limits.conf doesn't appear to have any effect
- To: pam-list@xxxxxxxxxx
- Subject: limits.conf doesn't appear to have any effect
- From: hbeaumont hbeaumont <ahlist@xxxxxxxxx>
- Date: Thu, 29 Dec 2005 11:59:18 -0500
- Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:mime-version:content-type; b=kdJmeB1bqqDQI7WWhWfE2GQareHNScsj8DA7Eio/zWEksLvoBvPzHPkw2jczEse/ByeCO4n6y4aNB0DxinbylcpE6tmpKsbYKqTnAFywZ99yFJni9R8wJrg0eqajk0zZOHYnnqpsAy89HYzJqt3WP1/yURkgT0QUm3JCZA8qbvg=
- Reply-to: Pluggable Authentication Modules <pam-list@xxxxxxxxxx>
- Sender: pam-list-bounces@xxxxxxxxxx
Hi,
On both redhat 7.3 and EL3 machines my entries of:
nobody hard cpu 1
do not seem to have effect.
I am trying to limit bad apache and apache-spawned processes (all ran as user nobody).
My only guess is that perhaps since apache starts out as root it is getting roots limits? I doubt this but it is all I can think of.
These are very stock installs of 7.3 and EL3 - is there anything I need to enable to make PAM limits effective? Do I need to reboot tthe machine?
After I make the changes I do stop and restart apache.
Also, until I can solve this, is anyone aware of a good script to kill off processes uses too much CPU time? Or a perl module that I can use to write something
to grab CPU times and do the killing in my own script?
Thanks!
P.S. I searched the archives but could not find anything. One recent message said to search the archives for a solution (another user posted a similiar question to mine)
but I could not find anything. Thanks again.
_______________________________________________
Pam-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/pam-list
[Home]
[Kernel List]
[Red Hat Install]
[Linux for the blind]
[Red Hat Watch List]
[Gimp]
[Kerberos: The Definitive Guide]
| &nsp; |
 |