[gentoo-user] Is grub2 stable and who uses it?

[gentoo-user] Is grub2 stable and who uses it?

[Dale]

Hi,

Subject line says it pretty well.  Is grub2 stable, who uses it and can 
you post your experience on the switching process?  Was it difficult?  
Easy?  Somewhere between?

Thinking about switching.  Get this over with before all the initramfs 
thingy kicks in.

BTW, I still haven't got that initramfs thingy finished.  :-(

Thanks.

Dale

:-)  :-)

Re: [gentoo-user] Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 790 bytes --]

On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:

> Subject line says it pretty well.  Is grub2 stable, who uses it and can 
> you post your experience on the switching process?  Was it difficult?  

I use it on my netbook, which I admittedly don't boot more than a couple
of times a month. It's stable, I can't comment on the switching process
as I used GRUB2 from the start with this machine, it seemed a good time
to get to grips with it.

GRUB2 is neither complicated nor difficult, but it is different. If you
try to think in terms of legacy GRUB, you will have more problems than if
you approach is as learning a new system.


-- 
Neil Bothwick

In the 60's people took acid to make the world weird.
Now the world is weird and people take Prozac to make it normal.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[Pandu Poluan]

[-- Attachment #1: Type: text/plain, Size: 834 bytes --]

On Oct 4, 2011 5:10 PM, "Neil Bothwick" <neil@digimed.co.uk> wrote:
>
> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>
> > Subject line says it pretty well.  Is grub2 stable, who uses it and can
> > you post your experience on the switching process?  Was it difficult?
>
> I use it on my netbook, which I admittedly don't boot more than a couple
> of times a month. It's stable, I can't comment on the switching process
> as I used GRUB2 from the start with this machine, it seemed a good time
> to get to grips with it.
>
> GRUB2 is neither complicated nor difficult, but it is different. If you
> try to think in terms of legacy GRUB, you will have more problems than if
> you approach is as learning a new system.
>

Kind of tangential...

Why does Gentoo still 'standardize' on grub instead of going forward with
grub2?

Rgds,

[-- Attachment #2: Type: text/html, Size: 1079 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[Michael Orlitzky]

On 10/04/2011 06:16 AM, Pandu Poluan wrote:
> 
> On Oct 4, 2011 5:10 PM, "Neil Bothwick" <neil@digimed.co.uk
> <mailto:neil@digimed.co.uk>> wrote:
>>
>> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>>
>> > Subject line says it pretty well.  Is grub2 stable, who uses it and can
>> > you post your experience on the switching process?  Was it difficult?
>>
>> I use it on my netbook, which I admittedly don't boot more than a couple
>> of times a month. It's stable, I can't comment on the switching process
>> as I used GRUB2 from the start with this machine, it seemed a good time
>> to get to grips with it.
>>
>> GRUB2 is neither complicated nor difficult, but it is different. If you
>> try to think in terms of legacy GRUB, you will have more problems than if
>> you approach is as learning a new system.
>>
> 
> Kind of tangential...
> 
> Why does Gentoo still 'standardize' on grub instead of going forward
> with grub2?

Grub2 is weird (coming from anything that isn't grub2), and if you mess
up the upgrade, you can't boot.

It's a support nightmare.

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Neil Bothwick <neil@digimed.co.uk> wrote:
> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>
>> Subject line says it pretty well.  Is grub2 stable, who uses it and can 
>> you post your experience on the switching process?  Was it difficult?  
>
> I use it on my netbook, which I admittedly don't boot more than a couple
> of times a month. It's stable, I can't comment on the switching process
> as I used GRUB2 from the start with this machine, it seemed a good time
> to get to grips with it.
>
> GRUB2 is neither complicated nor difficult, but it is different.

I've only used it on Ubuntu, and maybe it's just Ubuntu's
implementation -- but it was both complicated and difficult.  There
are 10X as many files, and to change anything you edit a whole set of
configuration files and run a utility that generates _another_ set of
configuration files.

Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.

> If you try to think in terms of legacy GRUB, you will have more
> problems than if you approach is as learning a new system.

At first glace, grub2 looks like a minature Unix installation whose
purpose is to boot a bigger Unix installation.  It's got it's own init
system and it's own set of init scripts.

-- 
Grant Edwards               grant.b.edwards        Yow! Could I have a drug
                                  at               overdose?
                              gmail.com            

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Dale]

Grant Edwards wrote:
> On 2011-10-04, Neil Bothwick<neil@digimed.co.uk>  wrote:
>> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>>
>>> Subject line says it pretty well.  Is grub2 stable, who uses it and can
>>> you post your experience on the switching process?  Was it difficult?
>> I use it on my netbook, which I admittedly don't boot more than a couple
>> of times a month. It's stable, I can't comment on the switching process
>> as I used GRUB2 from the start with this machine, it seemed a good time
>> to get to grips with it.
>>
>> GRUB2 is neither complicated nor difficult, but it is different.
> I've only used it on Ubuntu, and maybe it's just Ubuntu's
> implementation -- but it was both complicated and difficult.  There
> are 10X as many files, and to change anything you edit a whole set of
> configuration files and run a utility that generates _another_ set of
> configuration files.
>
> Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.
>
>> If you try to think in terms of legacy GRUB, you will have more
>> problems than if you approach is as learning a new system.
> At first glace, grub2 looks like a minature Unix installation whose
> purpose is to boot a bigger Unix installation.  It's got it's own init
> system and it's own set of init scripts.
>

Could this fix the mess with /usr and /var having to be on / or a 
initramfs?

Dale

:-)  :-)

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 7:35 AM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Neil Bothwick <neil@digimed.co.uk> wrote:
>> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>>
>>> Subject line says it pretty well.  Is grub2 stable, who uses it and can
>>> you post your experience on the switching process?  Was it difficult?
>>
>> I use it on my netbook, which I admittedly don't boot more than a couple
>> of times a month. It's stable, I can't comment on the switching process
>> as I used GRUB2 from the start with this machine, it seemed a good time
>> to get to grips with it.
>>
>> GRUB2 is neither complicated nor difficult, but it is different.
>
> I've only used it on Ubuntu, and maybe it's just Ubuntu's
> implementation -- but it was both complicated and difficult.  There
> are 10X as many files, and to change anything you edit a whole set of
> configuration files and run a utility that generates _another_ set of
> configuration files.
>
> Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.
>
>> If you try to think in terms of legacy GRUB, you will have more
>> problems than if you approach is as learning a new system.
>
> At first glace, grub2 looks like a minature Unix installation whose
> purpose is to boot a bigger Unix installation.  It's got it's own init
> system and it's own set of init scripts.

That it's not true. It connects to whatever init system do you have
(OpenRC, SysV, systemd, Upstart), and it has scripts to *generate* the
config file.

The thing is that GRUB2 needs to understand several filesystems to
grab the kernel image from. It also wants to be able to use a more
interesting resolution than 640x480. This means that it has to
reimplement all the code for any filesystem, and all the code for
video handling. The scripts are for semi automatic generation of the
config file, which is more complicated than the one from grub-legacy.
On the other hand, you only need to configure once, and run it every
time you compile and put a new kernel in /boot. But if you just change
your current kernel (same image file), you don't have to do anything.

Note that the version is 1.99, not 2.0. It is not finished: when 2.0
is reached, hopefully you will be able to disableat ./configure time
what video drivers and filesystems do you want to use. Also, the
scripts to generate the config file will be standardized by then.

However, in the last LPC, it was suggested that replicating filesystem
and video code on the kernel and grub was a terrible idea, and some
developers have suggested to use a /firstboot partition with a simple
filesystem, and populated with a kernel image and an initramfs. That
will mean that to boot Linux, we would use Linux.

You can read an article about it here: http://lwn.net/Articles/458789/

It was only a proposal: I don't know what will be the standard in the future.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> On Tue, Oct 4, 2011 at 7:35 AM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>> On 2011-10-04, Neil Bothwick <neil@digimed.co.uk> wrote:
>>> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>>>
>>>> Subject line says it pretty well. ??Is grub2 stable, who uses it and can
>>>> you post your experience on the switching process? ??Was it difficult?
>>>
>>> I use it on my netbook, which I admittedly don't boot more than a couple
>>> of times a month. It's stable, I can't comment on the switching process
>>> as I used GRUB2 from the start with this machine, it seemed a good time
>>> to get to grips with it.
>>>
>>> GRUB2 is neither complicated nor difficult, but it is different.
>>
>> I've only used it on Ubuntu, and maybe it's just Ubuntu's
>> implementation -- but it was both complicated and difficult. ??There
>> are 10X as many files, and to change anything you edit a whole set of
>> configuration files and run a utility that generates _another_ set of
>> configuration files.
>>
>> Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.
>>
>>> If you try to think in terms of legacy GRUB, you will have more
>>> problems than if you approach is as learning a new system.
>>
>> At first glace, grub2 looks like a minature Unix installation whose
>> purpose is to boot a bigger Unix installation. ??It's got it's own init
>> system and it's own set of init scripts.
>
> That it's not true. It connects to whatever init system do you have
> (OpenRC, SysV, systemd, Upstart),

I'm curious: what if you don't have one?  I use grub-legacy to boot
stuff other than Unix.

> and it has scripts to *generate* the config file.
>
> The thing is that GRUB2 needs to understand several filesystems to
> grab the kernel image from.

I understand why GRUB2 is complicated.  It's the statement that it's
not complicated that I was disagreeing with.

> It also wants to be able to use a more interesting resolution than
> 640x480.

That I don't understand. It's a bootloader.  It needs to allow you to
pick one of a handfull of choices and boot that choice.

> This means that it has to reimplement all the code for any
> filesystem,

That part I understand.

> and all the code for video handling.

I don't really understand the need for that, but I'm somebody who
still regularly uses a serial console.  [Insert the usual "I remember
when" grumbling here.]

[...]

> However, in the last LPC, it was suggested that replicating filesystem
> and video code on the kernel and grub was a terrible idea, and some
> developers have suggested to use a /firstboot partition with a simple
> filesystem, and populated with a kernel image and an initramfs. That
> will mean that to boot Linux, we would use Linux.

Yea, I've read about that.  The mind wobbles.  I suppose it's no worse
than VAXes having a PDP-11 inside to help it start up.  [I'm not
really sure that's true, but I heard it from several people who should
have known.]

-- 
Grant Edwards               grant.b.edwards        Yow! A dwarf is passing out
                                  at               somewhere in Detroit!
                              gmail.com            

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 11:31 AM, Grant Edwards
<grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>> On Tue, Oct 4, 2011 at 7:35 AM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>>> On 2011-10-04, Neil Bothwick <neil@digimed.co.uk> wrote:
>>>> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
>>>>
>>>>> Subject line says it pretty well. ??Is grub2 stable, who uses it and can
>>>>> you post your experience on the switching process? ??Was it difficult?
>>>>
>>>> I use it on my netbook, which I admittedly don't boot more than a couple
>>>> of times a month. It's stable, I can't comment on the switching process
>>>> as I used GRUB2 from the start with this machine, it seemed a good time
>>>> to get to grips with it.
>>>>
>>>> GRUB2 is neither complicated nor difficult, but it is different.
>>>
>>> I've only used it on Ubuntu, and maybe it's just Ubuntu's
>>> implementation -- but it was both complicated and difficult. ??There
>>> are 10X as many files, and to change anything you edit a whole set of
>>> configuration files and run a utility that generates _another_ set of
>>> configuration files.
>>>
>>> Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.
>>>
>>>> If you try to think in terms of legacy GRUB, you will have more
>>>> problems than if you approach is as learning a new system.
>>>
>>> At first glace, grub2 looks like a minature Unix installation whose
>>> purpose is to boot a bigger Unix installation. ??It's got it's own init
>>> system and it's own set of init scripts.
>>
>> That it's not true. It connects to whatever init system do you have
>> (OpenRC, SysV, systemd, Upstart),
>
> I'm curious: what if you don't have one?  I use grub-legacy to boot
> stuff other than Unix.

When I said "it connects", I mean "calls". The same way it calls
whatever thingy Window uses.

>> and it has scripts to *generate* the config file.
>>
>> The thing is that GRUB2 needs to understand several filesystems to
>> grab the kernel image from.
>
> I understand why GRUB2 is complicated.  It's the statement that it's
> not complicated that I was disagreeing with.
>
>> It also wants to be able to use a more interesting resolution than
>> 640x480.
>
> That I don't understand. It's a bootloader.  It needs to allow you to
> pick one of a handfull of choices and boot that choice.

I agree. That's why GRUB2 now is really 1.99, because it's not finished.

>> This means that it has to reimplement all the code for any
>> filesystem,
>
> That part I understand.
>
>> and all the code for video handling.
>
> I don't really understand the need for that, but I'm somebody who
> still regularly uses a serial console.  [Insert the usual "I remember
> when" grumbling here.]

Then stick with LILO or grub-legacy and root=UUID in your kernel command line.

> [...]
>
>> However, in the last LPC, it was suggested that replicating filesystem
>> and video code on the kernel and grub was a terrible idea, and some
>> developers have suggested to use a /firstboot partition with a simple
>> filesystem, and populated with a kernel image and an initramfs. That
>> will mean that to boot Linux, we would use Linux.
>
> Yea, I've read about that.  The mind wobbles.  I suppose it's no worse
> than VAXes having a PDP-11 inside to help it start up.  [I'm not
> really sure that's true, but I heard it from several people who should
> have known.]

I actually think is a good idea. I also think is not for everybody. As
I said, if the root=UUID kernel command line works, then nobody has
nothing to worry about anything: we would be able to use whatever boot
loader we want to, even LILO (if it still works).

Me, I want my laptop/desktop computers to have the best resolution
available from moment zero, even before loading the kernel, and not a
single flicker in my screen until my GNOME 3 is fully loaded. So I'm
gonna play with grub2 (or /firstboot, if it materializes) until it's
able to do that.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Alan McKinnon]

On Tue, 4 Oct 2011 08:08:16 -0700
Canek Peláez Valdés <caneko@gmail.com> wrote:

> > At first glace, grub2 looks like a minature Unix installation whose
> > purpose is to boot a bigger Unix installation.  It's got it's own
> > init system and it's own set of init scripts.  
> 
> That it's not true. It connects to whatever init system do you have
> (OpenRC, SysV, systemd, Upstart), and it has scripts to *generate* the
> config file.
> 
> The thing is that GRUB2 needs to understand several filesystems to
> grab the kernel image from. It also wants to be able to use a more
> interesting resolution than 640x480. This means that it has to
> reimplement all the code for any filesystem, and all the code for
> video handling.

Personally, I can't agree with this stance from the grub2 devs.

It's a bootloader. It is visible for 3 seconds at boot time. 

For driving the screen it should just use whatever facilities the
firmware one layer below it provides.

-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Alan McKinnon]

On Tue, 04 Oct 2011 09:53:07 -0500
Dale <rdalek1967@gmail.com> wrote:

> Grant Edwards wrote:
> > On 2011-10-04, Neil Bothwick<neil@digimed.co.uk>  wrote:
> >> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
> >>
> >>> Subject line says it pretty well.  Is grub2 stable, who uses it
> >>> and can you post your experience on the switching process?  Was
> >>> it difficult?
> >> I use it on my netbook, which I admittedly don't boot more than a
> >> couple of times a month. It's stable, I can't comment on the
> >> switching process as I used GRUB2 from the start with this
> >> machine, it seemed a good time to get to grips with it.
> >>
> >> GRUB2 is neither complicated nor difficult, but it is different.
> > I've only used it on Ubuntu, and maybe it's just Ubuntu's
> > implementation -- but it was both complicated and difficult.  There
> > are 10X as many files, and to change anything you edit a whole set
> > of configuration files and run a utility that generates _another_
> > set of configuration files.
> >
> > Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.
> >
> >> If you try to think in terms of legacy GRUB, you will have more
> >> problems than if you approach is as learning a new system.
> > At first glace, grub2 looks like a minature Unix installation whose
> > purpose is to boot a bigger Unix installation.  It's got it's own
> > init system and it's own set of init scripts.
> >
> 
> Could this fix the mess with /usr and /var having to be on / or a 
> initramfs?

No that's a completely different issue.

But the warped thinking that produces it is exactly the same.



-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[pk]

On 2011-10-04 20:56, Canek Peláez Valdés wrote:

Replying two mails in one...

Dale:
>Is grub2 stable, who uses it and can you post your experience on the
>switching process?

I use it (1.99-rc1, which is gone from Portage) for booting my UEFI
(with GPT partition table) motherboard until I can get coreboot running
on it... :-)

Experience wise it's, well, not much difference than old grub, with the
exception of the change in paradigm: you don't edit the config file
directly but instead edit "pre-config" files in order to get a working
solution. For me the default settings work fine (well, I haven't been
able to change the resolution); it finds my installed kernels in /boot
and put's them in the boot list (the boot screen list) together with a
single user version for rescue operations. IMO, it's over-complicated (I
agree with Grant) but if the default settings works (with tweaks) for you...

>Was it difficult?  Easy?  Somewhere between?

Hm... Well, see above...

Canek:
> Me, I want my laptop/desktop computers to have the best resolution
> available from moment zero, even before loading the kernel, and not a

I agree with this sentiment although I think that the video firmware (or
motherboard firmware) should handle this...

> single flicker in my screen until my GNOME 3 is fully loaded. So I'm

Yes, agree again, although I think Gnome (2,3+) is a festering piece of
#%!&... :-)

> gonna play with grub2 (or /firstboot, if it materializes) until it's

Ok, cool. Please share your experiences. I'll try playing (when I can
find the time) with coreboot and FILO:
http://www.coreboot.org/Payloads#FILO

Best regards

Peter K

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 1:11 PM, Alan McKinnon <alan.mckinnon@gmail.com> wrote:
> On Tue, 4 Oct 2011 08:08:16 -0700
> Canek Peláez Valdés <caneko@gmail.com> wrote:
>
>> > At first glace, grub2 looks like a minature Unix installation whose
>> > purpose is to boot a bigger Unix installation.  It's got it's own
>> > init system and it's own set of init scripts.
>>
>> That it's not true. It connects to whatever init system do you have
>> (OpenRC, SysV, systemd, Upstart), and it has scripts to *generate* the
>> config file.
>>
>> The thing is that GRUB2 needs to understand several filesystems to
>> grab the kernel image from. It also wants to be able to use a more
>> interesting resolution than 640x480. This means that it has to
>> reimplement all the code for any filesystem, and all the code for
>> video handling.
>
> Personally, I can't agree with this stance from the grub2 devs.
>
> It's a bootloader. It is visible for 3 seconds at boot time.

Some of us care about those 3 seconds, and the flickering of the
screen when going from bootloader to init splash to X. If you don't
care about those 3 seconds or the flickering, then simply don't use
grub2: keep using grub-legacy or lilo.

> For driving the screen it should just use whatever facilities the
> firmware one layer below it provides.

That's your opinion, and a respectable one. I agree not everybody will
(nor should) care about a pretty boot menu. However, many of us do.

I'm pretty sure when grub2 hits the 2.0 version it will be optional at
./configure time wether to use or not pretty graphics and a lot of
filesystems, or only VGA and ext2, and everything in between.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:

>>> That it's not true. It connects to whatever init system do you have
>>> (OpenRC, SysV, systemd, Upstart),
>>
>> I'm curious: what if you don't have one? ??I use grub-legacy to boot
>> stuff other than Unix.
>
> When I said "it connects", I mean "calls". The same way it calls
> whatever thingy Window uses.

Right.  And what about non-windows, non-Unix systems that don't have
any thingy to call?

>>> and it has scripts to *generate* the config file.
>>>
>>> The thing is that GRUB2 needs to understand several filesystems to
>>> grab the kernel image from.
>>
>> I understand why GRUB2 is complicated. ??It's the statement that it's
>> not complicated that I was disagreeing with.
>>
>>> It also wants to be able to use a more interesting resolution than
>>> 640x480.
>>
>> That I don't understand. It's a bootloader. ??It needs to allow you to
>> pick one of a handfull of choices and boot that choice.
>
> I agree. That's why GRUB2 now is really 1.99, because it's not finished.
>
>>> This means that it has to reimplement all the code for any
>>> filesystem,
>>
>> That part I understand.
>>
>>> and all the code for video handling.
>>
>> I don't really understand the need for that, but I'm somebody who
>> still regularly uses a serial console. ??[Insert the usual "I remember
>> when" grumbling here.]
>
> Then stick with LILO or grub-legacy and root=UUID in your kernel
> command line.

That's the plan for now, but if things go the way they usually do,
grub-legacy will get pulled out from under us before too long and
we'll be forced to either use grub2 or stop whinging and voluteer to
maintain grub-legacy.  :)

-- 
Grant Edwards               grant.b.edwards        Yow! Didn't I buy a 1951
                                  at               Packard from you last March
                              gmail.com            in Cairo?

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Alan McKinnon <alan.mckinnon@gmail.com> wrote:

> No that's a completely different issue.
>
> But the warped thinking that produces it is exactly the same.

QOTW!

-- 
Grant Edwards               grant.b.edwards        Yow! I need to discuss
                                  at               BUY-BACK PROVISIONS
                              gmail.com            with at least six studio
                                                   SLEAZEBALLS!!

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 1:50 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>
>>>> That it's not true. It connects to whatever init system do you have
>>>> (OpenRC, SysV, systemd, Upstart),
>>>
>>> I'm curious: what if you don't have one? ??I use grub-legacy to boot
>>> stuff other than Unix.
>>
>> When I said "it connects", I mean "calls". The same way it calls
>> whatever thingy Window uses.
>
> Right.  And what about non-windows, non-Unix systems that don't have
> any thingy to call?

Then you don't have an operating system.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> On Tue, Oct 4, 2011 at 1:50 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>
>>>>> That it's not true. It connects to whatever init system do you have
>>>>> (OpenRC, SysV, systemd, Upstart),
>>>>
>>>> I'm curious: what if you don't have one? ??I use grub-legacy to boot
>>>> stuff other than Unix.
>>>
>>> When I said "it connects", I mean "calls". The same way it calls
>>> whatever thingy Window uses.
>>
>> Right. ??And what about non-windows, non-Unix systems that don't have
>> any thingy to call?
>
> Then you don't have an operating system.

Yes, I do.  It just doesn't have any sort of "init" system that's
visible from a bootloader. Right now I use grub-legacy to boot
embedded applications written using the eCos RTOS via the el torito
state2.  I take it that won't be something grub2 is capable of doing?

Grub2 can only boot Windows or Unix?

-- 
Grant Edwards               grant.b.edwards        Yow! !  Up ahead!  It's a
                                  at               DONUT HUT!!
                              gmail.com            

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 2:02 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>> On Tue, Oct 4, 2011 at 1:50 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>>
>>>>>> That it's not true. It connects to whatever init system do you have
>>>>>> (OpenRC, SysV, systemd, Upstart),
>>>>>
>>>>> I'm curious: what if you don't have one? ??I use grub-legacy to boot
>>>>> stuff other than Unix.
>>>>
>>>> When I said "it connects", I mean "calls". The same way it calls
>>>> whatever thingy Window uses.
>>>
>>> Right. ??And what about non-windows, non-Unix systems that don't have
>>> any thingy to call?
>>
>> Then you don't have an operating system.
>
> Yes, I do.

Then any boot loader will need to call something to start it.
Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
OpenRC) is simply a program... that the Linux kernel itself executes.
That's the init= command line in the kernel.

The bootloader calls an operating system. The init system (if at all)
that the OS uses doesn't matter: so if you have an operating system,
any bootloader should be able to boot it (bearing things like being
able to understand the filesystem etc.)

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

[gentoo-user] Re: Is grub2 stable and who uses it?

[walt]

On 10/04/2011 07:53 AM, Dale wrote:

> Could this fix the mess with /usr and /var having to be on / or a initramfs?

I'm using grub2 because it fixes a different problem that has always needed an
initramfs--but not the recently lamented separate /var problem.

I have an outboard ESATA disk that I can plug into various machines for making
backups.  If the outboard disk is powered on during boot/reboot, the BIOS will
detect the disks in a different order so that old grub tries to load the boot
sector from the outboard disk instead of the internal one, and fails.

The answer is to let grub2 find the correct disk by checking the UUID of the
*partition table* on each disk, and then load the boot sector from only that
disk without even knowing the /dev/sd* name or the BIOS disk number.

I'm assuming/hoping that the new EFI mechanism will make all of this garbage
obsolete fairly soon.  Anyone here understand the basics of EFI and how it
might relate to these problems?

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:

> Then any boot loader will need to call something to start it.
> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
> OpenRC) is simply a program... that the Linux kernel itself executes.

I know.  What I don't understand is the statement that grub2 calls (or
connects to) the init system.

> That's the init= command line in the kernel.
>
> The bootloader calls an operating system. The init system (if at all)
> that the OS uses doesn't matter: so if you have an operating system,
> any bootloader should be able to boot it (bearing things like being
> able to understand the filesystem etc.)

I know how bootloaders like LILO and grub-legacy work.  What I don't
understand is the statement that grub2 is somehow aware of the booted
OS's init system.

-- 
Grant Edwards               grant.b.edwards        Yow! FROZEN ENTREES may
                                  at               be flung by members of
                              gmail.com            opposing SWANSON SECTS ...

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, walt <w41ter@gmail.com> wrote:

> The answer is to let grub2 find the correct disk by checking the UUID
> of the *partition table* on each disk, and then load the boot sector
> from only that disk without even knowing the /dev/sd* name or the
> BIOS disk number.
>
> I'm assuming/hoping that the new EFI mechanism will make all of this
> garbage obsolete fairly soon.

If Microsoft gets their way, EFI will indeed make all of this
obsolete, since it will (for all practical purposes) prohibit booting
anything except pre-configured factory-certified installations of
MS-Windows.

> Anyone here understand the basics of EFI and how it might relate to
> these problems?

-- 
Grant Edwards               grant.b.edwards        Yow! Didn't I buy a 1951
                                  at               Packard from you last March
                              gmail.com            in Cairo?

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>
>> Then any boot loader will need to call something to start it.
>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
>> OpenRC) is simply a program... that the Linux kernel itself executes.
>
> I know.  What I don't understand is the statement that grub2 calls (or
> connects to) the init system.
>
>> That's the init= command line in the kernel.
>>
>> The bootloader calls an operating system. The init system (if at all)
>> that the OS uses doesn't matter: so if you have an operating system,
>> any bootloader should be able to boot it (bearing things like being
>> able to understand the filesystem etc.)
>
> I know how bootloaders like LILO and grub-legacy work.  What I don't
> understand is the statement that grub2 is somehow aware of the booted
> OS's init system.

Oh. The configuration file of GRUB2 is autogenerated, and this means
that the init=systemd has to be passed to the kernel line.

In that sense, GRUB2 is "aware" of it.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Michael Schreckenbauer]

On Tuesday, 4. October 2011 14:14:24 Canek Peláez Valdés wrote:
> On Tue, Oct 4, 2011 at 2:02 PM, Grant Edwards <grant.b.edwards@gmail.com> 
wrote:
> > On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >> On Tue, Oct 4, 2011 at 1:50 PM, Grant Edwards <grant.b.edwards@gmail.com> 
wrote:
> >>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >>>>>> That it's not true. It connects to whatever init system do you
> >>>>>> have
> >>>>>> (OpenRC, SysV, systemd, Upstart),
> >>>>> 
> >>>>> I'm curious: what if you don't have one? ??I use grub-legacy to
> >>>>> boot
> >>>>> stuff other than Unix.
> >>>> 
> >>>> When I said "it connects", I mean "calls". The same way it calls
> >>>> whatever thingy Window uses.
> >>> 
> >>> Right. ??And what about non-windows, non-Unix systems that don't
> >>> have
> >>> any thingy to call?
> >> 
> >> Then you don't have an operating system.
> > 
> > Yes, I do.
> 
> Then any boot loader will need to call something to start it.
> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
> OpenRC) is simply a program... that the Linux kernel itself executes.
> That's the init= command line in the kernel.

Correct, the *kernel* executes it.

Quoted from an earlier mail in this thread:

"That it's not true. It connects to whatever init system do you have
(OpenRC, SysV, systemd, Upstart)"

The kernel executes the initsystem, the initsystem takes care of the rest. 
Care to explain, why grub2 needs to connect to (or call) the initsystem?

> Regards.

Best,
Michael

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 2:33 PM, Michael Schreckenbauer <grimlog@gmx.de> wrote:
> On Tuesday, 4. October 2011 14:14:24 Canek Peláez Valdés wrote:
>> On Tue, Oct 4, 2011 at 2:02 PM, Grant Edwards <grant.b.edwards@gmail.com>
> wrote:
>> > On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>> >> On Tue, Oct 4, 2011 at 1:50 PM, Grant Edwards <grant.b.edwards@gmail.com>
> wrote:
>> >>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>> >>>>>> That it's not true. It connects to whatever init system do you
>> >>>>>> have
>> >>>>>> (OpenRC, SysV, systemd, Upstart),
>> >>>>>
>> >>>>> I'm curious: what if you don't have one? ??I use grub-legacy to
>> >>>>> boot
>> >>>>> stuff other than Unix.
>> >>>>
>> >>>> When I said "it connects", I mean "calls". The same way it calls
>> >>>> whatever thingy Window uses.
>> >>>
>> >>> Right. ??And what about non-windows, non-Unix systems that don't
>> >>> have
>> >>> any thingy to call?
>> >>
>> >> Then you don't have an operating system.
>> >
>> > Yes, I do.
>>
>> Then any boot loader will need to call something to start it.
>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
>> OpenRC) is simply a program... that the Linux kernel itself executes.
>> That's the init= command line in the kernel.
>
> Correct, the *kernel* executes it.
>
> Quoted from an earlier mail in this thread:
>
> "That it's not true. It connects to whatever init system do you have
> (OpenRC, SysV, systemd, Upstart)"
>
> The kernel executes the initsystem, the initsystem takes care of the rest.
> Care to explain, why grub2 needs to connect to (or call) the initsystem?

It connects via the kernel via init=, as always. Maybe not the best
choice of words, but the important thing is that the statement about
GRUB2 having its "own init system and it's own set of init scripts" is
false. I noted the "connection" between the bootloader and the init
system (via the init= command line) to emphasize that GRUB2 has not
its own init system. Nor init scripts.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Michael Schreckenbauer]

On Tuesday, 4. October 2011 14:46:07 Canek Peláez Valdés wrote:
> On Tue, Oct 4, 2011 at 2:33 PM, Michael Schreckenbauer <grimlog@gmx.de> 
wrote:
> > On Tuesday, 4. October 2011 14:14:24 Canek Peláez Valdés wrote:
> >> On Tue, Oct 4, 2011 at 2:02 PM, Grant Edwards
> >> <grant.b.edwards@gmail.com>> 
> > wrote:
> >> > On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >> >> On Tue, Oct 4, 2011 at 1:50 PM, Grant Edwards
> >> >> <grant.b.edwards@gmail.com>> 
> > wrote:
> >> >>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >> >>>>>> That it's not true. It connects to whatever init system
> >> >>>>>> do you
> >> >>>>>> have
> >> >>>>>> (OpenRC, SysV, systemd, Upstart),
> >> >>>>> 
> >> >>>>> I'm curious: what if you don't have one? ??I use
> >> >>>>> grub-legacy to
> >> >>>>> boot
> >> >>>>> stuff other than Unix.
> >> >>>> 
> >> >>>> When I said "it connects", I mean "calls". The same way it
> >> >>>> calls
> >> >>>> whatever thingy Window uses.
> >> >>> 
> >> >>> Right. ??And what about non-windows, non-Unix systems that
> >> >>> don't
> >> >>> have
> >> >>> any thingy to call?
> >> >> 
> >> >> Then you don't have an operating system.
> >> > 
> >> > Yes, I do.
> >> 
> >> Then any boot loader will need to call something to start it.
> >> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
> >> OpenRC) is simply a program... that the Linux kernel itself executes.
> >> That's the init= command line in the kernel.
> > 
> > Correct, the *kernel* executes it.
> > 
> > Quoted from an earlier mail in this thread:
> > 
> > "That it's not true. It connects to whatever init system do you have
> > (OpenRC, SysV, systemd, Upstart)"
> > 
> > The kernel executes the initsystem, the initsystem takes care of the
> > rest. Care to explain, why grub2 needs to connect to (or call) the
> > initsystem?
> It connects via the kernel via init=, as always. Maybe not the best
> choice of words, but the important thing is that the statement about
> GRUB2 having its "own init system and it's own set of init scripts" is
> false. I noted the "connection" between the bootloader and the init
> system (via the init= command line) to emphasize that GRUB2 has not
> its own init system. Nor init scripts.

Ah, so no connection or call at all :) Thanks for clarifying

> Regards.

