Question

一段时间以来，我一直在尝试解决生产站点上的500和502错误的网关问题。我查看了nginx日志文件，发现那里有错误，但似乎找不到任何帮助。这是nginx的错误日志：

2019/04/28 20:52:10 [crit] 921#921: *1 connect() to unix:/pathto/django/project/projectname.sock failed (2: No such file or directory) while connecting to upstream, client: x.x.x.x, server: example.com, request: "GET / HTTP/1.1", upstream: "http://unix:/pathto/django/project/projectname.sock:/", host: "example.com"
2019/04/28 20:53:55 [error] 921#921: *17 upstream prematurely closed connection while reading response header from upstream, client: x.x.x.x, server: example.com, request: "POST /games/ HTTP/1.1", upstream: "http://unix:/pathto/django/project/projectname.sock:/x/", host: "example.com", referrer: "http://example.com/x/"
2019/04/28 20:55:57 [error] 921#921: *17 upstream prematurely closed connection while reading response header from upstream, client: x.x.x.x, server: example.com, request: "POST /games/ HTTP/1.1", upstream: "http://unix:/pathto/django/project/projectname.sock:/x/", host: "example.com", referrer: "http://example.com/x/"

我尝试增加Nginx和Gunicorn的超时时间，但是没有运气。由于多个API请求，我网站上的某些操作需要几分钟才能完成，我相信可能是时候了，但是我不确定。

我的网站可用的nginx文件：

server {
    listen 80;
    server_name example.com;

    location = /favicon.ico {
        return 204;
        access_log off;
        log_not_found off;
    }

    location /static/ {
        root /path/to/static/file;
    }

    location / {
        include proxy_params;
        proxy_pass 
        http://unix:/pathto/django/project/projectname.sock;
    }
}

我的gunicorn起始文件：


#!/bin/bash

NAME="projectname"
DIR=/path/to/project/root
USER=u
GROUP=u
WORKERS=3
BIND=unix:/home/u/run/gunicorn.sock
DJANGO_SETTINGS_MODULE=projectname.settings
DJANGO_WSGI_MODULE=projectname.wsgi
LOG_LEVEL=error

cd $DIR
source ../bin/activate

export DJANGO_SETTINGS_MODULE=$DJANGO_SETTINGS_MODULE
export PYTHONPATH=$DIR:$PYTHONPATH

exec ../bin/gunicorn ${DJANGO_WSGI_MODULE}:application \
  --name $NAME \
  --workers $WORKERS \
  --user=$USER \
  --group=$GROUP \
  --bind=$BIND \
  --log-level=$LOG_LEVEL \
  --log-file=-

如上所述，我的网站开始执行api操作时遇到了500多个错误和502个错误。这些操作涉及（在某些情况下）获取大量api数据并将其存储到我的项目模型中。在许多情况下，它将在操作进行约30秒后502。很多次，我第二次重试后，它可以按预期正常工作。此刻我很困，不胜感激，谢谢！

此外，如果有帮助，我使用了本指南来部署该项目：https://www.digitalocean.com/community/tutorials/how-to-set-up-django-with-postgres-nginx-and-gunicorn-on-ubuntu-16-04

在连接到nginx错误日志中的上游时，归档了djangoproject.sock文件（2：没有这样的文件或目录）

0 个答案: