apache宕机请教

benkernel

最近公司前端有部分web服务器的apache不定时的宕机，error_log里出现大量的抛错：
注：查看了访问日志很正常，error_log里没有发现memory耗尽的提示，message也没有异常
[warn] child process 14582 still did not exit, sending a SIGTERM
[warn] child process 14085 still did not exit, sending a SIGTERM
[warn] child process 14384 still did not exit, sending a SIGTERM
[warn] child process 14672 still did not exit, sending a SIGTERM
[warn] child process 14401 still did not exit, sending a SIGTERM
[warn] child process 14407 still did not exit, sending a SIGTERM
[warn] child process 14447 still did not exit, sending a SIGTERM
[warn] child process 14629 still did not exit, sending a SIGTERM
[warn] child process 14725 still did not exit, sending a SIGTERM
[warn] child process 14729 still did not exit, sending a SIGTERM
[warn] child process 14582 still did not exit, sending a SIGTERM
[warn] child process 14085 still did not exit, sending a SIGTERM
[warn] child process 14384 still did not exit, sending a SIGTERM
[warn] child process 14401 still did not exit, sending a SIGTERM
[warn] child process 14407 still did not exit, sending a SIGTERM
[warn] child process 14447 still did not exit, sending a SIGTERM
[warn] child process 14629 still did not exit, sending a SIGTERM
[warn] child process 14725 still did not exit, sending a SIGTERM
[warn] child process 14729 still did not exit, sending a SIGTERM
[warn] child process 14582 still did not exit, sending a SIGTERM
[warn] child process 14085 still did not exit, sending a SIGTERM
[warn] child process 14384 still did not exit, sending a SIGTERM
[warn] child process 14401 still did not exit, sending a SIGTERM
[warn] child process 14407 still did not exit, sending a SIGTERM
[warn] child process 14447 still did not exit, sending a SIGTERM
[warn] child process 14629 still did not exit, sending a SIGTERM
[warn] child process 14725 still did not exit, sending a SIGTERM
[warn] child process 14729 still did not exit, sending a SIGTERM
[error] child process 14582 still did not exit, sending a SIGKILL
[error] child process 14085 still did not exit, sending a SIGKILL
[error] child process 14384 still did not exit, sending a SIGKILL
[error] child process 14401 still did not exit, sending a SIGKILL
[error] child process 14407 still did not exit, sending a SIGKILL
[error] child process 14447 still did not exit, sending a SIGKILL
[error] child process 14629 still did not exit, sending a SIGKILL
[error] child process 14725 still did not exit, sending a SIGKILL
[error] child process 14729 still did not exit, sending a SIGKILL
[notice] caught SIGTERM, shutting down
然后apache就down掉了，无法正常访问，服务器上的监测和重启apache的脚本没有发挥作用
只有手动去重启apache才能恢复正常
在cu上search了好多帖子，没有找到产生这种情况的原因和解决方法

alvis

看日志是，你的 apache 的子进程无法正常终止，所以资源耗尽就 down 机了

lushy

apache产生的日志文件有多大？

benkernel

出现down机的时候，服务器负载很大，郁闷的是负载大的时候，服务器的监控重启脚本不能正常生效，在此之前，一直运行很正常的

日志文件没有很大了，错误日志是每天分析然后清空的，access访问日志是同步到日志服务器去分析的

alvis

方便的話，把 httpd.conf 貼過來

xuki

具休环境是怎么样的?使用了哪些模块?

benkernel

as4  apache2.2.4 .php 4.4.7 ,apache使用了expires,mod_evasive，差不多就这样
昨天晚上修改了一下监控脚本，在stop后面加了个kill，然后再start，貌似正常了，今天又出问题了，监控脚本无法stop，也无法kill，ps出现很多僵死的httpd进程。。。。

bend

可能是apache的日志文件达到2G了