| Capture core dump inside VE [message #8345] |
Thu, 16 November 2006 12:31 |
iziz
Messages: 10
Registered: October 2006
|
Junior Member |
|
|
Hello,
I'm trying to capture core dump file inside particular Virtual Enviroinment. I set up in root node kernel.core_pattern kernel parameter and I'm able to see the change inside VE with `sysctl -a | grep kernel.core_pattern`:
kernel.core_pattern = /var/tmp/core
I also set up ulimit -c unlimited in /etc/profile (commenting line with default `ulimit -S -c 0 > /dev/null 2>&1` command) and restarted application I want to debug.
But when I kill it with 11 signal for example, I do not see any core dumps in /var/tmp/core. And it does not seem that they are generated in some unknown location, since typically my application writes "possible core dump in /some/dir" if it really generates dump.
However above operations are valid way to switch core dumps on inside usual Linux machine.
What are the reasons for that? How to switch core dumps on in OpenVZ?
|
|
|
|
OpenVZ Forum
Members
Search
Help
Register
Login
Home
