On Sun, 26 Oct 2008 09:49:14 -0700, Darren New wrote:

> Jim Henderson wrote:
>> I'm now wondering if that's the case - if it uses ntfs.sys;
> 
> I might have been out of date. Maybe that was ntfs-2g or something.  I
> just know what I've heard people tell me.  My research (as of a day or
> two ago) tells me it's actually all native code, unencumbered by IP.

I've probably been one of those people telling you, because I thought it 
did use ntfs.sys.  :-)

>> Well, yes.  There was specific boot sector code in the MBR to do that,
> 
> Yeah, but every MBR I ever saw does it that way. It used to be the only
> way to boot a different OS.

Yeah, until we got to the fancy menus, which required additional code 
that wouldn't fit in one sector.

>>> Yes, MS's boot sector follows the standard that's been around for a
>>> decade longer than Linux has. GRUB's doesn't. What do you think MS is
>>> doing wrong here?
>> 
>> My understanding is that the Microsoft MBR (at least as included in
>> Vista, possibly with older versions as well) depends on that boot.ini
>> file, much as GRUB depends on the files in /boot/grub.
> 
> Maybe Vista, since it has a new boot record. I'll play with it to see.
> 
> But everything before that, from DOS thru Win95 through XP all just load
> the first sector of the active partition and jumps to it. Now, that
> said, the first sector of the XP partition relies on NTLDR and boot.ini
> being present. But you'd have to work really hard to fit NTFS into 480
> bytes, even if only to find boot.ini, you know?

True, now you mention it, I do recall this.

>> Thing is, it shouldn't require a change to the active partition list.
> 
> It doesn't *require* it. It *allows* it.

No, I mean in order to boot a different partition.

>> But also, GRUB can be told to change the default for a single boot only
>> - at least I seem to recall there's an option to do that.
> 
> Again, my Linux sysop knowledge is probably out of date. The version as
> of a couple years ago doesn't allow that, and I can't afford to risk
> breaking 60 production machines to check if it's better now. :-)

I'll have to see if I can find that one - I was sure I saw the option 
somewhere, but of course now I can't find it. :-)

>> Installing Linux (at least openSUSE) on a drive with Windows on it, the
>> installer will set Windows up as a menu option so you can select
>> either. Installing Windows after Linux, though, Windows won't add Linux
>> to the boot menu automatically.  I guess that's what I was trying to
>> say.
> 
> Oh, it's not automated, sure. But it's pretty trivial, compared to a lot
> of Linux things. :-)  If you wind up installing Windows after Linux, you
> probably are smart enough to type the two or three command-line bits it
> takes to get the Linux boot sector into Windows.

Probably, yes - though it's probably easier to modify the Grub setup at 
that point and just add the Windows install.  Most people I know who 
install Windows second, though, use a VM rather than a native boot.

>> I had thought GRUB on my system here was set up with the MBR, but I was
>> in fact mistaken - it's in the root partition instead.
> 
> I'm not sure what that means. I know *all* of GRUB doesn't fit in the
> MBR, any more than NTLDR does. It's just a question of whether GRUB's
> MBR does something necessary to make GRUB boot, or whether you can boot
> a GRUB partition off someone else's MBR. Right now, my tests (on SuSE
> 10.2) are telling me GRUB needs GRUB's MBR. It might be better now, or I
> might be doing something wrong.

It means that GRUB on my system here is using the active partition.  I 
don't know what's in the MBR on this laptop's hard drive, but the 
settings when I go into YaST and look indicate that it's installed to the 
first sector of the active partition.

Jim

Post a reply to this message

On Sun, 26 Oct 2008 09:40:46 -0700, Darren New wrote:

> Jim Henderson wrote:
>> YaST will let you do this.  I was just in there checking it out. :-)
> 
> Cool. Perhaps they improved that since I first tried it a couple years
> ago. I couldn't imagine nothing would do it. I'm pretty sure the
> command-line fdisk on Linux allows it too.

Oh, sure fdisk or cfdisk will let you change the active flag in a 
partition.

Jim

Post a reply to this message

On Sun, 26 Oct 2008 09:40:02 -0700, Darren New wrote:

> Jim Henderson wrote:
>>> I'll eat my underwears if they do. :D
>> You'll have to video that one. ;-)
> 
> In my experience, the proper response is
>     GIF! GIF! GIF!

