>> Would it be possible to extend rpm with an option to be used >> in kernel rpms, so that when people (mistakenly) run >> "rpm -Uvh kernel-2.6...", the old kernel rpm is not uninstalled? >Seems you are trying to code around a lack of common sense. > >People will only do -Uvh until it bites them, then never >again... It sucks, but sometimes the School of Hard Knocks is >the best way to get a lasting education. There are other approaches. Many distros seem to distribute kernel rpms named in such a way that a new kernel doesn't look (to rpm) like a newer version of an older one, and so updating in place just doesn't work. With -A running, install -B, reboot, if everything checkes out you can delete -A. I suspect this means that too many users have gotten bitten or they wouldn't be doing this. Now if distros would be a little more cautious what they add... maybe we could count on replacing the kernel with a supported update as not being a recipe for disaster. In the once-bitten camp, I once upgraded my kernel on a SUSE distro (one who doesn't do the "your can't upgrade in place" trick) - it was supposed to be just a security fix, but the new one absolutely wouldn't run on my hardware. Later digging turned out just a little bit more :-) than the security fix had gone in... And I never asked for "rpm -Uvh", for this one I used the Yast gui, which offered several things that ought to get upgraded for various serious-sounding reasons. The summary is, this is a distro problem to solve, and it's within their power to solve it without changing rpm. _______________________________________________ Rpm-list mailing list Rpm-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/rpm-list
