最新消息:20181230 VPS服务器已从Linode换到腾讯云香港,主题仍用朋友推荐的大前端D8

【已解决】Flask-Sockets的websocket访问出错:Error during WebSocket handshake: Unexpected response code 404

Flask crifan 704浏览 0评论

用Chrome的ws插件:

Simple WebSocket Client

去测试打开ws端口:

ws://115.29.173.126:21084/users/user-17681082-13ae-4a0c-a8bb-83f5814a1b22/p7Isn6ZJkj6Ot7mXAM2melDGT9imrhI9

但是出错:

WebSocket connection to ‘ws://115.29.173.126:21084/users/user-17681082-13ae-4a0c-a8bb-83f5814a1b22/p7Isn6ZJkj6Ot7mXAM2melDGT9imrhI9’ failed: Error during WebSocket handshake: Unexpected response code: 404

弹出对话框:

Simple WebSocket Client

undefined

log为:

index.js:15 WebSocket connection to ‘ws://115.29.173.126:21084/users/user-17681082-13ae-4a0c-a8bb-83f5814a1b22/p7Isn6ZJkj6Ot7mXAM2melDGT9imrhI9’ failed: Error during WebSocket handshake: Unexpected response code: 404
open @ index.js:15(anonymous function) @ index.js:95handle @ jquery-1.4.3.min.js:63n @ jquery-1.4.3.min.js:56
index.js:55 CLOSED: ws://115.29.173.126:21084/users/user-17681082-13ae-4a0c-a8bb-83f5814a1b22/p7Isn6ZJkj6Ot7mXAM2melDGT9imrhI9

不过访问同一服务器中另外的一个端口不同的(同样的代码的)Flask的app,则是可以正常打开websocket的:

OPENED: ws://115.29.173.126:21085/users/user-17681082-13ae-4a0c-a8bb-83f5814a1b22/IYdRjFnYdVjDkGIgxex1fVcJF7HkQ4tW
index.js:29 CLOSING …
index.js:55 CLOSED: ws://115.29.173.126:21085/users/user-17681082-13ae-4a0c-a8bb-83f5814a1b22/IYdRjFnYdVjDkGIgxex1fVcJF7HkQ4tW

对应的Flask的app是通过gunicorn和supervisor启动的:

supervisorctl -c supervisor.conf restart runningfast_staging

supervisorctl -c supervisor.conf restart runningfast_stable

配置是:

[program:runningfast_stable]
directory=/root/RunningFast/stable
command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21084 -k flask_sockets.worker run:app &
startsecs=0
stopwaitsecs=0
autostart=false
autorestart=false
stdout_logfile_maxbytes=1MB
stderr_logfile_maxbytes=1MB
stdout_logfile=/root/RunningFast/stable/logs/gunicorn.log
stderr_logfile=/root/RunningFast/stable/logs/gunicorn.err
[program:runningfast_staging]
directory=/root/RunningFast/staging
command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21085 -k flask_sockets.worker run:app &
startsecs=0
stopwaitsecs=0
autostart=false
autorestart=false
stdout_logfile_maxbytes=1MB
stderr_logfile_maxbytes=1MB
stdout_logfile=/root/RunningFast/staging/logs/gunicorn.log
stderr_logfile=/root/RunningFast/staging/logs/gunicorn.err

即:

command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21084 -k flask_sockets.worker run:app &

command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21085 -k flask_sockets.worker run:app &

21085端口的ws是好的。

21084的端口的ws是有问题的。

1.难道是因为:

同一个Flask的服务器中,只能运行单个ws?

 或许只能是:

最后一个启动的flask的app的ws可以使用?

那么去试试:

先:

supervisorctl -c supervisor.conf restart runningfast_staging

再:

supervisorctl -c supervisor.conf restart runningfast_stable

看看是不是:

stable的21084可用

staging的21085不可用

(RunningFast) ➜  staging supervisorctl -c supervisor.conf restart runningfast_staging
runningfast_staging: stopped
runningfast_staging: started
(RunningFast) ➜  staging cd ../stable
(RunningFast) ➜  stable supervisorctl -c supervisor.conf restart runningfast_stable  
runningfast_stable: stopped
runningfast_stable: started

结果还是21084不可用。。。

2.搜:

flask-sockets Error during WebSocket handshake Unexpected response code 404

Intermittent Error during WebSocket handshake: Unexpected response code: 400 on CloudBees – Stack Overflow

python – (tornadio2) failed: Error during WebSocket handshake: Unexpected response code: 403 – Stack Overflow

flask-sockets Unexpected response code 404

ws websocket Unexpected response code 404

ws Flask websocket Unexpected response code 404

Error during WebSocket handshake: Unexpected response code: 404 · Issue #165 · jrief/django-websocket-redis

“well, error 404 normally means that the endpoint wasn’t found. “

->404意味着端口没有找到

-》就是不存在,没有配置好

-》再去看看自己的配置,是否可能存在哪些问题

有空去试试:

去停止掉21804和21805这两个Flask的app

只运行21804的去试试:

/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21084 -k flask_sockets.worker run:app &

Flask websocket   code 404

Receiving 404 response code when opening websocket connection · Issue #38 · zeekay/flask-uwsgi-websocket

感觉好像和我这里错误有点类似或者有关系。

但是此处没法解决啊

python – Trouble deploying a Flask Websockets App on Heroku – Stack Overflow

draft-hixie-thewebsocketprotocol-55 – The WebSocket protocol

“The Web Socket protocol

                  draft-hixie-thewebsocketprotocol-55

Abstract

   The Web Sockets protocol enables two-way communication between a user

   agent running untrusted code running in a controlled environment to a

   remote host that has opted-in to communications from that code.  The

   security model used for this is the Origin-based security model

   commonly used by Web browsers.  The protocol consists of an initial

   handshake followed by basic message framing, layered over TCP.  The

   goal of this technology is to provide a mechanism for browser-based

   applications that need two-way communication with servers that does

   not rely on opening multiple HTTP connections (e.g. using

   XMLHttpRequest or <iframe>s and long polling).