LOL

Jim

Post a reply to this message

On Sun, 26 Oct 2008 09:51:04 -0700, Darren New wrote:

> Jim Henderson wrote:
>> I think this is because GRUB is actually a more complex piece of
>> software compared to LILO.
> 
> Yep.
> 
>> That's not how I remember it, but I generally didn't have multiple MS
>> OSes installed simultaneously.
> 
> Remember that MS was selling Xenix long before Linux was conceived.

True, but there again, the Xenix systems that I saw and worked on only 
ran Xenix....We actually used that for the "library" system in the 
computer lab at the school I went to.

Jim

Post a reply to this message

Jim Henderson wrote:
>>> Thing is, it shouldn't require a change to the active partition list.
>> It doesn't *require* it. It *allows* it.
> 
> No, I mean in order to boot a different partition.

Windows and GRUB are the same in this respect. If the active partition 
is, for some reason, unreadable, you won't get a menu to boot the other 
partition. However, you can use a non-OS tool (fdisk off a dos disk, 
say) to change the active partition and boot the other OS.

It's a two-stage boot. You're talking about the second stage. I'm 
talking about the first stage.

> Probably, yes - though it's probably easier to modify the Grub setup at 
> that point and just add the Windows install.  Most people I know who 
> install Windows second, though, use a VM rather than a native boot.

There are definitely more options now than there were, yes. :-)

> It means that GRUB on my system here is using the active partition.  I 
> don't know what's in the MBR on this laptop's hard drive, but the 
> settings when I go into YaST and look indicate that it's installed to the 
> first sector of the active partition.

It's most likely installed to both. It's a two-stage boot. Maybe that's 
what's confusing the conversation.

The first stage is to load the MBR off (basically) sector zero of the 
hard drive, which happens to include the partition table. The *stnadard* 
MBR then looks at the partition table, finds the active partition, moves 
itself out of the way, reads the first sector of the active partition 
into the same place it was, and jumps to it.

Only after that second step do you get any kind of menu. If you delete 
NTLDR or /boot off the active partition, you're not going to have an 
opportunity to boot off a different partition. With the standard MBR, 
you don't have to put NTLDR or /boot back, you just have to change the 
active flag.  Last I experiemented, the GRUB MBR booted the GRUB 
partition even if it wasn't marked active, which means if your GRUB 
partition is corrupt, you're SOL.

-- 
Darren New / San Diego, CA, USA (PST)

Post a reply to this message

On Sun, 26 Oct 2008 13:25:44 -0700, Darren New wrote:

> Jim Henderson wrote:
>>>> Thing is, it shouldn't require a change to the active partition list.
>>> It doesn't *require* it. It *allows* it.
>> 
>> No, I mean in order to boot a different partition.
> 
> Windows and GRUB are the same in this respect. If the active partition
> is, for some reason, unreadable, you won't get a menu to boot the other
> partition. However, you can use a non-OS tool (fdisk off a dos disk,
> say) to change the active partition and boot the other OS.
> 
> It's a two-stage boot. You're talking about the second stage. I'm
> talking about the first stage.

Oh, right - yes, and both platforms use the two-stage system these days.

>> Probably, yes - though it's probably easier to modify the Grub setup at
>> that point and just add the Windows install.  Most people I know who
>> install Windows second, though, use a VM rather than a native boot.
> 
> There are definitely more options now than there were, yes. :-)

Absolutely. :-)

>> It means that GRUB on my system here is using the active partition.  I
>> don't know what's in the MBR on this laptop's hard drive, but the
>> settings when I go into YaST and look indicate that it's installed to
>> the first sector of the active partition.
> 
> It's most likely installed to both. It's a two-stage boot. Maybe that's
> what's confusing the conversation.

Perhaps.

> The first stage is to load the MBR off (basically) sector zero of the
> hard drive, which happens to include the partition table. The *stnadard*
> MBR then looks at the partition table, finds the active partition, moves
> itself out of the way, reads the first sector of the active partition
> into the same place it was, and jumps to it.
>
> Only after that second step do you get any kind of menu. If you delete
> NTLDR or /boot off the active partition, you're not going to have an
> opportunity to boot off a different partition. With the standard MBR,
> you don't have to put NTLDR or /boot back, you just have to change the
> active flag.  Last I experiemented, the GRUB MBR booted the GRUB
> partition even if it wasn't marked active, which means if your GRUB
> partition is corrupt, you're SOL.