Best,
Michael

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>
>>> Then any boot loader will need to call something to start it.
>>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
>>> OpenRC) is simply a program... that the Linux kernel itself executes.
>>
>> I know. ??What I don't understand is the statement that grub2 calls (or
>> connects to) the init system.
>>
>>> That's the init= command line in the kernel.
>>>
>>> The bootloader calls an operating system. The init system (if at all)
>>> that the OS uses doesn't matter: so if you have an operating system,
>>> any bootloader should be able to boot it (bearing things like being
>>> able to understand the filesystem etc.)
>>
>> I know how bootloaders like LILO and grub-legacy work. ??What I don't
>> understand is the statement that grub2 is somehow aware of the booted
>> OS's init system.
>
> Oh. The configuration file of GRUB2 is autogenerated, and this means
> that the init=systemd has to be passed to the kernel line.
>
> In that sense, GRUB2 is "aware" of it.

So to use grub2 you have to replace the normal "init" program that's
started by the kernle as PID#1 with something else?

-- 
Grant Edwards               grant.b.edwards        Yow! Where does it go when
                                  at               you flush?
                              gmail.com            

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Tue, Oct 4, 2011 at 3:14 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>> On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>>
>>>> Then any boot loader will need to call something to start it.
>>>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
>>>> OpenRC) is simply a program... that the Linux kernel itself executes.
>>>
>>> I know. ??What I don't understand is the statement that grub2 calls (or
>>> connects to) the init system.
>>>
>>>> That's the init= command line in the kernel.
>>>>
>>>> The bootloader calls an operating system. The init system (if at all)
>>>> that the OS uses doesn't matter: so if you have an operating system,
>>>> any bootloader should be able to boot it (bearing things like being
>>>> able to understand the filesystem etc.)
>>>
>>> I know how bootloaders like LILO and grub-legacy work. ??What I don't
>>> understand is the statement that grub2 is somehow aware of the booted
>>> OS's init system.
>>
>> Oh. The configuration file of GRUB2 is autogenerated, and this means
>> that the init=systemd has to be passed to the kernel line.
>>
>> In that sense, GRUB2 is "aware" of it.
>
> So to use grub2 you have to replace the normal "init" program that's
> started by the kernle as PID#1 with something else?

No.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 1426 bytes --]

On Tue, 4 Oct 2011 14:35:42 +0000 (UTC), Grant Edwards wrote:

> I've only used it on Ubuntu, and maybe it's just Ubuntu's
> implementation -- but it was both complicated and difficult.  There
> are 10X as many files, and to change anything you edit a whole set of
> configuration files and run a utility that generates _another_ set of
> configuration files.

That's not strictly true. GRUB2 uses only one config file when booting,
grub.cfg, which is analogous to menu.lst. If you want you can edit this
directly. The rest of the files do not live on /boot and are used to
automatically generate grub.cfg if you want them too. This makes life
easy for distro installer writers as they don't need to worry about
scanning the hard disk to see what is installed and creating suitable
menu entries, they just run grub-install. That's why distros now tend to
play nicely with one another, instead of only setting up dual booting for
themselves and Windows.

The reason there are so many more files is because GRUB2 uses modules to
be able to boot from many more devices, such as RAID or LVM. They don't
all end up in /boot.

So it is bigger and more capable/automatable, but you can use it just
like legacy GRUB if you really want to. For most distros, GRUB2 makes a
lot of sense, but many of its capabilities have little relevance to Gentoo.


-- 
Neil Bothwick

"Criminal Lawyer" is a redundancy.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Dale]

Canek Peláez Valdés wrote:
> On Tue, Oct 4, 2011 at 2:33 PM, Michael Schreckenbauer<grimlog@gmx.de>  wrote:
>
>> Correct, the *kernel* executes it.
>>
>> Quoted from an earlier mail in this thread:
>>
>> "That it's not true. It connects to whatever init system do you have
>> (OpenRC, SysV, systemd, Upstart)"
>>
>> The kernel executes the initsystem, the initsystem takes care of the rest.
>> Care to explain, why grub2 needs to connect to (or call) the initsystem?
> It connects via the kernel via init=, as always. Maybe not the best
> choice of words, but the important thing is that the statement about
> GRUB2 having its "own init system and it's own set of init scripts" is
> false. I noted the "connection" between the bootloader and the init
> system (via the init= command line) to emphasize that GRUB2 has not
> its own init system. Nor init scripts.
>
> Regards.

I don't have that on mine.

title Gentoo
kernel (hd0,0)/bzImage-3.0.4-1 root=/dev/sda3

So I guess my grub is ignorant.  lol

Dale

:-)  :-)

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

[-- Attachment #1: Type: text/plain, Size: 1352 bytes --]

El 04/10/2011 17:09, "Dale" <rdalek1967@gmail.com> escribió:
>
> Canek Peláez Valdés wrote:
>>
>> On Tue, Oct 4, 2011 at 2:33 PM, Michael Schreckenbauer<grimlog@gmx.de>
 wrote:
>>
>>> Correct, the *kernel* executes it.
>>>
>>> Quoted from an earlier mail in this thread:
>>>
>>> "That it's not true. It connects to whatever init system do you have
>>> (OpenRC, SysV, systemd, Upstart)"
>>>
>>> The kernel executes the initsystem, the initsystem takes care of the
rest.
>>> Care to explain, why grub2 needs to connect to (or call) the initsystem?
>>
>> It connects via the kernel via init=, as always. Maybe not the best
>> choice of words, but the important thing is that the statement about
>> GRUB2 having its "own init system and it's own set of init scripts" is
>> false. I noted the "connection" between the bootloader and the init
>> system (via the init= command line) to emphasize that GRUB2 has not
>> its own init system. Nor init scripts.
>>
>> Regards.
>
>
> I don't have that on mine.
>
> title Gentoo
> kernel (hd0,0)/bzImage-3.0.4-1 root=/dev/sda3
>
> So I guess my grub is ignorant.  lol

If there is no init= command line argument, /sbin/init is the default. It
has been this way from the very beginning; systemd uses /sbin/systemd to be
able to be installed in parallel with SysV.

Regards.

[-- Attachment #2: Type: text/html, Size: 1816 bytes --]

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Alan McKinnon]

On Tue, 4 Oct 2011 14:32:51 -0700
Canek Peláez Valdés <caneko@gmail.com> wrote:

> On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards
> <grant.b.edwards@gmail.com> wrote:
> > On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >
> >> Then any boot loader will need to call something to start it.
> >> Understand this: any Linux/Unix init system (systemd, SysV,
> >> Upstart, OpenRC) is simply a program... that the Linux kernel
> >> itself executes.
> >
> > I know.  What I don't understand is the statement that grub2 calls
> > (or connects to) the init system.
> >
> >> That's the init= command line in the kernel.
> >>
> >> The bootloader calls an operating system. The init system (if at
> >> all) that the OS uses doesn't matter: so if you have an operating
> >> system, any bootloader should be able to boot it (bearing things
> >> like being able to understand the filesystem etc.)
> >
> > I know how bootloaders like LILO and grub-legacy work.  What I don't
> > understand is the statement that grub2 is somehow aware of the
> > booted OS's init system.
> 
> Oh. The configuration file of GRUB2 is autogenerated, and this means
> that the init=systemd has to be passed to the kernel line.
> 
> In that sense, GRUB2 is "aware" of it.

Possibly what you meant to say is that grub2 is not aware of the OS but
the grub2 installer does.

Like grub and lilo before it, the installer is a Linux app; and can
figure out the correct kernel parameters to use by examining the file
system


-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 1415 bytes --]

On Tue, 04 Oct 2011 23:33:34 +0200, Michael Schreckenbauer wrote:

> > Then any boot loader will need to call something to start it.
> > Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
> > OpenRC) is simply a program... that the Linux kernel itself executes.
> > That's the init= command line in the kernel.  
> 
> Correct, the *kernel* executes it.
> 
> Quoted from an earlier mail in this thread:
> 
> "That it's not true. It connects to whatever init system do you have
> (OpenRC, SysV, systemd, Upstart)"
> 
> The kernel executes the initsystem, the initsystem takes care of the
> rest. Care to explain, why grub2 needs to connect to (or call) the
> initsystem?

The confusion is caused by using grub to describe two different modes of
operation. the bootloader itself does not need access to anything but the
kernel and the initramfs , if used. The grub program, run from Linux to
set up the bootloader, does need access to your filesystem to be able to
do its job. That is not required for booting, which is why the code is
not in /boot.

The GRUB2 bootloader works in much the same way as the old one, with the
menu entry format being quite similar too. The difference is in the
automation stuff that non-genkernel or other distro users wouldn't be
interested in anyway.


-- 
Neil Bothwick

If you consult enough experts, you can confirm any opinion.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 674 bytes --]

On Tue, 4 Oct 2011 22:11:00 +0200, Alan McKinnon wrote:

> > The thing is that GRUB2 needs to understand several filesystems to
> > grab the kernel image from. It also wants to be able to use a more
> > interesting resolution than 640x480. This means that it has to
> > reimplement all the code for any filesystem, and all the code for
> > video handling.  
> 
> Personally, I can't agree with this stance from the grub2 devs.
> 
> It's a bootloader. It is visible for 3 seconds at boot time. 

I'd agree with you there, but I suppose the glossy distros want a glossy
boot screen.


-- 
Neil Bothwick

Facts are stubborn, but statistics are more pliable

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> On Tue, Oct 4, 2011 at 3:14 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>> On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>>>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>>>
>>>>> Then any boot loader will need to call something to start it.
>>>>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
>>>>> OpenRC) is simply a program... that the Linux kernel itself executes.
>>>>
>>>> I know. ??What I don't understand is the statement that grub2 calls (or
>>>> connects to) the init system.
>>>>
>>>>> That's the init= command line in the kernel.
>>>>>
>>>>> The bootloader calls an operating system. The init system (if at all)
>>>>> that the OS uses doesn't matter: so if you have an operating system,
>>>>> any bootloader should be able to boot it (bearing things like being
>>>>> able to understand the filesystem etc.)
>>>>
>>>> I know how bootloaders like LILO and grub-legacy work. ??What I don't
>>>> understand is the statement that grub2 is somehow aware of the booted
>>>> OS's init system.
>>>
>>> Oh. The configuration file of GRUB2 is autogenerated, and this means
>>> that the init=systemd has to be passed to the kernel line.
>>>
>>> In that sense, GRUB2 is "aware" of it.
>>
>> So to use grub2 you have to replace the normal "init" program that's
>> started by the kernle as PID#1 with something else?
>
> No.

I give up.  I've absolutely no idea what grub2 has to do with the OS's
init system, and none of what you've written makes any sense to me.

-- 
Grant Edwards               grant.b.edwards        Yow! We are now enjoying
                                  at               total mutual interaction in
                              gmail.com            an imaginary hot tub ...

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Pandu Poluan]

[-- Attachment #1: Type: text/plain, Size: 2305 bytes --]

On Oct 5, 2011 8:59 PM, "Grant Edwards" <grant.b.edwards@gmail.com> wrote:
>
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> > On Tue, Oct 4, 2011 at 3:14 PM, Grant Edwards <grant.b.edwards@gmail.com>
wrote:
> >> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >>> On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards <
grant.b.edwards@gmail.com> wrote:
> >>>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >>>>
> >>>>> Then any boot loader will need to call something to start it.
> >>>>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
> >>>>> OpenRC) is simply a program... that the Linux kernel itself
executes.
> >>>>
> >>>> I know. ??What I don't understand is the statement that grub2 calls
(or
> >>>> connects to) the init system.
> >>>>
> >>>>> That's the init= command line in the kernel.
> >>>>>
> >>>>> The bootloader calls an operating system. The init system (if at
all)
> >>>>> that the OS uses doesn't matter: so if you have an operating system,
> >>>>> any bootloader should be able to boot it (bearing things like being
> >>>>> able to understand the filesystem etc.)
> >>>>
> >>>> I know how bootloaders like LILO and grub-legacy work. ??What I don't
> >>>> understand is the statement that grub2 is somehow aware of the booted
> >>>> OS's init system.
> >>>
> >>> Oh. The configuration file of GRUB2 is autogenerated, and this means
> >>> that the init=systemd has to be passed to the kernel line.
> >>>
> >>> In that sense, GRUB2 is "aware" of it.
> >>
> >> So to use grub2 you have to replace the normal "init" program that's
> >> started by the kernle as PID#1 with something else?
> >
> > No.
>
> I give up.  I've absolutely no idea what grub2 has to do with the OS's
> init system, and none of what you've written makes any sense to me.

I think what he meant was:

The *installer* portion of grub2 is aware of which pid#0 is running when it
auto-creates the bootloader's configuration. That pid#0 is passed on to the
kernel by the bootloader.

The *bootloader* portion of grub2 don't know and don't care what is being
used as pid#0 by the OS. All it knows is that the installer portion has
specified something to be passed to the OS. And that's what it does, without
understanding anything about pid#0.

rgds,

[-- Attachment #2: Type: text/html, Size: 3366 bytes --]

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Sebastian Beßler]

[-- Attachment #1: Type: text/plain, Size: 520 bytes --]

Am 05.10.2011 15:55, schrieb Grant Edwards:

> I give up.  I've absolutely no idea what grub2 has to do with the OS's
> init system, and none of what you've written makes any sense to me.

It has NOTHING to do with it, or not more or less then lilo or grub1 or
any other bootloader. But the automagic of grub2 at install/setup time
is able to see which init system is used in the linuxes it reach on the
system and configure the init= kernel option for you.
Thats all folks!

Greetings

Sebastian Beßler


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 900 bytes --]

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-05, Pandu Poluan <pandu@poluan.info> wrote:

>> I give up.  I've absolutely no idea what grub2 has to do with the OS's
>> init system, and none of what you've written makes any sense to me.
>
> I think what he meant was:

I assume you mean PID#1 (typically /sbin/init).  On Unixes with PID#0,
it's usually the swapper or scheduler task that's internal to the
kernel.

> The *installer* portion of grub2 is aware of which pid#0 is running
> when it auto-creates the bootloader's configuration. That pid#0 is
> passed on to the kernel by the bootloader.

OK.  I that I understand.  It seems a bit redundant to me: I've been
running Linux since the 0.99 days and never had to pass init= to a
kernel.  But, I guess it won't hurt anything...

> The *bootloader* portion of grub2 don't know and don't care what is
> being used as pid#0 by the OS. All it knows is that the installer
> portion has specified something to be passed to the OS. And that's
> what it does, without understanding anything about pid#0.

And the set of init scripts that belong to grub2 are just to try to
auto-magically generate the config file?

-- 
Grant Edwards               grant.b.edwards        Yow! They collapsed
                                  at               ... like nuns in the
                              gmail.com            street ... they had no
                                                   teen appeal!

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Alan McKinnon]

On Wed, 5 Oct 2011 14:46:03 +0000 (UTC)
Grant Edwards <grant.b.edwards@gmail.com> wrote:

> > The *installer* portion of grub2 is aware of which pid#0 is running
> > when it auto-creates the bootloader's configuration. That pid#0 is
> > passed on to the kernel by the bootloader.  
> 
> OK.  I that I understand.  It seems a bit redundant to me: I've been
> running Linux since the 0.99 days and never had to pass init= to a
> kernel.  But, I guess it won't hurt anything.


It is indeed redundant and harmless. You no doubt already know the
kernel's logic for launching the first userspace app - three paths are
hardcoded and searched in sequence, first one found is launched. The
third one is /sbin/init

It makes for a wonderful prank, add "init=bin/bash" to someone's
menu.lst and watch them get confused at next reboot :-)

I suppose grub2 could search for and include a redundant init parameter
for the sake of consistency with cases where a non-standard init was in
use

 


-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 855 bytes --]

On Wed, 5 Oct 2011 14:46:03 +0000 (UTC), Grant Edwards wrote:

>> The *installer* portion of grub2 is aware of which pid#0 is running
> > when it auto-creates the bootloader's configuration. That pid#0 is
> > passed on to the kernel by the bootloader.  
> 
> OK.  I that I understand.  It seems a bit redundant to me: I've been
> running Linux since the 0.99 days and never had to pass init= to a
> kernel.  But, I guess it won't hurt anything...

That's because you are using the standard, hard-coded, init. GRUB2 is
trying to cope with all use cases, so it checks to see whether a
different system is in use and configures the boot menu accordingly.


