对HTTP请求限制和Signalr感到困惑 [英] confused about http request limits and signalr

问题描述

我们在带有Redis背板的应用程序服务上安装了Signalr，并有成千上万的客户端始终通过signalr连接到网站.

We have a signalr setup on an app service with a redis back plane and have thousands of clients that are always connected to the website via signalr. 

一切似乎都正常，但我想知道我们将首先遇到哪些限制-我找不到增加 appConcurrentRequestLimit的方法-通过控制台运行它告诉我" appConcurrentRequestLimit 找不到"

Everything seems to be working ok but i am wondering what limits we will hit first - I can't find a way to increase the appConcurrentRequestLimit - running this via the console tells me that "appConcurrentRequestLimit could not be found"

appcmd.exe设置配置/section:serverRuntime/appConcurrentRequestLimit:100000

如何增加对应用服务的限制?实际服务仅与Redis背板有1个连接吗?还是每个实例都被传入的每个请求击中，然后它们都击中了背板?

How do I increase that limit on an app service?  Does the actual service just have 1 connection to the redis back plane? Or every instance hit with each request that comes in and then they all hit the back plane? 

通过我找到的Kudu

Through Kudu I found

• 我无法终生发现-1意味着什么-我想这是无限的吗?

  I can't for the life of me find what -1 means - i guess it's unlimited?

  这些资料的文档似乎被炸成碎片，散布在非常广阔的区域.

  The documentation for this stuff seems to have been blown to pieces and scattered over a very wide area.

  推荐答案

  请查看下面的链接，其中包含VM出站连接限制.

  Please see the below link, which has the VM outbound connection limits.

  To avoid socket exhaustion scenarios, you'll need to make sure your app closes the connections or reuses them(i.e. connection pooling, connection persistence) rather than generating a new connection for each call.

  
  

  这篇关于对HTTP请求限制和Signalr感到困惑的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！