[Crash-utility] Re: crashdc : automatic data collection for new vmcores in text format

[Mike Snitzer]

On Fri, Sep 25 2009 at 10:55am -0400,
Louis Bouchard <louis.bouchard at hp.com> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Dave Anderson a écrit :
> > 
> > I'm guessing the run-*.sh scripts are easy to modify in order for
> > a user to tailor them to their needs?
> > 
> > Anyway, it sounds reasonable to me.
> > 
> > Can you work up a patch to the crash.spec file to create an additional
> > sub-package?  Presume a new crashdc subdirectory off the top-level
> > to contain the files so that tar.gz users will get them as well.
> > 
> > Thanks,
> >   Dave
> > 
> 
> The idea that I'm following is to keep crashdc generic so it can be used
> on the cmdline and used in an automated fashion.  The run-*.sh are where
> distro specific stuff gets put in (i.e. location of the dump directory,
> debuginfo files etc). This is the script that is invoked by the kdump
> mechanism.

Seems like you should have a config file also; so that each of the
run-*.sh don't contain path information, etc.. just the distro specific
mechanism.

make the configuration common (e.g. /etc/sysconfig/crashdc ?).  Have
templates as you like for each distro.  But config info does not belong
hidden in scripts in /usr/bin/

Mike