-- 
Neil Bothwick

GOTO: (n.) an efficient and general way of controlling a program, much
despised by academics and others whose brains have been ruined by
overexposure to Pascal.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[ny6p01]

On Tue, Oct 04, 2011 at 08:56:54AM -0400, Michael Orlitzky wrote:
> On 10/04/2011 06:16 AM, Pandu Poluan wrote:
> > 
> > On Oct 4, 2011 5:10 PM, "Neil Bothwick" <neil@digimed.co.uk
> > <mailto:neil@digimed.co.uk>> wrote:
> >>
> >> On Tue, 04 Oct 2011 04:49:56 -0500, Dale wrote:
> >>
> >> > Subject line says it pretty well.  Is grub2 stable, who uses it and can
> >> > you post your experience on the switching process?  Was it difficult?
> >>
> >> I use it on my netbook, which I admittedly don't boot more than a couple
> >> of times a month. It's stable, I can't comment on the switching process
> >> as I used GRUB2 from the start with this machine, it seemed a good time
> >> to get to grips with it.
> >>
> >> GRUB2 is neither complicated nor difficult, but it is different. If you
> >> try to think in terms of legacy GRUB, you will have more problems than if
> >> you approach is as learning a new system.
> >>
> > 
> > Kind of tangential...
> > 
> > Why does Gentoo still 'standardize' on grub instead of going forward
> > with grub2?
> 
> Grub2 is weird (coming from anything that isn't grub2), and if you mess
> up the upgrade, you can't boot.
> 
> It's a support nightmare.
> 

Wow - what an improvement. :|

Terry

Re: [gentoo-user] Is grub2 stable and who uses it?

[Jonas de Buhr]

Am Wed, 5 Oct 2011 09:20:07 -0700
schrieb ny6p01@gmail.com:

> > > Why does Gentoo still 'standardize' on grub instead of going
> > > forward with grub2?
> > 
> > Grub2 is weird (coming from anything that isn't grub2), and if you
> > mess up the upgrade, you can't boot.
> > 
> > It's a support nightmare.
> > 
> 
> Wow - what an improvement. :|

if you mess up the bootloader you can't boot. thats trivial, easy to
solve and not at all related to grub2.
im not defending grub2 (the little i know about it is what has been
said here) but sometimes things indeed need to change in order to
improve. 

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Wed, Oct 5, 2011 at 6:55 AM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>> On Tue, Oct 4, 2011 at 3:14 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>>> On Tue, Oct 4, 2011 at 2:24 PM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
>>>>> On 2011-10-04, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>>>>>
>>>>>> Then any boot loader will need to call something to start it.
>>>>>> Understand this: any Linux/Unix init system (systemd, SysV, Upstart,
>>>>>> OpenRC) is simply a program... that the Linux kernel itself executes.
>>>>>
>>>>> I know. ??What I don't understand is the statement that grub2 calls (or
>>>>> connects to) the init system.
>>>>>
>>>>>> That's the init= command line in the kernel.
>>>>>>
>>>>>> The bootloader calls an operating system. The init system (if at all)
>>>>>> that the OS uses doesn't matter: so if you have an operating system,
>>>>>> any bootloader should be able to boot it (bearing things like being
>>>>>> able to understand the filesystem etc.)
>>>>>
>>>>> I know how bootloaders like LILO and grub-legacy work. ??What I don't
>>>>> understand is the statement that grub2 is somehow aware of the booted
>>>>> OS's init system.
>>>>
>>>> Oh. The configuration file of GRUB2 is autogenerated, and this means
>>>> that the init=systemd has to be passed to the kernel line.
>>>>
>>>> In that sense, GRUB2 is "aware" of it.
>>>
>>> So to use grub2 you have to replace the normal "init" program that's
>>> started by the kernle as PID#1 with something else?
>>
>> No.
>
> I give up.  I've absolutely no idea what grub2 has to do with the OS's
> init system, and none of what you've written makes any sense to me.

Others explain it too. The GRUB2 bootloader has nothing to do with the
init system, except for passing thi init= command line, and it is not
required that it does.

The GRUB2 userspace, used to generate the config file, can be made
aware of what init system is to be used, making the bootloader to pass
the init= command line to the kernel.

That is all.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Wed, Oct 5, 2011 at 7:46 AM, Grant Edwards <grant.b.edwards@gmail.com> wrote:
> On 2011-10-05, Pandu Poluan <pandu@poluan.info> wrote:
>
>>> I give up.  I've absolutely no idea what grub2 has to do with the OS's
>>> init system, and none of what you've written makes any sense to me.
>>
>> I think what he meant was:
>
> I assume you mean PID#1 (typically /sbin/init).  On Unixes with PID#0,
> it's usually the swapper or scheduler task that's internal to the
> kernel.
>
>> The *installer* portion of grub2 is aware of which pid#0 is running
>> when it auto-creates the bootloader's configuration. That pid#0 is
>> passed on to the kernel by the bootloader.
>
> OK.  I that I understand.  It seems a bit redundant to me: I've been
> running Linux since the 0.99 days and never had to pass init= to a
> kernel.  But, I guess it won't hurt anything...
>
>> The *bootloader* portion of grub2 don't know and don't care what is
>> being used as pid#0 by the OS. All it knows is that the installer
>> portion has specified something to be passed to the OS. And that's
>> what it does, without understanding anything about pid#0.
>
> And the set of init scripts that belong to grub2 are just to try to
> auto-magically generate the config file?

With options from /etc/default/grub, yes. But please stop calling the
files in /etc/grub.d "init scripts". That's the whole reason I dragged
the init systems into the discussion: you said that GRUB2 "got it's
own initsystem and it's own set of init scripts."

And it's simply not true. Maybe with the best of intentions, but
that's disinformation.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

[gentoo-user] Re: Is grub2 stable and who uses it?

[Grant Edwards]

On 2011-10-05, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:

>> And the set of init scripts that belong to grub2 are just to try to
>> auto-magically generate the config file?
>
> With options from /etc/default/grub, yes. But please stop calling the
> files in /etc/grub.d "init scripts".

I'm not calling those "init scripts".  I'm referring to 

 /etc/init.d/grub-common

That's an executable /bin/sh shell script.  Don't know what that is
called if not an "init script".

And then there are these (also /bin/sh scripts):
 
 /etc/grub.d/00_header
 /etc/grub.d/05_debian_theme
 /etc/grub.d/10_linux
 /etc/grub.d/20_memtest86+
 /etc/grub.d/30_os-prober
 /etc/grub.d/40_custom

I assumed these were also some sort of init scripts, but I don't
really know when they get executed.

> That's the whole reason I dragged the init systems into the
> discussion: you said that GRUB2 "got it's own initsystem and it's own
> set of init scripts."

You forgot the part where I said "at first glance under Ubuntu, it
appears that" or somesuch.

> And it's simply not true. Maybe with the best of intentions, but
> that's disinformation.

To me, /etc/init.d/grub-common is an init script.

-- 
Grant Edwards               grant.b.edwards        Yow! Excuse me, but didn't
                                  at               I tell you there's NO HOPE
                              gmail.com            for the survival of OFFSET
                                                   PRINTING?

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Canek Peláez Valdés]

On Wed, Oct 5, 2011 at 11:42 AM, Grant Edwards
<grant.b.edwards@gmail.com> wrote:
> On 2011-10-05, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
>
>>> And the set of init scripts that belong to grub2 are just to try to
>>> auto-magically generate the config file?
>>
>> With options from /etc/default/grub, yes. But please stop calling the
>> files in /etc/grub.d "init scripts".
>
> I'm not calling those "init scripts".  I'm referring to
>
>  /etc/init.d/grub-common

I don't have that file, and it's not because Gentoo removes it: it was
probably added by the Ubuntu developers.

> That's an executable /bin/sh shell script.  Don't know what that is
> called if not an "init script".

But it's not part of GRUB2. If you had checked the project sources, or
the Gentoo ebuild, you would have realized that it is not a file from
the project.

> And then there are these (also /bin/sh scripts):
>
>  /etc/grub.d/00_header
>  /etc/grub.d/05_debian_theme
>  /etc/grub.d/10_linux
>  /etc/grub.d/20_memtest86+
>  /etc/grub.d/30_os-prober
>  /etc/grub.d/40_custom
>
> I assumed these were also some sort of init scripts, but I don't
> really know when they get executed.

That's the problem: you didn't know how the thing worked, and jumped
to conclusions.

>> That's the whole reason I dragged the init systems into the
>> discussion: you said that GRUB2 "got it's own initsystem and it's own
>> set of init scripts."
>
> You forgot the part where I said "at first glance under Ubuntu, it
> appears that" or somesuch.

You said that, but your next sentence was "It's got it's own init
system and it's own set of init scripts", unequivocally.

>> And it's simply not true. Maybe with the best of intentions, but
>> that's disinformation.
>
> To me, /etc/init.d/grub-common is an init script.

Maybe in Ubunt (and maybe not: distros this days throw every kind of
scripts in /etc/init.d, and Gentoo does this too, BTW), but again you
only took a quick look at how it's set in another distro, and jumped
to say that the project as a whole (and not the config from a
particular distro) "got it's own init system and it's own set of init
scripts". To me, that's the definition of spreading disinformation:
not looking for all the info, and stating that such and such is or is
not when it's simply not true.

It's the same history as the myth that /var will not longer be able to
be on its own partition: it keeps popping up in many threads, and it's
also simply not true.

Again, I don't think you did it on purpose with the intention of smear
GRUB2 (that was my "with the best of intentions" part), but *it is*
disinformation.

To finish: GRUB2 does not need or have init scripts, it doesn't have
it's own init system, and if your setup works with GRUB, it will work
in GRUB2, but you will probably need to learn a new way to configure
it. The other way around is not true: GRUB will not support all the
setups that GRUB2 will, unless someone steps up and writes the code
for it.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Alan McKinnon]

On Wed, 5 Oct 2011 12:10:45 -0700
Canek Peláez Valdés <caneko@gmail.com> wrote:

> On Wed, Oct 5, 2011 at 11:42 AM, Grant Edwards
> <grant.b.edwards@gmail.com> wrote:
> > On 2011-10-05, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:
> >
> >>> And the set of init scripts that belong to grub2 are just to try
> >>> to auto-magically generate the config file?
> >>
> >> With options from /etc/default/grub, yes. But please stop calling
> >> the files in /etc/grub.d "init scripts".
> >
> > I'm not calling those "init scripts".  I'm referring to
> >
> >  /etc/init.d/grub-common
> 
> I don't have that file, and it's not because Gentoo removes it: it was
> probably added by the Ubuntu developers.

True:

! /bin/sh
### BEGIN INIT INFO
# Provides:          grub-common
# Required-Start:    $all
# Required-Stop:
# Default-Start:     2 3 4 5
# Default-Stop:
# Short-Description: Record successful boot for GRUB
# Description:       GRUB displays the boot menu at the next boot if it
#                    believes that the previous boot failed. This script
#                    informs it that the system booted successfully.
### END INIT INFO




-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Is grub2 stable and who uses it?

[Peter Humphrey]

[-- Attachment #1: Type: text/plain, Size: 559 bytes --]

On Wednesday 05 October 2011 17:47:21 Jonas de Buhr wrote:

> sometimes things indeed need to change in order to improve.

I remember things being improved. While I was in my 50s I was continually 
faced with youngsters' ideas for improving the company's methods. Stupid, 
every one. When challenged, they couldn't say how their proposed new 
"solutions" would lead to specified gains by anybody, but the changes were 
forced through anyway. This isn't get-up-and-go; it's I've-got-to-make-my-
mark.

Pathetic.

-- 
Rgds
Peter		Linux Counter 5290, 1994-04-23

[-- Attachment #2: Type: text/html, Size: 2738 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 1060 bytes --]

On Thu, 6 Oct 2011 00:18:49 +0100, Peter Humphrey wrote:

> I remember things being improved. While I was in my 50s I was
> continually faced with youngsters' ideas for improving the company's
> methods. Stupid, every one. When challenged, they couldn't say how
> their proposed new "solutions" would lead to specified gains by
> anybody, but the changes were forced through anyway. This isn't
> get-up-and-go; it's I've-got-to-make-my- mark.
> 
> Pathetic.

But not comparable. The reasons for the changes in GRUB2 have been given.

I may not like how they have implemented everything, splitting the
settings between config files in two separate directories means you
always look in the wrong place first, but I understand the need for the
changes even if I don't need the vast majority of them.

Problems with upgrading from GRUB1 to GRUB2 are irrelevant. If you have a
working legacy installation, there is absolutely no reason to change
beyond curiosity.


-- 
Neil Bothwick

Barnum was wrong....it's more like every 30 seconds!

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[gentoo-user] OT: change and improvement (was: Is grub2 stable and who uses it?)

[Jonas de Buhr]

Am Thu, 6 Oct 2011 00:18:49 +0100
schrieb Peter Humphrey <peter@humphrey.ukfsn.org>:

> On Wednesday 05 October 2011 17:47:21 Jonas de Buhr wrote:
> 
> > sometimes things indeed need to change in order to improve.
> 
> I remember things being improved. While I was in my 50s I was
> continually faced with youngsters' ideas for improving the company's
> methods. Stupid, every one. When challenged, they couldn't say how
> their proposed new "solutions" would lead to specified gains by
> anybody, but the changes were forced through anyway. This isn't
> get-up-and-go; it's I've-got-to-make-my- mark.

i never said change equals improvement or change for the sake of change
is a good thing. and i pointed out that i cannot judge this for grub2
since i don't know it. 
but the reverse is true too - rejecting every change just because its
not "how we always did it" potentially keeps you from making important
development. 

most of the "oh it's so weird"-whining often comes from just not being
used to it. flip your door lock upside down - you'll hate it with
passion for a week and then you won't even notice. flip it again and
the process will repeat.

at your age you should be able to look at things with a little more
distance and insight instead of ripping statements out of context and
insulting people.

that said, even without context, that statement is still true. you just
derived wrong statements from it using flawed logic.

/jonas

Re: [gentoo-user] Is grub2 stable and who uses it?

[Alan McKinnon]

On Thu, 6 Oct 2011 09:12:37 +0100
Neil Bothwick <neil@digimed.co.uk> wrote:

> On Thu, 6 Oct 2011 00:18:49 +0100, Peter Humphrey wrote:
> 
> > I remember things being improved. While I was in my 50s I was
> > continually faced with youngsters' ideas for improving the company's
> > methods. Stupid, every one. When challenged, they couldn't say how
> > their proposed new "solutions" would lead to specified gains by
> > anybody, but the changes were forced through anyway. This isn't
> > get-up-and-go; it's I've-got-to-make-my- mark.
> > 
> > Pathetic.
> 
> But not comparable. The reasons for the changes in GRUB2 have been
> given.
> 
> I may not like how they have implemented everything, splitting the
> settings between config files in two separate directories means you
> always look in the wrong place first, but I understand the need for
> the changes even if I don't need the vast majority of them.
> 
> Problems with upgrading from GRUB1 to GRUB2 are irrelevant. If you
> have a working legacy installation, there is absolutely no reason to
> change beyond curiosity.


Well the only constant is change, right?

The trick is to spot the difference between change for the sake of
change and change that does make sense. This usually means getting
inside someone's head, which makes life fun. Doubly so if the change
proposer works in sales....


-- 
Alan McKinnnon
alan.mckinnon@gmail.com

[gentoo-user] Re: Is grub2 stable and who uses it?

[Harry Putnam]

Grant Edwards <grant.b.edwards@gmail.com> writes:

> I've only used it on Ubuntu, and maybe it's just Ubuntu's
> implementation -- but it was both complicated and difficult.  There
> are 10X as many files, and to change anything you edit a whole set of
> configuration files and run a utility that generates _another_ set of
> configuration files.
>
> Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.

100 % agreement here.  I'm running Debian which has moved over to
grub2 in the newest (wheezy) install media.

Here is just a factual count of files to bare out the point.

  sudo mount /boot
  find /boot -type f|wc -l

   225
  
  225 files in boot.  And that isn't all that are involved, there
  are others elsewhere on the file system.

Just for grub users info a real list is inlined at the end. 

In truth, I've only had to make one small edit (It was very esoteric
and hard to find info about).

I wanted to boot to console which required me to change one line in a
file.  /etc/default/grub
 (original line)
  GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"
 (edited line)
  GRUB_CMDLINE_LINUX_DEFAULT="quiet text"

It caused some kind of grief and came up as error on boot messages but
still worked.

To avoid unforeseen problems I put it back to original state and just
took `gdm' out of play (named gdm3 on debian). I then `startx' into
fluxbox.  

However to edit the boot process you make changes to certain files
like the one above and then run `update-grub' which generates the
files in /boot  or at least some of them.

/boot/grub/grub.cfg itself is 111 lines.  Its not the most complex
script going but for me it would take some serious study for an hour
or more to figure out what is happening in it.  But of course you are
not supposed to edit grub.cfg directly.

It is orders of magnitude more complicated in my opinion... I'm not
sure what the advantages are supposed to be.

Also, its not clear how one would install a new kernel.  Where to put
the information and so forth.

