Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 1 Aug 2002 20:18:59 +0100 (BST)
From:      William Palfreman <william@palfreman.com>
To:        lists@brenius.com
Cc:        questions@FreeBSD.ORG
Subject:   Re: Very High HTTPD Usage (Pls help, urgent)
Message-ID:  <20020801201301.V8173-100000@bell.lan.palfreman.com>
In-Reply-To: <006101c23978$97a2e8d0$7b01a8c0@afi>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 1 Aug 2002 lists@brenius.com wrote:

> 12:26PM  up 124 days,  6:47, 2 users, load averages: 12.31, 12.66, 12.17
>
> Usually around 0.20- 0.30 load average...
>
> last pid: 53735;  load averages: 13.83, 13.01, 12.33
> up 124+06:48:28 12:27:22
> 94 processes:  16 running, 78 sleeping
> CPU states: 27.2% user,  0.0% nice, 72.4% system,  0.4% interrupt,  0.0% idle
> Mem: 324M Active, 63M Inact, 89M Wired, 23M Cache, 61M Buf, 992K Free
> Swap: 720M Total, 176M Used, 544M Free, 24% Inuse
>
>   PID USERNAME     PRI NICE  SIZE    RES STATE    TIME   WCPU    CPU COMMAND
> 42591 nobody        61   0  5732K  3728K RUN      3:16  7.18%  7.18% httpd
> 47033 nobody        60   0  5720K  3720K RUN      3:15  7.03%  7.03% httpd
> 47750 nobody        60   0  5612K  3600K RUN      3:17  6.98%  6.98% httpd
> 47736 nobody        60   0  5720K  3716K RUN      3:17  6.79%  6.79% httpd
> 47749 nobody        59   0  5720K  3716K RUN      3:16  6.79%  6.79% httpd
> 49201 username        59   0 19204K 13412K RUN      3:23  6.54%  6.54% procmail
> 47755 nobody        59   0  5600K  3572K RUN      3:17  6.54%  6.54% httpd
> 47035 nobody        58   0  5720K  3716K RUN      3:15  6.49%  6.49% httpd
> 49166 nobody        59   0  5600K  3532K RUN      3:16  6.40%  6.40% httpd
> 47563 nobody        58   0  5720K  3724K RUN      3:17  6.10%  6.10% httpd
>
> How can I troubleshoot this and track down the culprit. It has been at
> this for the last 10 minutes and is really slowing down the server.
> Please help!

I'm getting the same thing with my Slash installation every day about 2
o'clock in the morning.  Except for me it is a perl process that is
wreaking havoc, swallowing all the swap then getting killed off by the
OS.

The only workaround I've done so far is a script to check that at least
one perl is running and to restart the service if it dies.  Of course
the sensible think to do would be to take the probelm to
www.slashcode.com, and I may yet do that.

As for you, are you doing anything funny with your webserver?  I used to
et this when I ran mod_jk.  You might have to just restart apache.

Bill.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020801201301.V8173-100000>