太多CLOSE_WAIT在码头9.0.3 [英] Too many Close_wait in jetty 9.0.3

问题描述

我们已经部署在码头9.0.3 server.For验证和部署在码头服务授权的应用程序REST网关是Apache的commons-httpclient3.1照顾。
当有并发线程击中休息网关，很多CLOSE_WAIT的数量庞大的在REST网关机器堆积如山。
这是打开发送请求为enabler（终点）的HTTP连接是通过调用阿帕奇公地releaseConnections方法正确关闭。
请让我知道如何处理这些CLOSE_WAIT为什么连接在CLOSE_WAIT结束了任何线索。

We have an application REST gateway deployed over jetty 9.0.3 server.For authentication and authorization of services deployed in jetty is taken care by apache commons-httpclient3.1. when there are huge number of concurrent threads hitting Rest gateway, lot of CLOSE_WAIT are piling up in REST gateway machine. Http connections which are opened for sending request to enabler(END POINT)are closed properly by calling releaseConnections method of apache commons. Please let me know how to handle these CLOSE_WAIT or any clues why connections are ending up in CLOSE_WAIT.

推荐答案

该CLOSE_WAIT连接时被Apache公地HTTP客户端3.1提供closeIdleConnections（0）消失了。但现在太多的TIME_WAIT越来越建立和后端我收到类似地址已在使用错误：（

The close_wait connections vanished when closeIdleConnections(0) provided by Apache commons http Client 3.1. But now too many TIME_WAIT are getting established and back end i am getting errors like "Address already in use" :(

最后CLOSE_WAIT连接没有使用closeIdleConnections后可见（0）。希望这个解决方案可以帮助别人：）

Finally CLOSE_WAIT connections were not seen after using closeIdleConnections(0). Hope this solution might help others :)

这篇关于太多CLOSE_WAIT在码头9.0.3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！