-------        ---------       ---=---       ---------      --------

find /boot -type f

/boot/System.map-3.0.0-1-686-pae
/boot/config-3.0.0-1-686-pae
/boot/grub/device.map
/boot/grub/915resolution.mod
/boot/grub/acpi.mod
/boot/grub/affs.mod
/boot/grub/afs.mod
/boot/grub/afs_be.mod
/boot/grub/aout.mod
/boot/grub/at_keyboard.mod
/boot/grub/ata.mod
/boot/grub/ata_pthru.mod
/boot/grub/befs.mod
/boot/grub/befs_be.mod
/boot/grub/biosdisk.mod
/boot/grub/bitmap.mod
/boot/grub/bitmap_scale.mod
/boot/grub/blocklist.mod
/boot/grub/boot.mod
/boot/grub/bsd.mod
/boot/grub/btrfs.mod
/boot/grub/bufio.mod
/boot/grub/cat.mod
/boot/grub/chain.mod
/boot/grub/cmostest.mod
/boot/grub/cmp.mod
/boot/grub/configfile.mod
/boot/grub/cpio.mod
/boot/grub/cpuid.mod
/boot/grub/crypto.mod
/boot/grub/cs5536.mod
/boot/grub/date.mod
/boot/grub/datehook.mod
/boot/grub/datetime.mod
/boot/grub/dm_nv.mod
/boot/grub/drivemap.mod
/boot/grub/echo.mod
/boot/grub/efiemu.mod
/boot/grub/elf.mod
/boot/grub/example_functional_test.mod
/boot/grub/ext2.mod
/boot/grub/extcmd.mod
/boot/grub/fat.mod
/boot/grub/font.mod
/boot/grub/fshelp.mod
/boot/grub/functional_test.mod
/boot/grub/gcry_arcfour.mod
/boot/grub/gcry_blowfish.mod
/boot/grub/gcry_camellia.mod
/boot/grub/gcry_cast5.mod
/boot/grub/gcry_crc.mod
/boot/grub/gcry_des.mod
/boot/grub/gcry_md4.mod
/boot/grub/gcry_md5.mod
/boot/grub/gcry_rfc2268.mod
/boot/grub/gcry_rijndael.mod
/boot/grub/gcry_rmd160.mod
/boot/grub/gcry_seed.mod
/boot/grub/gcry_serpent.mod
/boot/grub/gcry_sha1.mod
/boot/grub/gcry_sha256.mod
/boot/grub/gcry_sha512.mod
/boot/grub/gcry_tiger.mod
/boot/grub/gcry_twofish.mod
/boot/grub/gcry_whirlpool.mod
/boot/grub/gettext.mod
/boot/grub/gfxmenu.mod
/boot/grub/gfxterm.mod
/boot/grub/gptsync.mod
/boot/grub/gzio.mod
/boot/grub/halt.mod
/boot/grub/hashsum.mod
/boot/grub/hdparm.mod
/boot/grub/hello.mod
/boot/grub/help.mod
/boot/grub/hexdump.mod
/boot/grub/hfs.mod
/boot/grub/hfsplus.mod
/boot/grub/iorw.mod
/boot/grub/iso9660.mod
/boot/grub/jfs.mod
/boot/grub/jpeg.mod
/boot/grub/keylayouts.mod
/boot/grub/keystatus.mod
/boot/grub/legacycfg.mod
/boot/grub/linux.mod
/boot/grub/linux16.mod
/boot/grub/loadenv.mod
/boot/grub/loopback.mod
/boot/grub/ls.mod
/boot/grub/lsacpi.mod
/boot/grub/lsapm.mod
/boot/grub/lsmmap.mod
/boot/grub/lspci.mod
/boot/grub/lvm.mod
/boot/grub/mdraid09.mod
/boot/grub/mdraid1x.mod
/boot/grub/memdisk.mod
/boot/grub/memrw.mod
/boot/grub/minicmd.mod
/boot/grub/mmap.mod
/boot/grub/minix.mod
/boot/grub/minix2.mod
/boot/grub/msdospart.mod
/boot/grub/multiboot.mod
/boot/grub/multiboot2.mod
/boot/grub/nilfs2.mod
/boot/grub/normal.mod
/boot/grub/ntfs.mod
/boot/grub/ntfscomp.mod
/boot/grub/ntldr.mod
/boot/grub/ohci.mod
/boot/grub/part_acorn.mod
/boot/grub/part_amiga.mod
/boot/grub/part_apple.mod
/boot/grub/part_bsd.mod
/boot/grub/part_gpt.mod
/boot/grub/part_msdos.mod
/boot/grub/part_sun.mod
/boot/grub/part_sunpc.mod
/boot/grub/parttool.mod
/boot/grub/password.mod
/boot/grub/password_pbkdf2.mod
/boot/grub/pbkdf2.mod
/boot/grub/pci.mod
/boot/grub/play.mod
/boot/grub/png.mod
/boot/grub/probe.mod
/boot/grub/pxe.mod
/boot/grub/pxecmd.mod
/boot/grub/raid.mod
/boot/grub/raid5rec.mod
/boot/grub/raid6rec.mod
/boot/grub/read.mod
/boot/grub/reboot.mod
/boot/grub/regexp.mod
/boot/grub/reiserfs.mod
/boot/grub/relocator.mod
/boot/grub/scsi.mod
/boot/grub/search.mod
/boot/grub/search_fs_file.mod
/boot/grub/search_fs_uuid.mod
/boot/grub/search_label.mod
/boot/grub/sendkey.mod
/boot/grub/serial.mod
/boot/grub/setjmp.mod
/boot/grub/setpci.mod
/boot/grub/sfs.mod
/boot/grub/sleep.mod
/boot/grub/squash4.mod
/boot/grub/tar.mod
/boot/grub/terminal.mod
/boot/grub/terminfo.mod
/boot/grub/test.mod
/boot/grub/test_blockarg.mod
/boot/grub/testload.mod
/boot/grub/tga.mod
/boot/grub/trig.mod
/boot/grub/true.mod
/boot/grub/udf.mod
/boot/grub/ufs1.mod
/boot/grub/ufs2.mod
/boot/grub/uhci.mod
/boot/grub/usb.mod
/boot/grub/usb_keyboard.mod
/boot/grub/usbms.mod
/boot/grub/usbserial_common.mod
/boot/grub/usbserial_ftdi.mod
/boot/grub/usbserial_pl2303.mod
/boot/grub/usbtest.mod
/boot/grub/vbe.mod
/boot/grub/vga.mod
/boot/grub/vga_text.mod
/boot/grub/video.mod
/boot/grub/video_bochs.mod
/boot/grub/video_cirrus.mod
/boot/grub/video_fb.mod
/boot/grub/videoinfo.mod
/boot/grub/videotest.mod
/boot/grub/xfs.mod
/boot/grub/xnu.mod
/boot/grub/xnu_uuid.mod
/boot/grub/xzio.mod
/boot/grub/zfs.mod
/boot/grub/zfsinfo.mod
/boot/grub/command.lst
/boot/grub/crypto.lst
/boot/grub/fs.lst
/boot/grub/moddep.lst
/boot/grub/partmap.lst
/boot/grub/parttool.lst
/boot/grub/terminal.lst
/boot/grub/video.lst
/boot/grub/boot.img
/boot/grub/cdboot.img
/boot/grub/diskboot.img
/boot/grub/g2hdr.img
/boot/grub/grldr.img
/boot/grub/kernel.img
/boot/grub/lnxboot.img
/boot/grub/pxeboot.img
/boot/grub/efiemu32.o
/boot/grub/efiemu64.o
/boot/grub/locale/ast.mo
/boot/grub/locale/ca.mo
/boot/grub/locale/da.mo
/boot/grub/locale/de.mo
/boot/grub/locale/fi.mo
/boot/grub/locale/fr.mo
/boot/grub/locale/hu.mo
/boot/grub/locale/id.mo
/boot/grub/locale/it.mo
/boot/grub/locale/ko.mo
/boot/grub/locale/nl.mo
/boot/grub/locale/pl.mo
/boot/grub/locale/ru.mo
/boot/grub/locale/sv.mo
/boot/grub/locale/uk.mo
/boot/grub/locale/vi.mo
/boot/grub/locale/zh_CN.mo
/boot/grub/core.img
/boot/grub/grubenv
/boot/grub/grub.cfg
/boot/vmlinuz-3.0.0-1-686-pae
/boot/initrd.img-3.0.0-1-686-pae

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Tanstaafl]

Ugh....

Guess if Gentoo ever removes Grub1 I'll have to switch to Lilo or 
something else - I loathe complicated, especially when there is no good 
reason...

On 2011-10-06 8:51 AM, Harry Putnam <reader@newsguy.com> wrote:
> Grant Edwards<grant.b.edwards@gmail.com>  writes:
>
>> I've only used it on Ubuntu, and maybe it's just Ubuntu's
>> implementation -- but it was both complicated and difficult.  There
>> are 10X as many files, and to change anything you edit a whole set of
>> configuration files and run a utility that generates _another_ set of
>> configuration files.
>>
>> Compared to "vi /boot/grub/menu.lst; reboot", that's complicated.
>
> 100 % agreement here.  I'm running Debian which has moved over to
> grub2 in the newest (wheezy) install media.
>
> Here is just a factual count of files to bare out the point.
>
>    sudo mount /boot
>    find /boot -type f|wc -l
>
>     225
>
>    225 files in boot.  And that isn't all that are involved, there
>    are others elsewhere on the file system.
>
> Just for grub users info a real list is inlined at the end.
>
> In truth, I've only had to make one small edit (It was very esoteric
> and hard to find info about).
>
> I wanted to boot to console which required me to change one line in a
> file.  /etc/default/grub
>   (original line)
>    GRUB_CMDLINE_LINUX_DEFAULT="quiet splash"
>   (edited line)
>    GRUB_CMDLINE_LINUX_DEFAULT="quiet text"
>
> It caused some kind of grief and came up as error on boot messages but
> still worked.
>
> To avoid unforeseen problems I put it back to original state and just
> took `gdm' out of play (named gdm3 on debian). I then `startx' into
> fluxbox.
>
> However to edit the boot process you make changes to certain files
> like the one above and then run `update-grub' which generates the
> files in /boot  or at least some of them.
>
> /boot/grub/grub.cfg itself is 111 lines.  Its not the most complex
> script going but for me it would take some serious study for an hour
> or more to figure out what is happening in it.  But of course you are
> not supposed to edit grub.cfg directly.
>
> It is orders of magnitude more complicated in my opinion... I'm not
> sure what the advantages are supposed to be.
>
> Also, its not clear how one would install a new kernel.  Where to put
> the information and so forth.
>
> -------        ---------       ---=---       ---------      --------
>
> find /boot -type f
>
> /boot/System.map-3.0.0-1-686-pae
> /boot/config-3.0.0-1-686-pae
> /boot/grub/device.map
> /boot/grub/915resolution.mod
> /boot/grub/acpi.mod
> /boot/grub/affs.mod
> /boot/grub/afs.mod
> /boot/grub/afs_be.mod
> /boot/grub/aout.mod
> /boot/grub/at_keyboard.mod
> /boot/grub/ata.mod
> /boot/grub/ata_pthru.mod
> /boot/grub/befs.mod
> /boot/grub/befs_be.mod
> /boot/grub/biosdisk.mod
> /boot/grub/bitmap.mod
> /boot/grub/bitmap_scale.mod
> /boot/grub/blocklist.mod
> /boot/grub/boot.mod
> /boot/grub/bsd.mod
> /boot/grub/btrfs.mod
> /boot/grub/bufio.mod
> /boot/grub/cat.mod
> /boot/grub/chain.mod
> /boot/grub/cmostest.mod
> /boot/grub/cmp.mod
> /boot/grub/configfile.mod
> /boot/grub/cpio.mod
> /boot/grub/cpuid.mod
> /boot/grub/crypto.mod
> /boot/grub/cs5536.mod
> /boot/grub/date.mod
> /boot/grub/datehook.mod
> /boot/grub/datetime.mod
> /boot/grub/dm_nv.mod
> /boot/grub/drivemap.mod
> /boot/grub/echo.mod
> /boot/grub/efiemu.mod
> /boot/grub/elf.mod
> /boot/grub/example_functional_test.mod
> /boot/grub/ext2.mod
> /boot/grub/extcmd.mod
> /boot/grub/fat.mod
> /boot/grub/font.mod
> /boot/grub/fshelp.mod
> /boot/grub/functional_test.mod
> /boot/grub/gcry_arcfour.mod
> /boot/grub/gcry_blowfish.mod
> /boot/grub/gcry_camellia.mod
> /boot/grub/gcry_cast5.mod
> /boot/grub/gcry_crc.mod
> /boot/grub/gcry_des.mod
> /boot/grub/gcry_md4.mod
> /boot/grub/gcry_md5.mod
> /boot/grub/gcry_rfc2268.mod
> /boot/grub/gcry_rijndael.mod
> /boot/grub/gcry_rmd160.mod
> /boot/grub/gcry_seed.mod
> /boot/grub/gcry_serpent.mod
> /boot/grub/gcry_sha1.mod
> /boot/grub/gcry_sha256.mod
> /boot/grub/gcry_sha512.mod
> /boot/grub/gcry_tiger.mod
> /boot/grub/gcry_twofish.mod
> /boot/grub/gcry_whirlpool.mod
> /boot/grub/gettext.mod
> /boot/grub/gfxmenu.mod
> /boot/grub/gfxterm.mod
> /boot/grub/gptsync.mod
> /boot/grub/gzio.mod
> /boot/grub/halt.mod
> /boot/grub/hashsum.mod
> /boot/grub/hdparm.mod
> /boot/grub/hello.mod
> /boot/grub/help.mod
> /boot/grub/hexdump.mod
> /boot/grub/hfs.mod
> /boot/grub/hfsplus.mod
> /boot/grub/iorw.mod
> /boot/grub/iso9660.mod
> /boot/grub/jfs.mod
> /boot/grub/jpeg.mod
> /boot/grub/keylayouts.mod
> /boot/grub/keystatus.mod
> /boot/grub/legacycfg.mod
> /boot/grub/linux.mod
> /boot/grub/linux16.mod
> /boot/grub/loadenv.mod
> /boot/grub/loopback.mod
> /boot/grub/ls.mod
> /boot/grub/lsacpi.mod
> /boot/grub/lsapm.mod
> /boot/grub/lsmmap.mod
> /boot/grub/lspci.mod
> /boot/grub/lvm.mod
> /boot/grub/mdraid09.mod
> /boot/grub/mdraid1x.mod
> /boot/grub/memdisk.mod
> /boot/grub/memrw.mod
> /boot/grub/minicmd.mod
> /boot/grub/mmap.mod
> /boot/grub/minix.mod
> /boot/grub/minix2.mod
> /boot/grub/msdospart.mod
> /boot/grub/multiboot.mod
> /boot/grub/multiboot2.mod
> /boot/grub/nilfs2.mod
> /boot/grub/normal.mod
> /boot/grub/ntfs.mod
> /boot/grub/ntfscomp.mod
> /boot/grub/ntldr.mod
> /boot/grub/ohci.mod
> /boot/grub/part_acorn.mod
> /boot/grub/part_amiga.mod
> /boot/grub/part_apple.mod
> /boot/grub/part_bsd.mod
> /boot/grub/part_gpt.mod
> /boot/grub/part_msdos.mod
> /boot/grub/part_sun.mod
> /boot/grub/part_sunpc.mod
> /boot/grub/parttool.mod
> /boot/grub/password.mod
> /boot/grub/password_pbkdf2.mod
> /boot/grub/pbkdf2.mod
> /boot/grub/pci.mod
> /boot/grub/play.mod
> /boot/grub/png.mod
> /boot/grub/probe.mod
> /boot/grub/pxe.mod
> /boot/grub/pxecmd.mod
> /boot/grub/raid.mod
> /boot/grub/raid5rec.mod
> /boot/grub/raid6rec.mod
> /boot/grub/read.mod
> /boot/grub/reboot.mod
> /boot/grub/regexp.mod
> /boot/grub/reiserfs.mod
> /boot/grub/relocator.mod
> /boot/grub/scsi.mod
> /boot/grub/search.mod
> /boot/grub/search_fs_file.mod
> /boot/grub/search_fs_uuid.mod
> /boot/grub/search_label.mod
> /boot/grub/sendkey.mod
> /boot/grub/serial.mod
> /boot/grub/setjmp.mod
> /boot/grub/setpci.mod
> /boot/grub/sfs.mod
> /boot/grub/sleep.mod
> /boot/grub/squash4.mod
> /boot/grub/tar.mod
> /boot/grub/terminal.mod
> /boot/grub/terminfo.mod
> /boot/grub/test.mod
> /boot/grub/test_blockarg.mod
> /boot/grub/testload.mod
> /boot/grub/tga.mod
> /boot/grub/trig.mod
> /boot/grub/true.mod
> /boot/grub/udf.mod
> /boot/grub/ufs1.mod
> /boot/grub/ufs2.mod
> /boot/grub/uhci.mod
> /boot/grub/usb.mod
> /boot/grub/usb_keyboard.mod
> /boot/grub/usbms.mod
> /boot/grub/usbserial_common.mod
> /boot/grub/usbserial_ftdi.mod
> /boot/grub/usbserial_pl2303.mod
> /boot/grub/usbtest.mod
> /boot/grub/vbe.mod
> /boot/grub/vga.mod
> /boot/grub/vga_text.mod
> /boot/grub/video.mod
> /boot/grub/video_bochs.mod
> /boot/grub/video_cirrus.mod
> /boot/grub/video_fb.mod
> /boot/grub/videoinfo.mod
> /boot/grub/videotest.mod
> /boot/grub/xfs.mod
> /boot/grub/xnu.mod
> /boot/grub/xnu_uuid.mod
> /boot/grub/xzio.mod
> /boot/grub/zfs.mod
> /boot/grub/zfsinfo.mod
> /boot/grub/command.lst
> /boot/grub/crypto.lst
> /boot/grub/fs.lst
> /boot/grub/moddep.lst
> /boot/grub/partmap.lst
> /boot/grub/parttool.lst
> /boot/grub/terminal.lst
> /boot/grub/video.lst
> /boot/grub/boot.img
> /boot/grub/cdboot.img
> /boot/grub/diskboot.img
> /boot/grub/g2hdr.img
> /boot/grub/grldr.img
> /boot/grub/kernel.img
> /boot/grub/lnxboot.img
> /boot/grub/pxeboot.img
> /boot/grub/efiemu32.o
> /boot/grub/efiemu64.o
> /boot/grub/locale/ast.mo
> /boot/grub/locale/ca.mo
> /boot/grub/locale/da.mo
> /boot/grub/locale/de.mo
> /boot/grub/locale/fi.mo
> /boot/grub/locale/fr.mo
> /boot/grub/locale/hu.mo
> /boot/grub/locale/id.mo
> /boot/grub/locale/it.mo
> /boot/grub/locale/ko.mo
> /boot/grub/locale/nl.mo
> /boot/grub/locale/pl.mo
> /boot/grub/locale/ru.mo
> /boot/grub/locale/sv.mo
> /boot/grub/locale/uk.mo
> /boot/grub/locale/vi.mo
> /boot/grub/locale/zh_CN.mo
> /boot/grub/core.img
> /boot/grub/grubenv
> /boot/grub/grub.cfg
> /boot/vmlinuz-3.0.0-1-686-pae
> /boot/initrd.img-3.0.0-1-686-pae
>
>

