# Core dumps

**URL:** <https://www.ruby-forum.com/t/core-dumps/207469>\
**Category:** NGINX\
**Created:** [May 26, 2011, 10:07am UTC](https://www.ruby-forum.com/t/core-dumps/207469 "2011-05-26T10:07:30Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![dubstep](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/d/cdc98d/32.png) [@dubstep](https://www.ruby-forum.com/u/dubstep)\
**Post date:** [May 26, 2011, 10:07am UTC](https://www.ruby-forum.com/t/core-dumps/207469/1 "2011-05-26T10:07:30Z")

</div>

Hi  
I’m trying to get a core dump out of nginx but it isn’t working…

I have done the following:

Made directory with permissions:  
drwxr-xr-x 2 root root 4.0K 2011-05-26 00:32 nginx-core

changed config:  
worker\_rlimit\_core 512M;  
working\_directory /home/dcstaff/nginx-core/;

but no dump is made when it crashes:  
2011/05/26 06:46:50 [alert] 20526#0: worker process 20547 exited on  
signal 11

Is there any other trick to it?

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [May 26, 2011, 10:11am UTC](https://www.ruby-forum.com/t/core-dumps/207469/2 "2011-05-26T10:11:51Z")

</div>

I should say this is on linux - ubuntu 10.10

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [May 26, 2011, 10:26am UTC](https://www.ruby-forum.com/t/core-dumps/207469/3 "2011-05-26T10:26:42Z")

</div>

Hello!

On Thu, May 26, 2011 at 08:06:37AM +0000, Richard K. wrote:

> working\_directory /home/dcstaff/nginx-core/;
> 
> but no dump is made when it crashes:  
> 2011/05/26 06:46:50 [alert] 20526#0: worker process 20547 exited on signal 11
> 
> Is there any other trick to it?

Worker processes are usually run under non-root, so it’s unlikely  
that write permissions for root will be enough.

Maxim D.

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [May 26, 2011, 11:03am UTC](https://www.ruby-forum.com/t/core-dumps/207469/4 "2011-05-26T11:03:12Z")

</div>

Sorry, I chose a bad example  
I have 3 machines where it’s not dumping  
The other 2 had the correct permissions already  
Should it core dump for signal 11 of a worker?

> Worker processes are usually run under non-root, so it’s unlikely  
> that write permissions for root will be enough.

* * *

nginx mailing list  
[removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)

> **[mailman.nginx.org archive](http://mailman.nginx.org/pipermail/)**

---

<div class="post-metadata">

**Author:** ![rkoskela](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/r/ce73a5/32.png) [@rkoskela](https://www.ruby-forum.com/u/rkoskela)\
**Post date:** [May 26, 2011, 12:37pm UTC](https://www.ruby-forum.com/t/core-dumps/207469/5 "2011-05-26T12:37:49Z")

</div>

On Thu, May 26, 2011 at 09:02:26AM +0000, Richard K. wrote:

> Sorry, I chose a bad example  
> I have 3 machines where it’s not dumping  
> The other 2 had the correct permissions already  
> Should it core dump for signal 11 of a worker?

Probably you need to set also:  
echo 1 \> /proc/sys/fs/suid\_dumpable

> > Worker processes are usually run under non-root, so it’s unlikely  
> > that write permissions for root will be enough.

–  
Igor S.