想办法去给Socket输出log

看看有没有对应的配置参数。

kennethreitz/flask-sockets: Elegant WebSockets for your Flask apps.

flask-sockets not work

flask-sockets two app

python – How to integrate flask and flask_sockets into a single app running under uwsgi – Stack Overflow

->

WebSocket support — uWSGI 2.0 documentation

python – flask-sockets and ZeroMQ – Stack Overflow

flask-sockets Using worker flask_sockets.worker

Issue with flask-sockets’s concurrent connection served by GUnicorn – Q&A Overflow

->

winkidney/flask-sockets-tornado: webscoket in flask with tornado style : )

->

ulimit – How do I increase the open files limit for a non-root user? – Ask Ubuntu

Gunicorn worker-connections

Settings — Gunicorn 19.6.0 documentation

worker_class

-k STRING, –worker-class STRING

sync

The type of workers to use.

The default class (sync) should handle most “normal” types of workloads. You’ll want to read Design for information on when you might want to choose one of the other worker classes.

A string referring to one of the following bundled classes:

sync

eventlet – Requires eventlet >= 0.9.7

gevent – Requires gevent >= 0.13

tornado – Requires tornado >= 0.2

gthread – Python 2 requires the futures package to be installed

gaiohttp – Requires Python 3.4 and aiohttp >= 0.21.5

Optionally, you can provide your own worker by giving Gunicorn a Python path to a subclass of gunicorn.workers.base.Worker. This alternative syntax will load the gevent class: gunicorn.workers.ggevent.GeventWorker. Alternatively, the syntax can also load the gevent class with egg:gunicorn#gevent.

worker_connections

–worker-connections INT

1000

The maximum number of simultaneous clients.

This setting only affects the Eventlet and Gevent worker types.“

flask sockets 404

Recently Active ‘flask-sockets’ Questions – Stack Overflow

WebSocket Security | Heroku Dev Center

然后去:

重新改了端口配置,从21084改为21086:

directory=/root/RunningFast/stable[program:runningfast_stable]
directory=/root/RunningFast/stable
command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21086 -k flask_sockets.worker run:app &

再去重载配置,再去运行:

(RunningFast) ➜  stable supervisorctl -c supervisor.conf restart runningfast_stable
runningfast_stable: ERROR (not running)
runningfast_stable: started
(RunningFast) ➜  stable supervisorctl -c supervisor.conf restart runningfast_stable
runningfast_stable: stopped
runningfast_stable: started
(RunningFast) ➜  stable supervisorctl -c supervisor.conf status runningfast_stable       
runningfast_stable               RUNNING   pid 24252, uptime 0:00:20

结果就可以打开新的21086的ws端口了:

-》推测之前的配置有误?

还是21084被其他进程,程序占用了?

重新改回来21084端口:

command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21084 -k flask_sockets.worker run:app &

重载配置:

(RunningFast) ➜  stable supervisorctl -c supervisor.conf reload                     
Restarted supervisord
(RunningFast) ➜  stable supervisorctl -c supervisor.conf restart runningfast_stable
runningfast_stable: ERROR (not running)
runningfast_stable: started

再试试

果然就可以了。

【总结】

此处,对于同样的一套Flask的app,

之所以21085的端口的ws可以正常访问,

而21084端口的ws无法访问

原因是:

估计是之前不知道哪次,把21084的配置改出问题了。但是没有及时改回来。

所以有问题。

重新确认配置:

附上完整的配置:

[program:runningfast_stable]
directory=/root/RunningFast/stable
command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21084 -k flask_sockets.worker run:app &
startsecs=0
stopwaitsecs=0
autostart=false
autorestart=false
stdout_logfile_maxbytes=1MB
stderr_logfile_maxbytes=1MB
stdout_logfile=/root/RunningFast/stable/logs/gunicorn.log
stderr_logfile=/root/RunningFast/stable/logs/gunicorn.err
[program:runningfast_staging]
directory=/root/RunningFast/staging
command=/root/Envs/RunningFast/bin/gunicorn -w 4 -b 0.0.0.0:21085 -k flask_sockets.worker run:app &
startsecs=0
stopwaitsecs=0
autostart=false
autorestart=false
stdout_logfile_maxbytes=1MB
stderr_logfile_maxbytes=1MB
stdout_logfile=/root/RunningFast/staging/logs/gunicorn.log
stderr_logfile=/root/RunningFast/staging/logs/gunicorn.err

再去reload再运行:

(RunningFast) ➜  stable supervisorctl -c supervisor.conf reload                     
Restarted supervisord
(RunningFast) ➜  stable supervisorctl -c supervisor.conf restart runningfast_stable
runningfast_stable: ERROR (not running)
runningfast_stable: started
(RunningFast) ➜  stable supervisorctl -c supervisor.conf restart runningfast_stable
runningfast_stable: stopped
runningfast_stable: started

然后21084端口的ws就正常工作了:

转载请注明:在路上 » 【已解决】Flask-Sockets的websocket访问出错:Error during WebSocket handshake: Unexpected response code 404

发表我的评论
取消评论

表情

Hi,您需要填写昵称和邮箱!

  • 昵称 (必填)
  • 邮箱 (必填)
  • 网址
65 queries in 0.129 seconds, using 9.57MB memory