Re: [gentoo-user] Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 528 bytes --]

On Thu, 6 Oct 2011 13:01:11 +0200, Alan McKinnon wrote:

> Well the only constant is change, right?

It is now that c is in doubt :)

> The trick is to spot the difference between change for the sake of
> change and change that does make sense. This usually means getting
> inside someone's head, which makes life fun. Doubly so if the change
> proposer works in sales....

What about marketing, or do you not even attempt it then?


-- 
Neil Bothwick

Zmodem has bigger bits, softer blocks, and tighter ASCII

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 892 bytes --]

On Thu, 06 Oct 2011 07:51:04 -0500, Harry Putnam wrote:

> /boot/grub/grub.cfg itself is 111 lines.  Its not the most complex
> script going but for me it would take some serious study for an hour
> or more to figure out what is happening in it.  But of course you are
> not supposed to edit grub.cfg directly.

But you can if you wish.
> 
> It is orders of magnitude more complicated in my opinion... I'm not
> sure what the advantages are supposed to be.

Better automation.

> Also, its not clear how one would install a new kernel.  Where to put
> the information and so forth.

See above. You don't put anything anywhere when installing a new kernel,
just run grub-update and it will be found and added to the menu. At the
same time, any old kernels you have deleted will be removed.


-- 
Neil Bothwick

System halted - hit any Microsoft employee to continue.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[Alan McKinnon]

On Thu, 6 Oct 2011 14:47:26 +0100
Neil Bothwick <neil@digimed.co.uk> wrote:

> On Thu, 6 Oct 2011 13:01:11 +0200, Alan McKinnon wrote:
> 
> > Well the only constant is change, right?
> 
> It is now that c is in doubt :)

ooooooooooo! wicked pun!

Let's hold a poll. How many list readers get it?

> > The trick is to spot the difference between change for the sake of
> > change and change that does make sense. This usually means getting
> > inside someone's head, which makes life fun. Doubly so if the change
> > proposer works in sales....
> 
> What about marketing, or do you not even attempt it then?

Where I work, marketing are forbidden from not only coming up with
ideas, but from communicating with techies at all. Our marketing dept
is old-fashioned - promote stuff we already have, get the public to
think well of the company, dream up a new logo every 5 years.


-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Is grub2 stable and who uses it?

[Michael Mol]

On Thu, Oct 6, 2011 at 11:17 AM, Alan McKinnon <alan.mckinnon@gmail.com> wrote:
> On Thu, 6 Oct 2011 14:47:26 +0100
> Neil Bothwick <neil@digimed.co.uk> wrote:
>> On Thu, 6 Oct 2011 13:01:11 +0200, Alan McKinnon wrote:
>>
>> > Well the only constant is change, right?
>>
>> It is now that c is in doubt :)
>
> ooooooooooo! wicked pun!
>
> Let's hold a poll. How many list readers get it?

It's not c which is in doubt, but its function as an upper limit. ;)

-- 
:wq

Re: [gentoo-user] Is grub2 stable and who uses it?

[Alan McKinnon]

On Thu, 6 Oct 2011 11:25:04 -0400
Michael Mol <mikemol@gmail.com> wrote:

> On Thu, Oct 6, 2011 at 11:17 AM, Alan McKinnon
> <alan.mckinnon@gmail.com> wrote:
> > On Thu, 6 Oct 2011 14:47:26 +0100
> > Neil Bothwick <neil@digimed.co.uk> wrote:
> >> On Thu, 6 Oct 2011 13:01:11 +0200, Alan McKinnon wrote:
> >>
> >> > Well the only constant is change, right?
> >>
> >> It is now that c is in doubt :)
> >
> > ooooooooooo! wicked pun!
> >
> > Let's hold a poll. How many list readers get it?
> 
> It's not c which is in doubt, but its function as an upper limit. ;)
> 

Boys and girls,

Allow me to introduce to you the third member of the "Alan and Neil
Wise-cracking Wise-ass Club":

Michael Mol!

<round of applause>

-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] Is grub2 stable and who uses it?

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 335 bytes --]

On Thu, 6 Oct 2011 11:25:04 -0400, Michael Mol wrote:

> > Let's hold a poll. How many list readers get it?  
> 
> It's not c which is in doubt, but its function as an upper limit. ;)

You're right, c isn't in doubt :P


-- 
Neil Bothwick

Minds are like parachutes; they only function when fully open. * Sir
James Dewar

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[Sebastian Beßler]

[-- Attachment #1: Type: text/plain, Size: 503 bytes --]

Am 06.10.2011 17:25, schrieb Michael Mol:

>> Let's hold a poll. How many list readers get it?
> 
> It's not c which is in doubt, but its function as an upper limit. ;)

And even if the results a true and not an error then there is a list of
possible explanations. Most of them leave c as upper limit in place.
Quantum Mechanics has many aces in his sleeve like micro wormholes or
traveling through higher dimensions. All these dimensions from string
theorie must be good for something ;-)


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 900 bytes --]

[gentoo-user] Re: Is grub2 stable and who uses it?

[Harry Putnam]

Neil Bothwick <neil@digimed.co.uk> writes:

>> Also, its not clear how one would install a new kernel.  Where to put
>> the information and so forth.
>
> See above. You don't put anything anywhere when installing a new kernel,
> just run grub-update and it will be found and added to the menu. At the
> same time, any old kernels you have deleted will be removed.

Well now, that is an improvement.  But surely the kernel needs to be
put on /boot?

Re: [gentoo-user] Re: Is grub2 stable and who uses it?

[Alan McKinnon]

On Thu, 06 Oct 2011 12:48:08 -0500
Harry Putnam <reader@newsguy.com> wrote:

> Neil Bothwick <neil@digimed.co.uk> writes:
> 
> >> Also, its not clear how one would install a new kernel.  Where to
> >> put the information and so forth.
> >
> > See above. You don't put anything anywhere when installing a new
> > kernel, just run grub-update and it will be found and added to the
> > menu. At the same time, any old kernels you have deleted will be
> > removed.
> 
> Well now, that is an improvement.  But surely the kernel needs to be
> put on /boot?

That's not a function of grub-install, it's a function of make install



-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
> 
> most of the "oh it's so weird"-whining often comes from just not being
> used to it. flip your door lock upside down - you'll hate it with
> passion for a week and then you won't even notice. flip it again and
> the process will repeat.
> 

But if someone else snuck into your house and flipped your locks every week?

This one change won't be catastrophic, but I will probably spend a good
eight hours researching, testing, implementing, and documenting it. In
the end, *if everything goes according to plan*, stuff will work exactly
how it does now.

If Grub were the only package to do this -- fine, whatever. But next
week it will be something else. I don't know what my point is, but it
feels good to bitch about it.

Re: [gentoo-user] OT: change and improvement

[Dale]

Michael Orlitzky wrote:
> On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
>> most of the "oh it's so weird"-whining often comes from just not being
>> used to it. flip your door lock upside down - you'll hate it with
>> passion for a week and then you won't even notice. flip it again and
>> the process will repeat.
>>
> But if someone else snuck into your house and flipped your locks every week?
>
> This one change won't be catastrophic, but I will probably spend a good
> eight hours researching, testing, implementing, and documenting it. In
> the end, *if everything goes according to plan*, stuff will work exactly
> how it does now.
>
> If Grub were the only package to do this -- fine, whatever. But next
> week it will be something else. I don't know what my point is, but it
> feels good to bitch about it.
>
>

This is how I feel about the initramfs thingy and /usr and /var.  What 
is next?  I am pretty sure it will be something tho.

Dale

:-)  :-)

Re: [gentoo-user] OT: change and improvement

[Mick]

[-- Attachment #1: Type: Text/Plain, Size: 1298 bytes --]

On Thursday 06 Oct 2011 20:42:43 Dale wrote:
> Michael Orlitzky wrote:
> > On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
> >> most of the "oh it's so weird"-whining often comes from just not being
> >> used to it. flip your door lock upside down - you'll hate it with
> >> passion for a week and then you won't even notice. flip it again and
> >> the process will repeat.
> > 
> > But if someone else snuck into your house and flipped your locks every
> > week?
> > 
> > This one change won't be catastrophic, but I will probably spend a good
> > eight hours researching, testing, implementing, and documenting it. In
> > the end, *if everything goes according to plan*, stuff will work exactly
> > how it does now.
> > 
> > If Grub were the only package to do this -- fine, whatever. But next
> > week it will be something else. I don't know what my point is, but it
> > feels good to bitch about it.
> 
> This is how I feel about the initramfs thingy and /usr and /var.  What
> is next?  I am pretty sure it will be something tho.

I share your pain.  :-(

I'm not sure if this a sign of me getting (even) older, or Linux maturing and 
in doing so it caters less and less for Gentoo geeky users and more and more 
for mainstream ignoramuses.  :p

-- 
Regards,
Mick

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] OT: change and improvement

[Jonas de Buhr]

Am Thu, 06 Oct 2011 15:27:14 -0400
schrieb Michael Orlitzky <michael@orlitzky.com>:

> On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
> > 
> > most of the "oh it's so weird"-whining often comes from just not
> > being used to it. flip your door lock upside down - you'll hate it
> > with passion for a week and then you won't even notice. flip it
> > again and the process will repeat.
> > 
> 
> But if someone else snuck into your house and flipped your locks
> every week?
> 
> This one change won't be catastrophic, but I will probably spend a
> good eight hours researching, testing, implementing, and documenting
> it. In the end, *if everything goes according to plan*, stuff will
> work exactly how it does now.

nothing forces you to switch to grub2.

> If Grub were the only package to do this -- fine, whatever. But next
> week it will be something else. I don't know what my point is, but it
> feels good to bitch about it.

don't get me started, the suppressed memories about HAL-config or
broken suse-10 usermount may come back ;)

but grub2 config doesn't look all that bad to me...
http://www.linuxfromscratch.org/lfs/view/development/chapter08/grub.html
seems like you can have a single grub.cfg and ignore the rest

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/06/2011 05:00 PM, Jonas de Buhr wrote:
> Am Thu, 06 Oct 2011 15:27:14 -0400
> schrieb Michael Orlitzky <michael@orlitzky.com>:
> 
>> On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
>>>
>>> most of the "oh it's so weird"-whining often comes from just not
>>> being used to it. flip your door lock upside down - you'll hate it
>>> with passion for a week and then you won't even notice. flip it
>>> again and the process will repeat.
>>>
>>
>> But if someone else snuck into your house and flipped your locks
>> every week?
>>
>> This one change won't be catastrophic, but I will probably spend a
>> good eight hours researching, testing, implementing, and documenting
>> it. In the end, *if everything goes according to plan*, stuff will
>> work exactly how it does now.
> 
> nothing forces you to switch to grub2.
> 

True in theory, but not in practice. Legacy grub will go away
eventually. If we have some grub-legacy and some grub2 installs, we have
to support (document, test, take out to dinner occasionally) both, which
is probably going to be more effort than just moving to grub2 after I
figure out how it works.

Either way is going to require a non-zero amount of work, while zero is
the amount of work I would prefer to do.

Re: [gentoo-user] OT: change and improvement

[Paul Hartman]

On Thu, Oct 6, 2011 at 4:23 PM, Michael Orlitzky <michael@orlitzky.com> wrote:
> Either way is going to require a non-zero amount of work, while zero is
> the amount of work I would prefer to do.

Words to live by. :)

Re: [gentoo-user] OT: change and improvement

[Canek Peláez Valdés]

On Thu, Oct 6, 2011 at 2:23 PM, Michael Orlitzky <michael@orlitzky.com> wrote:
> On 10/06/2011 05:00 PM, Jonas de Buhr wrote:
>> Am Thu, 06 Oct 2011 15:27:14 -0400
>> schrieb Michael Orlitzky <michael@orlitzky.com>:
>>
>>> On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
>>>>
>>>> most of the "oh it's so weird"-whining often comes from just not
>>>> being used to it. flip your door lock upside down - you'll hate it
>>>> with passion for a week and then you won't even notice. flip it
>>>> again and the process will repeat.
>>>>
>>>
>>> But if someone else snuck into your house and flipped your locks
>>> every week?
>>>
>>> This one change won't be catastrophic, but I will probably spend a
>>> good eight hours researching, testing, implementing, and documenting
>>> it. In the end, *if everything goes according to plan*, stuff will
>>> work exactly how it does now.
>>
>> nothing forces you to switch to grub2.
>>
>
> True in theory, but not in practice. Legacy grub will go away
> eventually.

Technically, it's already gone. It's on life-support: the developers
of grub-legacy are the same of GRUB2, and they are only accepting bug
fixes, not new features (I believe, someone correct me if I'm
mistaken).

The good news is that your current hardware (and also the new, for the
next few months) will probably work OK with grub-legacy. The bad news
is that machines with EFI and UEFI will need to use GRUB2 (again, that
is what I understand, correct me if I'm wrong).

> If we have some grub-legacy and some grub2 installs, we have
> to support (document, test, take out to dinner occasionally) both, which
> is probably going to be more effort than just moving to grub2 after I
> figure out how it works.

Exactly. From my point of view, better to start moving to GRUB2 now
(except for critical systems), to got it mastered when it hits stable.

> Either way is going to require a non-zero amount of work, while zero is
> the amount of work I would prefer to do.

We have a say in México: "el que quiera azul celeste, que le cueste".
It's basically the same as "there's no such thing as a free lunch":
everything costs, maybe in money, maybe in time, maybe in work, and
possibly on all of the above.

But hey, at least you don't have to write your own boot loader.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] OT: change and improvement

[Jonas de Buhr]

Am Thu, 06 Oct 2011 17:23:32 -0400
schrieb Michael Orlitzky <michael@orlitzky.com>:

> > nothing forces you to switch to grub2.
> > 
> 
> True in theory, but not in practice. 

the purpose of theory is to predict what happens in practice. if it 
does't, the theory is wrong.