Well, only as far as that disk is - a bootable rescue disk can fix that, 
but also in my experience GRUB can get to a point where you can type in 
commands to boot the system.  I used to have to do that with RH9 systems 
I'd imaged on occasion.  So even if it didn't read the menu, I could get 
the system up to a point where I could fix the menu.

Jim

Post a reply to this message

Darren New wrote:
> 
> Yeah. Funny enough, people complain about Vista not supporting hardware
> right, but don't think that's a problem for Linux. :-) "Just buy the
> right hardware to start with."  My machines tend to last a number of
> years, so "buy the right hardware" isn't always possible for me.
> 

To exactly how many pieces of hardware have you lost the support on
Linux while upgrading to a newer version of Linux? The difference
between Linux and Vista here is that if your device worked with Zoot
(RedHat 6.2) and 2.2 -series kernel, it most probably still works with
newest SuSE and 2.6 -series kernel - at least if you're still able to
physically plug the device in.

If your hw works with Linux now, it'll work for the number of years,
even if you do update your software. As far as I've understood, if your
hw works with 2k or XP, you can't be sure it'll still work with Vista.

And yes, I'll have to admit - em8300 -support took a while to work in
64-bit. That's the only case I've hit my head with.

-- 
Eero "Aero" Ahonen
   http://www.zbxt.net
      aer### [at] removethiszbxtnetinvalid

Post a reply to this message

Jim Henderson wrote:
> 
> I've also heard that some of the newer HP laserjet printers don't have 
> open specs so there aren't drivers yet for them.  That one I haven't 
> confirmed, just have heard about.

That must be the cheapest bonk-line LJ's. All real-printer LJ's I've
seen have talked PostScript and various PCL versions and are supported
very widely.

> 
> Jim


-- 
Eero "Aero" Ahonen
   http://www.zbxt.net
      aer### [at] removethiszbxtnetinvalid

Post a reply to this message

Jim Henderson wrote:
> 
> Oh, that's not Grub, that's SYSLINUX - Grub is hard-disk only to the best 
> of my knowledge.
> 

I'm not sure about CD-ROM, but at least Grub is able to boot via PXE.

> Jim


-- 
Eero "Aero" Ahonen
   http://www.zbxt.net
      aer### [at] removethiszbxtnetinvalid

Post a reply to this message

Eero Ahonen <aer### [at] removethiszbxtnetinvalid> wrote:
> To exactly how many pieces of hardware have you lost the support on
> Linux while upgrading to a newer version of Linux? The difference
> between Linux and Vista here is that if your device worked with Zoot
> (RedHat 6.2) and 2.2 -series kernel, it most probably still works with
> newest SuSE and 2.6 -series kernel - at least if you're still able to
> physically plug the device in.

  I think one advantage of Linux over Windows is that Linux has absolutely
no need to try to "sell" new versions of itself. This allows for a much
more gradual development of the entire system, and there's no need to
artificially make a "new version of linux" (if we even can rationally talk
about one) look&feel different from earlier versions. On the contrary, one
of the strengths of Linux is precisely that it doesn't need to try to be
different with each new version, so it can not only keep backwards
compatibility at software level, but also at the user interface level.
Any changes in usage are caused by necessity, not by an attempt to be
different for the sake of being different.

  Not so with Windows. Microsoft *must* sell a completely brand new version
of Windows each n years. Of course people wouldn't buy the new version if
it was basically identical to the old version, with just internal invisible
improvements, and perhaps a few new programs (which would work in the old
version anyways). Thus MS has to make new versions of Windows look&feel
different from earlier versions. MS has to create the illusion that the
new version is much better than the old version, and that people should
definitely upgrade. That sells.
  The sad thing that changing the look&feel often means artificial changes
which are not for the better. It also sometimes means breaking backwards
compatibility with older software and even hardware.
  Another way of selling the new version is to include something which
does not work in the older version (DirectX10 anyone?)

  The sad thing is that people are ready to conform.

-- 
                                                          - Warp

Post a reply to this message