On Tuesday 12 April 2005 23:46, Mike Bird wrote: >On Tue, 2005-04-12 at 19:23, Peter Jones wrote: >> I'm choosing to continue not resurrecting lilo, since none of the >> group of you who want it back are willing to provide any technical >> reasons whatsoever. > >Historically Grub has been unreliable. [See bugzillas for Fedora, >Debian, GNU, etc, etc.] If someone tested the latest version of > Grub for six months and found no bugs at all, that would not change > the fact that historically Grub has been notoriously unreliable. I beg your pardon! I will submit that only those who do not want to understand how grub works will ever have any problems with it. Once I learned howto edit my own grub.conf, it has never been a problem. Flat statement. >Historically Lilo has been reliable. Even a five year old version > is works fine. If you don't screw up the src file that is, or insatll it on a system whose boot partition lives beyond the 1024th cylinder for that 5year old version. Screw up an entry in grub.conf, and the previous one is still bootable so its an easy fix once you realise your fat fingers didn't type what you told them to in adding the new entry. That happens to me more often than I want to admit at my age. >Now, after five years of missing features and poor reliability, you >claim that Grub is as reliable as Lilo. I doubt it - and experience >dictates scepticism here - but nobody yet knows for certain. Its far more reliable. >Does your opinion merit our trying Grub again sometime? Yes. > >Does your opinion justify throwing out Lilo? No. It's an > unjustifiable risk based upon the packages' respective histories. I threw lilo out with the bath water back about RH7.3, and have never had any reason to regret that, ever. >The WMD have now morphed into the difficulty of running Lilo from >Anaconda. (Seems incredible to me but I'll allow you some leeway > here since I haven't actually tried to write code to invoke Lilo > from Anaconda.) Well, if they don't get along, add a --play-nicely > patch to Lilo. It's a lot easier to munge a command-line interface > in Linux space than to reinvent all of Lilo's wheels in boot space. > >Or work on Grub if Redhat shareholders don't object. Use Grub if > you prefer it. Make Grub the default for newbies if you like. > Just don't deliberately break things for those of us who need the > reliable service that Lilo provides. > >YOU are the one who broke things. YOU are the one who has provided > no justification for breaking things. WE aren't trying to dictate > which boot loader you use. WE just want YOU to stop breaking > things. (Is this getting too personal?) Please, look in the mirror before you accuse someone of breaking grub. >Lilo worked fine until you removed it. Put it back. Don't mess > with it. Focus on Grub. Try not to think about Lilo. Let the > scripts compile it with the rest of FC4 and ship it on the damned > CD's. Now, thats not to say that someone trying to write a grub.conf updater thats at least as smart as someone fam with vim can't screw tyour grub.conf up. Thats why I *never*, _ever_ do a make install when building a new kernel, and frankly, I have never used rpm to update a kernel for exactly the same reason. Those scripts are carved for the lowest common denominator, and can very well hose your ability to recover by booting the old install. Thats the rpm's post processing script at fault, and no fault of grub. Your are pointing the finger at the wrong perp. >--Mike Bird -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) 99.34% setiathome rank, not too shabby for a WV hillbilly Yahoo.com and AOL/TW attorneys please note, additions to the above message by Gene Heskett are: Copyright 2005 by Maurice Eugene Heskett, all rights reserved.