>Legacy grub will go away
> eventually. 

>If we have some grub-legacy and some grub2 installs, 

why would you do that?

> Either way is going to require a non-zero amount of work, while zero
> is the amount of work I would prefer to do.

have a look at the link i posted. if you really want to keep it simple,
grub2 can be configured with a single grub.cfg file.

gentoo regularly requires some manual work during upgrades. i really
don't get which this is such a big deal.

Re: [gentoo-user] Is grub2 stable and who uses it?

[Peter Humphrey]

[-- Attachment #1: Type: text/plain, Size: 777 bytes --]

On Thursday 06 October 2011 17:40:33 Sebastian Beßler wrote:
> Am 06.10.2011 17:25, schrieb Michael Mol:
> >> Let's hold a poll. How many list readers get it?
> > 
> > It's not c which is in doubt, but its function as an upper limit. ;)
> 
> And even if the results a true and not an error then there is a list of
> possible explanations. Most of them leave c as upper limit in place.
> Quantum Mechanics has many aces in his sleeve like micro wormholes or
> traveling through higher dimensions. All these dimensions from string
> theorie must be good for something ;-)

Just think of the consequences if c is not the ultimate speed limit. I am, 
and they're so numerous that I can't even contemplate them all.

-- 
Rgds
Peter		Linux Counter 5290, 1994-04-23

[-- Attachment #2: Type: text/html, Size: 3776 bytes --]

Re: [gentoo-user] Is grub2 stable and who uses it?

[Michael Mol]

[-- Attachment #1: Type: text/plain, Size: 1160 bytes --]

On Oct 6, 2011 8:10 PM, "Peter Humphrey" <peter@humphrey.ukfsn.org> wrote:
>
> On Thursday 06 October 2011 17:40:33 Sebastian Beßler wrote:
>
> > Am 06.10.2011 17:25, schrieb Michael Mol:
>
> > >> Let's hold a poll. How many list readers get it?
>
> > >
>
> > > It's not c which is in doubt, but its function as an upper limit. ;)
>
> >
>
> > And even if the results a true and not an error then there is a list of
>
> > possible explanations. Most of them leave c as upper limit in place.
>
> > Quantum Mechanics has many aces in his sleeve like micro wormholes or
>
> > traveling through higher dimensions. All these dimensions from string
>
> > theorie must be good for something ;-)
>
>
> Just think of the consequences if c is not the ultimate speed limit. I am,
and they're so numerous that I can't even contemplate them all.

Avoiding fantasizing about FTL physical travel, FTL information transfer
using nutrinos strikes me as the most fascinating prospect. Imagine being
able to set up a direct point-to-point link through the globe. Skip surface
latency, get a direct Gigabit link between, e.g. Australia and the US.

[-- Attachment #2: Type: text/html, Size: 1471 bytes --]

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/06/11 19:42, Jonas de Buhr wrote:
> 
>> If we have some grub-legacy and some grub2 installs, 
> 
> why would you do that?

Eventually, grub2 will be all that's available from portage. At that
point, I can either,

1) Install grub2 on some machines.

2) Maintain grub-legacy (and install media) myself.


> have a look at the link i posted. if you really want to keep it simple,
> grub2 can be configured with a single grub.cfg file.

How much work would it be for you to,

  * Learn grub2
  * Travel to my office here in Baltimore
  * Test it on all combinations of hardware that we currently run
  * Document the standard config and any special cases
  * Upgrade a bunch of my servers at 4am?

If you still think it's "not much" then I'd be happy to have you do it
while I drink margaritas.

[gentoo-user] Re: OT: change and improvement

[Grant Edwards]

On 2011-10-06, Jonas de Buhr <jonas.de.buhr@gmx.net> wrote:
> Am Thu, 06 Oct 2011 15:27:14 -0400 schrieb Michael Orlitzky <michael@orlitzky.com>:
>
>> On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
>>
>>> most of the "oh it's so weird"-whining often comes from just not
>>> being used to it. flip your door lock upside down - you'll hate it
>>> with passion for a week and then you won't even notice. flip it
>>> again and the process will repeat.

Weird I can deal with.  It's the "order of magnitude more complicated"
that gets annoying.  Granted, in the case of Grub2, most of that is
due to the fact that I encounter in mainly on Ubuntu systems.  I don't
know how they manage it, but every time the Ubunutu folks have a
decision to make, they pick a direction exactly opposite that which I
would choose.  Still, I have to give them credit for still making a
usable single-CD system when everybody else requires 4 CDs or a DVD.

>> But if someone else snuck into your house and flipped your locks
>> every week?
>> 
>> This one change won't be catastrophic, but I will probably spend a
>> good eight hours researching, testing, implementing, and documenting
>> it. In the end, *if everything goes according to plan*, stuff will
>> work exactly how it does now.
>
> nothing forces you to switch to grub2.

Not this week, no.

It'll happen...

> don't get me started, the suppressed memories about HAL-config or
> broken suse-10 usermount may come back ;)

HAL... shudder.

-- 
Grant

[gentoo-user] Re: OT: change and improvement

[Grant Edwards]

On 2011-10-06, Jonas de Buhr <jonas.de.buhr@gmx.net> wrote:
> Am Thu, 06 Oct 2011 17:23:32 -0400
> schrieb Michael Orlitzky <michael@orlitzky.com>:
>
>> > nothing forces you to switch to grub2.
>> > 
>> 
>> True in theory, but not in practice. 
>
> the purpose of theory is to predict what happens in practice. if it 
> does't, the theory is wrong.
>
>>Legacy grub will go away
>> eventually. 
>
>>If we have some grub-legacy and some grub2 installs, 
>
> why would you do that?

Some distros don't give you a choice.

-- 
Grant

Re: [gentoo-user] OT: change and improvement

[Dale]

Mick wrote:
> On Thursday 06 Oct 2011 20:42:43 Dale wrote:
>> Michael Orlitzky wrote:
>>> On 10/06/2011 04:20 AM, Jonas de Buhr wrote:
>>>> most of the "oh it's so weird"-whining often comes from just not being
>>>> used to it. flip your door lock upside down - you'll hate it with
>>>> passion for a week and then you won't even notice. flip it again and
>>>> the process will repeat.
>>> But if someone else snuck into your house and flipped your locks every
>>> week?
>>>
>>> This one change won't be catastrophic, but I will probably spend a good
>>> eight hours researching, testing, implementing, and documenting it. In
>>> the end, *if everything goes according to plan*, stuff will work exactly
>>> how it does now.
>>>
>>> If Grub were the only package to do this -- fine, whatever. But next
>>> week it will be something else. I don't know what my point is, but it
>>> feels good to bitch about it.
>> This is how I feel about the initramfs thingy and /usr and /var.  What
>> is next?  I am pretty sure it will be something tho.
> I share your pain.  :-(
>
> I'm not sure if this a sign of me getting (even) older, or Linux maturing and
> in doing so it caters less and less for Gentoo geeky users and more and more
> for mainstream ignoramuses.  :p
>

I was thinking more like windoze really.  If windoze starts having mount 
points like Linux, things could start changing.  ^_^   Think about it, 
windoze currently has to have its stuff on the C drive and Linux can be 
spread out over many drives and you can mount things wherever you want.  
Linux is going the way of windoze then windoze would be going the way of 
Linux.  Weird huh?

Dale

:-)  :-)

Re: [gentoo-user] Re: OT: change and improvement

[Dale]

Grant Edwards wrote:
> On 2011-10-06, Jonas de Buhr<jonas.de.buhr@gmx.net>  wrote:
>
>> don't get me started, the suppressed memories about HAL-config or
>> broken suse-10 usermount may come back ;)
> HAL... shudder.
>

Let's not get started on hal.  Pardon me while I go pray to the 
porcelain gods.  :/

Dale

:-)  :-)

Re: [gentoo-user] Re: OT: change and improvement

[Michael Mol]

[-- Attachment #1: Type: text/plain, Size: 565 bytes --]

On Oct 6, 2011 11:34 PM, "Dale" <rdalek1967@gmail.com> wrote:
>
> Grant Edwards wrote:
>>
>> On 2011-10-06, Jonas de Buhr<jonas.de.buhr@gmx.net>  wrote:
>>
>>> don't get me started, the suppressed memories about HAL-config or
>>> broken suse-10 usermount may come back ;)
>>
>> HAL... shudder.
>>
>
> Let's not get started on hal.  Pardon me while I go pray to the porcelain
gods.  :/

I'm really sorry, but I missed that show. I was in Ubuntu-land learning not
to tweak system behaviors.  WTF was the deal with HAL? What was it supposed
to do? Why didn't it work?

[-- Attachment #2: Type: text/html, Size: 845 bytes --]

Re: [gentoo-user] Re: OT: change and improvement

[Michael Orlitzky]

On 10/06/2011 11:39 PM, Michael Mol wrote:
> 
>>> HAL... shudder.
>>>
>>
>> Let's not get started on hal.  Pardon me while I go pray to the
> porcelain gods.  :/
> 
> I'm really sorry, but I missed that show. I was in Ubuntu-land learning
> not to tweak system behaviors.  WTF was the deal with HAL? What was it
> supposed to do? Why didn't it work?
> 

From the Wikipedia page,

  HAL in [Linux]:

  HAL became operational in Urbana, Illinois at the HAL Plant at the
  University of Illinois, where the ILLIAC computers were assembled.
  The film states that this occurred in 1992, while the book gives 1997
  as HAL's birthyear. In [Linux], HAL begins to malfunction in subtle
  ways, and as a result, the decision is made to shut HAL down in order
  to prevent more serious malfunctions. The sequence of events and
  manner in which HAL is shut down differs between the novel and film
  versions of the story.

Re: [gentoo-user] Re: OT: change and improvement

[Michael Mol]

[-- Attachment #1: Type: text/plain, Size: 1126 bytes --]

On Oct 7, 2011 12:44 AM, "Michael Orlitzky" <michael@orlitzky.com> wrote:
>
> On 10/06/2011 11:39 PM, Michael Mol wrote:
> >
> >>> HAL... shudder.
> >>>
> >>
> >> Let's not get started on hal.  Pardon me while I go pray to the
> > porcelain gods.  :/
> >
> > I'm really sorry, but I missed that show. I was in Ubuntu-land learning
> > not to tweak system behaviors.  WTF was the deal with HAL? What was it
> > supposed to do? Why didn't it work?
> >
>
> From the Wikipedia page,
>
>  HAL in [Linux]:
>
>  HAL became operational in Urbana, Illinois at the HAL Plant at the
>  University of Illinois, where the ILLIAC computers were assembled.
>  The film states that this occurred in 1992, while the book gives 1997
>  as HAL's birthyear. In [Linux], HAL begins to malfunction in subtle
>  ways, and as a result, the decision is made to shut HAL down in order
>  to prevent more serious malfunctions. The sequence of events and
>  manner in which HAL is shut down differs between the novel and film
>  versions of the story.

Heh. I doubt Mr. Clark was imagining hardware abstractions while he was
exploring AI malfunction. :)

[-- Attachment #2: Type: text/html, Size: 1492 bytes --]

Re: [gentoo-user] Re: OT: change and improvement

[Canek Peláez Valdés]

On Thu, Oct 6, 2011 at 8:39 PM, Michael Mol <mikemol@gmail.com> wrote:
>
> On Oct 6, 2011 11:34 PM, "Dale" <rdalek1967@gmail.com> wrote:
>>
>> Grant Edwards wrote:
>>>
>>> On 2011-10-06, Jonas de Buhr<jonas.de.buhr@gmx.net>  wrote:
>>>
>>>> don't get me started, the suppressed memories about HAL-config or
>>>> broken suse-10 usermount may come back ;)
>>>
>>> HAL... shudder.
>>>
>>
>> Let's not get started on hal.  Pardon me while I go pray to the porcelain
>> gods.  :/
>
> I'm really sorry, but I missed that show. I was in Ubuntu-land learning not
> to tweak system behaviors.  WTF was the deal with HAL? What was it supposed
> to do? Why didn't it work?

HAL was the Hardware Abstraction Layer. The idea was that it would be
an index of the computer's hardware, so high level parts of the stack
would know what hardware and capabilities were available, without the
need of communicate with the kernel directly. This had the (perceived)
advantage of bein "portable", so for example a program using HAL could
run in any OS. Even Windows was theoretically included.

But it ended being *HUGE*, because it needed to handle a lot of
different hardware and Operating System quirks, and the design was
rigid and not very scalable. Also, it turns out that a lot of
Operating Systems do not have the necessary developers to fill the
specifications required by HAL: if I remember correctly, Solaris and
the *BSD family of systems implemented only a fraction of what was
needed. So, as is usually the case with "portable" code, the software
using HAL had to use the minimum common denominator.

In the end, udev was making huge advances, and HAL could not keep up
simply because the other Operating Systems didn't have similar
capabilities, so the consumers of HAL (desktop systems, mostly)
started to use udev directly. That was when the shit hit the fan: if
the purpose of HAL was to mantain "portability", but the biggest and
most active developer community (Linux) refused to use it since it
didn't allowed them to use the full capabilities of the operating
system, then it had (literally) no reason to live. So the HAL
mantainers saw the error of their ways, and they deprecated it, saying
to the user space developers that, in Linux, they should use udev, and
in other Operating Systems whatever was equivalent, if any.

It was really fast, if I remember correctly: one day half the programs
in my computers used HAL, and the next every single one of them
stopped using it. In Gentoo in particular was pretty rough, since the
X.org version that used HAL had just become stable (which was kinda
difficult to transition to), and next thing you know, you again had to
transition, this time to a HAL-free X.org. A lot of users got really
angry in Gentoo because of that.

I remember it was a couple of days getting everything back to normal,
but it required taking down at least one production system, and
basically rebuilding everything on the desktop stack in a workstation
computer (since the desktop was the most heavy user of HAL). But I had
it worse with the glitz cairo USE flag, or the several expat problems
that always arise when you update that library. HAL was bad, but
(IMHO) not *that* bad.

In the end, everything is fine I believe. The desktops (all of them)
have made a lot of advances in Linux since they no longer had to use
the minimum common denominator, and it showed a lot of devs that the
mythical grail of "portability" is not that great at the end of the
day. That's one of the reasons the developer of systemd refuses to
make it portable to other Unixes: it allows him to use all the great
and specific features of Linux, and not to worry about other systems.
If other systems want something similar to systemd, they have the
liberty of programming themselves, of course.

This is how I remember it. If I made I mistake, someone please correct me.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] OT: change and improvement

[Jonas de Buhr]

Am 07.10.2011 02:55, schrieb Michael Orlitzky:
> On 10/06/11 19:42, Jonas de Buhr wrote:
>>> If we have some grub-legacy and some grub2 installs,
>> why would you do that?
> Eventually, grub2 will be all that's available from portage. At that
> point, I can either,
>
> 1) Install grub2 on some machines.
>
> 2) Maintain grub-legacy (and install media) myself.

i really don't think thats the way its going to be. i think there will 
be grub and grub2 in portage potentially forever. like with python 2 and 3.

even if not, 2) takes you one cp command and a little bit of disk space 
for the grub tarball.

>
>
>> have a look at the link i posted. if you really want to keep it simple,
>> grub2 can be configured with a single grub.cfg file.
> How much work would it be for you to,
>
>    * Learn grub2
if you're ok with the single .cfg: done.
>    * Travel to my office here in Baltimore
first class flight please.
>    * Test it on all combinations of hardware that we currently run
>    * Document the standard config and any special cases
ok.
>    * Upgrade a bunch of my servers at 4am?
why not choose a convenient time to upgrade?
>
> If you still think it's "not much" then I'd be happy to have you do it
> while I drink margaritas.
no, i still don't think its as much of a big deal as you make of it. 
about as much work as a kernel upgrade. let's wait for grub2 to go 
stable before you send me that ticket ;)

Re: [gentoo-user] Is grub2 stable and who uses it?

[Alan McKinnon]

On Fri, 7 Oct 2011 01:08:24 +0100
Peter Humphrey <peter@humphrey.ukfsn.org> wrote:

> On Thursday 06 October 2011 17:40:33 Sebastian Beßler wrote:
> > Am 06.10.2011 17:25, schrieb Michael Mol:
> > >> Let's hold a poll. How many list readers get it?
> > > 
> > > It's not c which is in doubt, but its function as an upper
> > > limit. ;)
> > 
> > And even if the results a true and not an error then there is a
> > list of possible explanations. Most of them leave c as upper limit
> > in place. Quantum Mechanics has many aces in his sleeve like micro
> > wormholes or traveling through higher dimensions. All these
> > dimensions from string theorie must be good for something ;-)
> 
> Just think of the consequences if c is not the ultimate speed limit.
> I am, and they're so numerous that I can't even contemplate them all.

