FC4t2 no good without LILO

[Peter Jones]

On Tue, 2005-04-12 at 10:21 -0700, Mike Bird wrote:
> > Which of the 82 Grub bugs prevents it from being used for serious work?
> 
> 1) That basic Bugzilla queries yield 82 Grub bugs versus 15 Lilo bugs is
> illustrative of their relative levels of reliability.

No, that's just willfully stupid.  Go read my last reply to you.

> 2) The general flakiness of Grub makes it unusable for serious work. 
> One simply cannot afford plane tickets (or even trouble tickets) every
> time Grub fails.

You haven't established that any such general flakiness exists.  So
until you're prepared to actually tell us about some *real* problem,
please stop trolling fedora-test-list, and any other lists as well.

[silliness removed]
> 3) One of the most serious ongoing problems with Grub has been the
> flakiness of the software RAID support.  This was first added in 0.90
> (July 2001) but has ever since been a constant source of Bugzilla
> reports [...]

It wasn't added in any meaningful way until early this year, and it
hasn't particularly been a problem; there have been bugs here and there
in applications that interact with grub, but they've (mostly) been
fixed.  It's new code, bugs happen.  Somebody notices, and they get
fixed.  That's how this works.

The one serious problem with it that I know of currently, I'd be fixing
right now if you weren't tempting me so very strongly to waste my time.
And it's not even among your completely bogus 82.

> [...] - reports which are not included in the list of 82 because they
> were closed when the March 16th patch was written.  Now maybe that patch
> will solve all known problems

It implements actual support /boot on a RAID 1 device.  This is a new
feature.  If you know about problems with it, you're the only one, so
quit saying they're known and file a bug telling me what they are.

> , but I'm certainly not going to bet the
> farm on it without months of testing.  Why is Fedora betting its farm? 

We've tested it for years, and know its strengths and weaknesses, and
those of LILO.  We picked the better solution.  You clearly don't like
it, but you've made no actual technical points against it.

Guess what?  The part of all this where you simply refuse to make any
meaningful technical point at all turns out to be a pretty abysmal plan
in convincing me to switch things back to LILO.

> Lilo's RPM takes only 547k on the CD.

Thank you for inventing a nice strawman so you could knock it down
yourself.

Size isn't the issue; supportability and usability are.  LILO isn't
supportable.  And I mean supportable in *every* since.  Installing it in
software is uglier than grub, and is also incredibly failure prone.
Installing kernels with lilo is basically unscriptable, and certainly
unreliable.  It's not something you can depend on at all.  Debugging
what's going wrong without seeing the hardware itself tends to be neigh
on impossible.  And, of course, once you've managed to screw up a kernel
installation, you can no longer boot the box at all.

Grub has exactly one of these problems -- you can mess up installing it
to the disk.  But you have to do it once, when you install the OS.  Not
every time you want to boot a new kernel.  Not every time you add a new
device that you decide needs to be in the initrd.

> 4) And this brings us again to the incredible lack of judgment
> manifested by the Fedora Core team.

If it weren't completely bull, yeah, that'd demonstrate our "incredible
lack of judgment" amazingly.

> First do no harm - if you don't understand the issues relating to Lilo
> and Grub don't mess with them.

Seriously, have you even thought about this issue, or do you just like
to flame and troll a lot?
-- 
        Peter