[core-updates] Due to flawed bootstrap tarball, full rebuild needed

[core-updates] Due to flawed bootstrap tarball, full rebuild needed

[Mark H Weaver]

Hello Guix,

It turns out that the MesCC bootstrap tarball includes store references,
and moreover it references a different store item than the MesCC
bootstrap tarball that I built locally on my own machine.  See
<https://issues.guix.gnu.org/issue/36747> for details and a proposed
patch from Janneke to fix the problem.

In my opinion, we should update the bootstrap binaries and rebuild
'core-updates' from scratch.  What do other people think?

       Mark

Re: [core-updates] Due to flawed bootstrap tarball, full rebuild needed

[Jan Nieuwenhuizen]

Mark H Weaver writes:

> It turns out that the MesCC bootstrap tarball includes store references,
> and moreover it references a different store item than the MesCC
> bootstrap tarball that I built locally on my own machine.  See
> <https://issues.guix.gnu.org/issue/36747> for details and a proposed
> patch from Janneke to fix the problem.
>
> In my opinion, we should update the bootstrap binaries and rebuild
> 'core-updates' from scratch.  What do other people think?

Yes, I second that. I have just successfully built `hello' again with
the new set of tarballs; IWBN if you could verify the new, stripped
set.

janneke

-- 
Jan Nieuwenhuizen <janneke@gnu.org> | GNU LilyPond http://lilypond.org
Freelance IT http://JoyofSource.com | Avatar® http://AvatarAcademy.com

Re: [core-updates] Due to flawed bootstrap tarball, full rebuild needed

[Chris Marusich]

[-- Attachment #1: Type: text/plain, Size: 584 bytes --]

Mark H Weaver <mhw@netris.org> writes:

> It turns out that the MesCC bootstrap tarball includes store references,
> and moreover it references a different store item than the MesCC
> bootstrap tarball that I built locally on my own machine.  See
> <https://issues.guix.gnu.org/issue/36747> for details and a proposed
> patch from Janneke to fix the problem.
>
> In my opinion, we should update the bootstrap binaries and rebuild
> 'core-updates' from scratch.  What do other people think?

I think it's a good idea, and I'm glad you took the time to check.

-- 
Chris

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 832 bytes --]

Re: [core-updates] Due to flawed bootstrap tarball, full rebuild needed

[Marius Bakke]

Chris Marusich <cmmarusich@gmail.com> writes:

> Mark H Weaver <mhw@netris.org> writes:
>
>> It turns out that the MesCC bootstrap tarball includes store references,
>> and moreover it references a different store item than the MesCC
>> bootstrap tarball that I built locally on my own machine.  See
>> <https://issues.guix.gnu.org/issue/36747> for details and a proposed
>> patch from Janneke to fix the problem.
>>
>> In my opinion, we should update the bootstrap binaries and rebuild
>> 'core-updates' from scratch.  What do other people think?
>
> I think it's a good idea, and I'm glad you took the time to check.

I agree 100%.  Thank you Mark!

Re: [core-updates] Due to flawed bootstrap tarball, full rebuild needed

[Mark H Weaver]

Marius Bakke <mbakke@fastmail.com> writes:

> Chris Marusich <cmmarusich@gmail.com> writes:
>
>> Mark H Weaver <mhw@netris.org> writes:
>>
>>> It turns out that the MesCC bootstrap tarball includes store references,
>>> and moreover it references a different store item than the MesCC
>>> bootstrap tarball that I built locally on my own machine.  See
>>> <https://issues.guix.gnu.org/issue/36747> for details and a proposed
>>> patch from Janneke to fix the problem.
>>>
>>> In my opinion, we should update the bootstrap binaries and rebuild
>>> 'core-updates' from scratch.  What do other people think?
>>
>> I think it's a good idea, and I'm glad you took the time to check.
>
> I agree 100%.  Thank you Mark!

Great, thanks!

I hope that it will take only a few days, or perhaps a week, to verify
and finalize the new bootstrap binaries.  In the meantime, I suggest
that you continue your work on core-updates with no regard to this
issue, and continuing to build it out on Berlin.

The good news is that the only cost should be the time it takes for
Berlin to rebuild 'core-updates'.  It should not affect anything else
beyond the early bootstrap, except of course for different hashes
everywhere.

       Mark