linux – Apache性能问题,在“1000 total children”之后,Apache不再响应HTTP请求.不是MaxClients问题？

在CentOS 5.8上运行Apache 2.2.3 mod_PHP.

在流量很大的每一天,我们都遇到了Apache不再响应任何http请求的问题.

它听起来像是一个标准的MaxClients问题,但似乎并非如此.

此外,在此期间登录到机器的负载平均值低于1,并且仍然有足够的RAM可用.

查看/ var / log / httpd / error_log我注意到以下模式：

[Mon Apr 30 07:00:34 2012] [info] server seems busy,(you may need to increaseStartServers,or Min/MaxSpareServers),spawning 32 children,there are 0 idle,and 905 @R_755_10586@l children
[Mon Apr 30 07:00:35 2012] [info] server seems busy,(you may need to increase StartServers,and 937 @R_755_10586@l children
[Mon Apr 30 07:00:36 2012] [info] server seems busy,and 969 @R_755_10586@l children
[Mon Apr 30 07:00:37 2012] [info] server seems busy,there are 35 idle,and 1001 @R_755_10586@l children

[Mon Apr 30 07:00:42 2012] [debug] mpm_common.c(663): (70007)The timeout specified has expired: connect to listener on [::]:80 <br>
[Mon Apr 30 07:00:49 2012] [debug] mpm_common.c(663): (70007)The timeout specified has expired: connect to listener on [::]:80 <br>
[Mon Apr 30 07:00:56 2012] [debug] mpm_common.c(663): (70007)The timeout specified has expired: connect to listener on [::]:80 <br>
[Mon Apr 30 07:01:03 2012] [debug] mpm_common.c(663): (70007)The timeout specified has expired: connect to listener on [::]:80 <br>

每天几次,在1000名儿童之后,Apache停止响应并且必须重新启动以便再次工作.

我已经回顾了几周前的error_log,它是相同的模式,服务器击中了1000个孩子,然后立即吐出
[debug] mpm_common.c(663):( 70007)指定的超时已到期：连接到[::]：80错误消息上的监听器并停止响应.

然而,服务器上的负载非常低……
即使我尝试请求一个简单的index.html文件,它也会超时.

以下是配置中的相关部分：

Timeout 45
KeepAlive On
MaxKeepAliverequests 10000
KeepAliveTimeout 3

<IfModule prefork.c>
StartServers      80
MinSpareServers   50
MaxSpareServers  120
ServerLimit     3500
MaxClients      3500
MaxrequestsPerChild  2000
</IfModule>

任何人都知道为什么在停止处理更多请求之前Apache能够达到的神奇数量为1000？

或者如何理解(70007)指定的超时已过期：连接到[::]：80消息上的监听器？

指的是什么“超时指定”？

我仔细检查了Max Open Files,它之前是1024,但现在是16384,仍然是同样的问题.