You are viewing a plain text version of this content. The canonical link for it is here.
Posted to bugs@httpd.apache.org by bu...@apache.org on 2003/05/06 00:55:12 UTC

DO NOT REPLY [Bug 19687] New: - idle httpd childs consuming 100% cpu

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=19687>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=19687

idle httpd childs consuming 100% cpu

           Summary: idle httpd childs consuming 100% cpu
           Product: Apache httpd-1.3
           Version: 1.3.27
          Platform: Other
        OS/Version: Linux
            Status: NEW
          Severity: Critical
          Priority: Other
         Component: core
        AssignedTo: bugs@httpd.apache.org
        ReportedBy: fluke@krneki.org


A few of httpd childs hit 99% cpu load and they don't exit ever until I manually
kill them (or run apachectl restart).
By comparing PIDs, server-status reports apache is sending reply, however tcp
connection with an IP (which requested url) is closed long ago.

Srv	PID	Acc		M	CPU	SS	Req	Conn	Child	Slot	Host		VHost		Request
11-14	14140	0/1082/5900	W 	32.23	1810	0	0.0	4.67	22.16 	206.207.111.75
domain.com	GET /images/pix.gif HTTP/1.0
29-14	22392	0/22/4990	W 	1.31	4824	0	0.0	0.07	19.57 	207.148.214.34	domain.com
GET /search.php HTTP/1.0
16-14	30498	0/11/11624	W 	0.36	41025	0	0.0	0.04	44.39 	217.167.40.29	domain.com
GET /css/demo.css HTTP/1.0

This is very annoying since httpd childs starts accumulating, and over a few
hours/days period server hits very high load (40+).

Right now I have "apachectl restart" in my crontab every few hours, but that's
not really the solution I want.
Perhaps there's a way of forcefully killing a child after 30 minutes, or after
it consumes certain amount of cpu time.
I'm using apache 1.3.27 with php 4.3.1.

I tested apache 2.0.45 (+php 4.3.1) and it seems it handles this kind of lockups
somewhat better, most high cpu childs die, but some still stay running.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org