Re: [Crash-utility] How to work with crash utility.

Hi Nishant, Welcome to the group! Your usecase for using Crash is very apt and you can look for the following steps to start with: - Did you configure Kdump utility? In case you did not: Kdump utility lets you take a core-dump of the whole kernel once it hits a panic/oops situation, and save the same in a specified location. It is available as part of kexec-tools package. - You may want to try a dry run by inducing a panic with echo 'c' > /proc/sysrq-trigger and checking if kdump captured a dump file in the directory specified in the /etc/kdump.conf or not. - Once the above two basic steps are done, you are now set to analyze your problem with the captured context in the coredump file using Crash. - Install the kernel-debuginfo packages for the exact version of *kernel you are trying to debug.* You can find the debuginfo packages in a yum/fedora public repos. - Now start Crash with the two important parameters: coredump file (called as namelist file) location and place where you kernel-debuginfo file (vmlinu *x*). You can find the later /usr/lib/debug/lib/modules/2.6.29.4-167.fc11.x86_64/vmlinux (note that the exact kernel version might be different) Now the command to use is: crash <path to the vmlinux file> <path to your coredump file> Hope this helps in getting you started. Regards, Ratnam On Fri, Feb 4, 2011 at 1:08 PM, nishant mungse <nishantmungse(a)gmail.com&gt;wrote: