Jan Engelhardt wrote:
Think outside rpm for a change. The less steps in the install, the more likely to be successful.
Actually just thought about it -- When I auto-update my packages, new kernels are often installed as well -- the suse way. They are completely ignored. They fail to replace my current kernel (which is good for me. I suppose it is bad, but I've come to expect and somewhat rely on the fact that installing a suse kernel will have no effect on my boot process... That's how reliable the suse boot install is. I DO have my /etc/sysconfig/bootload configed for the correct loader... Its just that the code your boot installer relies on to insert it's new kernel into a bootable position in the updated system has been broken for several releases. So you want to talk chances of your install being successful or reliability? Let's not -- then I'd have to worry that it might get it right and I might have to actually do something about it. That bug was the thing that kicked me back from using grub which I used for a while (it was documented in one of the references I posted yesturday about XFS support being hosed due to grub. Another grub bug -- I changed my boot partition. so new grub installs were into the new place, but boot kept using the old boot partition (which thankfully I hadn't wiped yet). Switching back to lilo solved both problems. It did its updates to the partition I was actually going to boot from -- not the one that grub kept seeing as my boot disk. So "think outside grub for a change" (especially it's boot paradigm. It reliably doesn't work in various situations. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org