Permit me to direct you to the pearl of wisdom about that in the movie
K-Pax



-- 
Alan McKinnnon
alan.mckinnon@gmail.com

Re: [gentoo-user] OT: change and improvement

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 639 bytes --]

On Thu, 06 Oct 2011 20:55:05 -0400, Michael Orlitzky wrote:

> How much work would it be for you to,
> 
>   * Learn grub2

About the same as it would be for you, very little.

The config has moved from /boot/grub to /etc/grub.d and the syntax has
changed slightly.

Note that, BY DEFAULT, grub2 comes with a file for you to add your manual
configurations. If it weren't for the confusing idea of putting some of
the settings in a different location, the learning curve would be trivial.


-- 
Neil Bothwick

"Everything takes longer than expected, even when you take
  into account Hoffstead's Law." - Hoffstead's Law

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/07/2011 04:47 AM, Neil Bothwick wrote:
> On Thu, 06 Oct 2011 20:55:05 -0400, Michael Orlitzky wrote:
> 
>> How much work would it be for you to,
>>
>>   * Learn grub2
> 
> About the same as it would be for you, very little.

Granted, this is the easy one on my list.

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/07/2011 03:36 AM, Jonas de Buhr wrote:
> Am 07.10.2011 02:55, schrieb Michael Orlitzky:
>> On 10/06/11 19:42, Jonas de Buhr wrote:
>>>> If we have some grub-legacy and some grub2 installs,
>>> why would you do that?
>> Eventually, grub2 will be all that's available from portage. At that
>> point, I can either,
>>
>> 1) Install grub2 on some machines.
>>
>> 2) Maintain grub-legacy (and install media) myself.
> 
> i really don't think thats the way its going to be. i think there will 
> be grub and grub2 in portage potentially forever. like with python 2 and 3.
> 
> even if not, 2) takes you one cp command and a little bit of disk space 
> for the grub tarball.

Python2 will stick around because most packages (portage!) don't work
with python3. Grub doesn't have the same problem.

(2) requires me to at least,

  * Figure out how to build a Gentoo install CD
  * Fork grub-legacy on our servers somewhere
  * Test it against all future kernel releases
  * Document why we're doing this, and how to do the first three steps.


>>    * Upgrade a bunch of my servers at 4am?
> why not choose a convenient time to upgrade?

4am *is* the convenient time to upgrade.


>> If you still think it's "not much" then I'd be happy to have you do it
>> while I drink margaritas.
> no, i still don't think its as much of a big deal as you make of it. 
> about as much work as a kernel upgrade. let's wait for grub2 to go 
> stable before you send me that ticket ;)

This fails as a debate strategy since you wouldn't have to pay my
mortgage and feed me if you screwed up =)

Kernel upgrades usually take me a full day. I get to skip most of the
documentation step, but have to deal with heterogeneous configs. I'm not
saying that this is some huge problem on a cosmic scale, but it is going
to waste a day and risk downtime for no user-visible benefit.

Re: [gentoo-user] OT: change and improvement

[Alan Mackenzie]

Hi, Michael.

On Thu, Oct 06, 2011 at 08:55:05PM -0400, Michael Orlitzky wrote:
> On 10/06/11 19:42, Jonas de Buhr wrote:

> >> If we have some grub-legacy and some grub2 installs, 

> > why would you do that?

> Eventually, grub2 will be all that's available from portage. At that
> point, I can either,

> 1) Install grub2 on some machines.

> 2) Maintain grub-legacy (and install media) myself.

> > have a look at the link i posted. if you really want to keep it simple,
> > grub2 can be configured with a single grub.cfg file.

> How much work would it be for you to,

>   * Learn grub2
>   * Travel to my office here in Baltimore
>   * Test it on all combinations of hardware that we currently run
>   * Document the standard config and any special cases
>   * Upgrade a bunch of my servers at 4am?

> If you still think it's "not much" then I'd be happy to have you do it
> while I drink margaritas.

Why don't you upgrade to lilo instead?  It's got a single configuration
file which is short and relatively simple, and it just works.  My
lilo.conf has 50 non-comment/space lines, and that includes 8 kernel
versions.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: [gentoo-user] OT: change and improvement

[Brennan Shacklett]

[-- Attachment #1: Type: text/plain, Size: 206 bytes --]

> Python2 will stick around because most packages (portage!) don't work
> with python3. Grub doesn't have the same problem.
>
>
Just to be pedantic, portage uses python3 if the python3 use flag is
enabled.

[-- Attachment #2: Type: text/html, Size: 394 bytes --]

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/07/2011 10:29 AM, Alan Mackenzie wrote:
> 
> Why don't you upgrade to lilo instead?  It's got a single configuration
> file which is short and relatively simple, and it just works.  My
> lilo.conf has 50 non-comment/space lines, and that includes 8 kernel
> versions.
> 

Why don't I avoid $difficult_thing by doing
$equally_difficult_other_thing? =)

If I thought lilo was going to be around longer than grub2, it might
make sense. I'm gambling on that one.

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/07/2011 10:46 AM, Brennan Shacklett wrote:
> 
>     Python2 will stick around because most packages (portage!) don't work
>     with python3. Grub doesn't have the same problem.
> 
> 
> Just to be pedantic, portage uses python3 if the python3 use flag is
> enabled.
> 

That's not being pedantic; it looks like I'm outdated on this one.

Re: [gentoo-user] OT: change and improvement

[Dale]

Michael Orlitzky wrote:
> On 10/07/2011 10:29 AM, Alan Mackenzie wrote:
>> Why don't you upgrade to lilo instead?  It's got a single configuration
>> file which is short and relatively simple, and it just works.  My
>> lilo.conf has 50 non-comment/space lines, and that includes 8 kernel
>> versions.
>>
> Why don't I avoid $difficult_thing by doing
> $equally_difficult_other_thing? =)
>
> If I thought lilo was going to be around longer than grub2, it might
> make sense. I'm gambling on that one.
>
>

Not to mention you have to run lilo when you update kernels and other 
things too.  I used lilo when I first started using Linux.  I can't 
imagine me ever going back to that thing.

Dale

:-)  :-)

Re: [gentoo-user] OT: change and improvement

[Paul Hartman]

On Fri, Oct 7, 2011 at 10:18 AM, Dale <rdalek1967@gmail.com> wrote:
> Michael Orlitzky wrote:
>>
>> On 10/07/2011 10:29 AM, Alan Mackenzie wrote:
>>>
>>> Why don't you upgrade to lilo instead?  It's got a single configuration
>>> file which is short and relatively simple, and it just works.  My
>>> lilo.conf has 50 non-comment/space lines, and that includes 8 kernel
>>> versions.
>>>
>> Why don't I avoid $difficult_thing by doing
>> $equally_difficult_other_thing? =)
>>
>> If I thought lilo was going to be around longer than grub2, it might
>> make sense. I'm gambling on that one.
>>
>>
>
> Not to mention you have to run lilo when you update kernels and other things
> too.  I used lilo when I first started using Linux.  I can't imagine me ever
> going back to that thing.

Back in the mid 90's I used to use a boot floppy to dual-boot, I think
it was LILO... Floppy in, it booted Debian, floppy out, it booted
OS/2. One day, someone took the floppy and reformatted it to copy some
files onto it; that seriously hindered my ability to boot Linux at
that time. :)

Re: [gentoo-user] OT: change and improvement

[Jonas de Buhr]

Am Fri, 07 Oct 2011 09:59:54 -0400
schrieb Michael Orlitzky <michael@orlitzky.com>:

> On 10/07/2011 03:36 AM, Jonas de Buhr wrote:
> > Am 07.10.2011 02:55, schrieb Michael Orlitzky:
> >> On 10/06/11 19:42, Jonas de Buhr wrote:
> >>>> If we have some grub-legacy and some grub2 installs,
> >>> why would you do that?
> >> Eventually, grub2 will be all that's available from portage. At
> >> that point, I can either,
> >>
> >> 1) Install grub2 on some machines.
> >>
> >> 2) Maintain grub-legacy (and install media) myself.
> > 
> > i really don't think thats the way its going to be. i think there
> > will be grub and grub2 in portage potentially forever. like with
> > python 2 and 3.
> > 
> > even if not, 2) takes you one cp command and a little bit of disk
> > space for the grub tarball.
> 
> Python2 will stick around because most packages (portage!) don't work
> with python3. Grub doesn't have the same problem.
> 
> (2) requires me to at least,
> 
>   * Figure out how to build a Gentoo install CD
>   * Fork grub-legacy on our servers somewhere

build a package and put it on an ftp. monitor the bug-grub mailing list
for critical bugs. repeat.
or mirror it daily and autobuild the package.
no need to build an install cd or fork grub. but it is true that this is
less convenient than portage. its probably less work to switch to grub2
IF grub legacy really ever gets thrown out of portage. mixing both is
the worst idea in my opinion.

>   * Test it against all future kernel releases
>   * Document why we're doing this, and how to do the first three
> steps.
> 
> 
> >>    * Upgrade a bunch of my servers at 4am?
> > why not choose a convenient time to upgrade?
> 
> 4am *is* the convenient time to upgrade.
> 
> >> If you still think it's "not much" then I'd be happy to have you
> >> do it while I drink margaritas.
> > no, i still don't think its as much of a big deal as you make of
> > it. about as much work as a kernel upgrade. let's wait for grub2 to
> > go stable before you send me that ticket ;)
> 
> This fails as a debate strategy since you wouldn't have to pay my
> mortgage and feed me if you screwed up =)

hey, that was *your* idea in the first place ;)

> Kernel upgrades usually take me a full day. I get to skip most of the
> documentation step, but have to deal with heterogeneous configs. 

out of interest: why do you have different configs? even if you have
different hardware you could still build a "one fits all"-kernel. or
are they that specialized?

>I'm
> not saying that this is some huge problem on a cosmic scale, but it
> is going to waste a day and risk downtime for no user-visible benefit.

lets just agree on that. im kinda tired of this discussion. there's
nothing we can do about it anyway.

Re: [gentoo-user] OT: change and improvement

[Michael Orlitzky]

On 10/07/2011 12:16 PM, Jonas de Buhr wrote:
> 
> out of interest: why do you have different configs? even if you have
> different hardware you could still build a "one fits all"-kernel. or
> are they that specialized?
> 

We share kernel config whenever possible, but there are a few cases
where they have to diverge. Basically any option that can't be compiled
as a module is a candidate. Off the top of my head,

  * We've got x86/amd64
  * Intel/AMD
  * A couple with RAID hardware that can't have its module installed.
  * One server with a tulip NIC that can't use a particular driver.
  * A set where hyperthreading needs to be disabled
  * A virtual machine host that needs certain hardening features
    disabled
  * A separate config for VM guests
  * Headless vs. GUI requires more grsec/pax tweaking
  * Different HZ settings. Power management in general depends on what
    the box will be doing.
  * A firewall with no non-essential modules available

We keep the configs in git, so if two are similar I can usually just
pull the last changeset (after a make oldconfig) over. What sucks is
testing, and of course driving to work to reboot everything off-hours.


> lets just agree on that. im kinda tired of this discussion. there's
> nothing we can do about it anyway.
> 

Agreed.

[gentoo-user] Re: OT: change and improvement

[Grant Edwards]

On 2011-10-07, Canek Pel??ez Vald??s <caneko@gmail.com> wrote:

> In the end, udev was making huge advances, and HAL could not keep up
> simply because the other Operating Systems didn't have similar
> capabilities, so the consumers of HAL (desktop systems, mostly)
> started to use udev directly. That was when the shit hit the fan: if
> the purpose of HAL was to mantain "portability", but the biggest and
> most active developer community (Linux) refused to use it since it
> didn't allowed them to use the full capabilities of the operating
> system, then it had (literally) no reason to live. So the HAL
> mantainers saw the error of their ways, and they deprecated it,
> saying to the user space developers that, in Linux, they should use
> udev, and in other Operating Systems whatever was equivalent, if any.
>
> It was really fast, if I remember correctly: one day half the
> programs in my computers used HAL, and the next every single one of
> them stopped using it. In Gentoo in particular was pretty rough,
> since the X.org version that used HAL had just become stable (which
> was kinda difficult to transition to), and next thing you know, you
> again had to transition, this time to a HAL-free X.org. A lot of
> users got really angry in Gentoo because of that.

Some of us grumpy-old-guy types refused to play nice and didn't make
either transition.  We disabled HAL support in X.org [after X.org had
stopped working when HAL became the default].  Once we had HAL
disabled, we continued to to use the trusty old xorg.conf file until
the whole HAL thing blew over.

I'm going to ignore grub2 for as long as I can, but I don't think it's
going away the way HAL did... ;)

-- 
Grant Edwards               grant.b.edwards        Yow! I'd like some JUNK
                                  at               FOOD ... and then I want to
                              gmail.com            be ALONE --

[gentoo-user] Re: OT: change and improvement

[Grant Edwards]

On 2011-10-07, Michael Orlitzky <michael@orlitzky.com> wrote:
> On 10/07/2011 03:36 AM, Jonas de Buhr wrote:
>> Am 07.10.2011 02:55, schrieb Michael Orlitzky:
>>> On 10/06/11 19:42, Jonas de Buhr wrote:
>>>>> If we have some grub-legacy and some grub2 installs,
>>>> why would you do that?
>>> Eventually, grub2 will be all that's available from portage. At that
>>> point, I can either,
>>>
>>> 1) Install grub2 on some machines.
>>>
>>> 2) Maintain grub-legacy (and install media) myself.
>> 
>> i really don't think thats the way its going to be. i think there will 
>> be grub and grub2 in portage potentially forever. like with python 2 and 3.
>> 
>> even if not, 2) takes you one cp command and a little bit of disk space 
>> for the grub tarball.
>
> Python2 will stick around because most packages (portage!) don't work
> with python3. Grub doesn't have the same problem.
>
> (2) requires me to at least,
>
>   * Figure out how to build a Gentoo install CD
>   * Fork grub-legacy on our servers somewhere
>   * Test it against all future kernel releases
>   * Document why we're doing this, and how to do the first three steps.
>
>
>>>    * Upgrade a bunch of my servers at 4am?
>> why not choose a convenient time to upgrade?
>
> 4am *is* the convenient time to upgrade.

And usually on a weekend, so when the whole thing goes sideways you've
got at least one day to fix it before "regular business hours" start.
Unless it's a "consumer" server not a "business" server, then you
don't have a "weekend" for fixing stuff that goes wrong.

-- 
Grant Edwards               grant.b.edwards        Yow! ... he dominates the
                                  at               DECADENT SUBWAY SCENE.
                              gmail.com            

Re: [gentoo-user] Re: OT: change and improvement

[Dale]

Grant Edwards wrote:
> On 2011-10-07, Canek Pel??ez Vald??s<caneko@gmail.com>  wrote:
>
>> In the end, udev was making huge advances, and HAL could not keep up
>> simply because the other Operating Systems didn't have similar
>> capabilities, so the consumers of HAL (desktop systems, mostly)
>> started to use udev directly. That was when the shit hit the fan: if
>> the purpose of HAL was to mantain "portability", but the biggest and
>> most active developer community (Linux) refused to use it since it
>> didn't allowed them to use the full capabilities of the operating
>> system, then it had (literally) no reason to live. So the HAL
>> mantainers saw the error of their ways, and they deprecated it,
>> saying to the user space developers that, in Linux, they should use
>> udev, and in other Operating Systems whatever was equivalent, if any.
>>
>> It was really fast, if I remember correctly: one day half the
>> programs in my computers used HAL, and the next every single one of
>> them stopped using it. In Gentoo in particular was pretty rough,
>> since the X.org version that used HAL had just become stable (which
>> was kinda difficult to transition to), and next thing you know, you
>> again had to transition, this time to a HAL-free X.org. A lot of
>> users got really angry in Gentoo because of that.
> Some of us grumpy-old-guy types refused to play nice and didn't make
> either transition.  We disabled HAL support in X.org [after X.org had
> stopped working when HAL became the default].  Once we had HAL
> disabled, we continued to to use the trusty old xorg.conf file until
> the whole HAL thing blew over.
>
> I'm going to ignore grub2 for as long as I can, but I don't think it's
> going away the way HAL did... ;)
>

Some of us old farts tried hal and got bit, hard.  It's the one time I 
can say Linux failed me big time.  I felt like I was running windoze.  O_O

Dale

:-)  :-)