /etc/nginx/conf.d/*.conf;
    include /etc/nginx/sites-enabled/*;

    upstream sanic-test {
        server 127.0.0.1:8080;
    }
}

Sanic + Gunicorn 的 Nginx 配置:

server {
    listen 8081;
    listen [::]:8081;

    server_name sanic-test.com www.sanic-test.com;

    location / {
        aio threads=sanic_thread_pool;
        proxy_pass http://127.0.0.1:8080;
    }
}
4

1 回答 1

0

这可能是因为http://nginx.org/r/proxy_buffering默认设置为on,例如,当您使用http://nginx.org/r/proxy_pass时。

通常,nginx 应该控制后端的背压,因此,缓冲非常有意义,因为您不希望真正的后端受到Slowloris攻击向量的影响。同样,您应该进行缓存并限制与 nginx 背后的真实后端的连接数量,因此,您将所有内容设置为最大值但未能禁用缓冲的测试在现实世界场景中只是一个不切实际的条件,因此,您在指标中得到的数字非常差。

如果您只是想查看简单地向 HTTP 堆栈添加另一层对性能的影响,您应该proxy_buffering off;在使用proxy_pass. 否则,测试应该更现实:你的真实后端每秒处理的请求不应该超过http://nginx.org/r/proxy_temp_path指定的存储设备的 IO 参数。

于 2019-02-04T23:04:24.550 回答