Bug 166285 - When I change "stack size" on 65536 using ulimit or files /etc/security/limits.conf aplications like cpanel there starts making problems, when adding subdomains using cpanel I get errors
Summary: When I change "stack size" on 65536 using ulimit or files /etc/security/limit...
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: glibc
Version: 4.0
Hardware: All
OS: Linux
medium
high
Target Milestone: ---
: ---
Assignee: Jakub Jelinek
QA Contact: Brian Brock
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2005-08-18 19:09 UTC by Michal
Modified: 2007-11-30 22:07 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2005-08-18 19:24:58 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Description Michal 2005-08-18 19:09:00 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; pl-PL; rv:1.7.10) Gecko/20050717 Firefox/1.0.6

Description of problem:
When I change "stack size" on 65536 using ulimit or files /etc/security/limits.conf aplications like cpanel there starts making problems, when adding subdomains using cpanel I get errors 

"
Bind reloading on imn1 using rndc zone: [leaseweb.imn.pl]
Error reloading bind on imn1: socket.c:2367: isc_thread_create() failed
rndc: create socket manager: unexpected error
The subdomain, sdc.leaseweb.imn.pl has been added.
"


strace:
http://tech.imn.pl/redhat/cpanel-strace.txt


FTP Server (pureftpd) with standard config and stack size (10240) when trying uploading file on server I get segfault

Strace:
http://tech.imn.pl/redhat/pureftpd-strace.txt


I get an idea to set up the higher value of stack size, I added to limits.conf 

*               soft    nofile          65536
*               hard    nofile          65536
*               soft    stack           65536
*               hard    stack           65536


I restarted server and testing what will be going on, and then starts showing seriusly porblems, ssl stop working in Cpanel, and using exim, could not send any mails.

Version-Release number of selected component (if applicable):
glibc-2.3.4-2.9

How reproducible:
Always

Steps to Reproduce:
1. -
2.
3.
  

Additional info:

Comment 1 Jakub Jelinek 2005-08-18 19:24:58 UTC
This has nothing to do with glibc actually.
If you look at the strace, you'll see:
[pid 23597] mmap2(NULL, 67112960, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = -1 ENOMEM (Cannot allocate memory)
pthread_create tried to allocate thread's stack, but there wasn't enough
memory.  This depends on what
/proc/sys/vm/overcommit_memory
/proc/sys/vm/overcommit_ratio
setting you are using, how much RAM + swap do you have and how many and how
memory hungry other processes are.  In any case that is a kernel decision.
You can try echo 1 > /proc/sys/vm/overcommit_memory or increase your swap size.

Anyway, if your ftp daemon needs more than 10MB of stack, especially if that's
remotely triggerable how much will it actually need, I'd call it severely broken.
If there are apps where it is justified to run them with bigger stack, either
(if that is just non-initial thread stack needs, you can use
pthread_attr_setstacksize, or you can just increase the hard limit systemwide
and run only the particular program with bigger soft stack limit).


Note You need to log in before you can comment on or make changes to this bug.