On Tue, Aug 15, 2006 at 07:21:29PM +0200, Henning Sprang wrote: > And what about checking the return code of debootstrap(if it has a usable)?
It already should detect a failure in the case that debootstrap gives an error result. Just the display of the output was broken. I'm not 100% sure if this new addition will cause all errors to be displayed since debootstrap has a tendency to not give useful return codes and that makes detecting failures .. tricky. > (BTW: please don't fix the error handling "too good", because right now > I am depending on the possibility to run copy with /dev/null or > --mirror=file:///dev/null to achieve the goal of creating only configs > and disk files but not installing anything... see the other bug report > about that :) ) That shoulkdn't be a problem. The isntallation of the system is done by executing the xt-install-system script and that will only attempt one installation method. Either --copy, --untar, --rpmstrap, or --debootstrap. So debootstrap error detection and improvement(s) should have no effect on the other methods. Steve -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
