apache與其他webserver結合時，無法啟動

今天早上apache無法訪問，查看日誌發現

[Thu Feb 21 09:47:48 2008] [error] (111)Connection refused: proxy: HTTP: attempt to connect to 127.0.0.1:3000 (localhost) failed
[Thu Feb 21 09:47:48 2008] [error] ap_proxy_connect_backend disabling worker for (localhost)
[Thu Feb 21 09:47:48 2008] [error] proxy: HTTP: disabled connection for (localhost)

3000連接埠啟動並執行是ruby的radiant cms系統，所以便懷疑radiant啟動失敗，便手工啟動該服務
[root@localhost ~]# service mongrel_cluster start
Starting all mongrel_clusters…
** !!! PID file tmp/pids/mongrel.3000.pid already exists.  Mongrel could be running already.  Check your log/mongrel.3000.log for errors.
** !!! Exiting with error.  You must stop mongrel and clear the .pid before I’ll attempt a start.
** !!! PID file tmp/pids/mongrel.4000.pid already exists.  Mongrel could be running already.  Check your log/mongrel.4000.log for errors.
** !!! Exiting with error.  You must stop mongrel and clear the .pid before I’ll attempt a start.
[root@localhost ~]# service mongrel_cluster start
Starting all mongrel_clusters…
** !!! PID file tmp/pids/mongrel.4000.pid already exists.  Mongrel could be running already.  Check your log/mongrel.4000.log for errors.
** !!! Exiting with error.  You must stop mongrel and clear the .pid before I’ll attempt a start.

看意思應該是一些臨時檔案存在，所以無法啟動，到該目錄下刪除，重新啟動，果然可以。

發現apache如果在httpd中轉向到其他web server，如果那個web server啟動失敗，則同時會導致apache無法啟動，

所以如果在linux下使用service自動啟動各種服務，需要注意服務的啟動順序。