Re: locally built netboot image errors out at partman stage

2026-10-27 Thread Marc Haber

Hi,

after letting this rest for a while I am now ready for another round of 
trying.


On Wed, Jul 15, 2026 at 06:23:48PM +0200, Cyril Brulebois wrote:

Marc Haber  (2026-07-15):

when I boot a qemu machine from
https://d-i.debian.org/daily-images/amd64/daily/netboot/mini.iso, everything
is fine. I can proceed until partman comes up (I abort then, partman is what
interests me).

When I check out
https://salsa.debian.org/installer-team/debian-installer.git, do make
reallyclean, make build_netboot and boot from that mini.iso, at the stage
where partman is supposed to come up, I get a red screen saying that
partitioning disks failed.


Since you're not mentioning switching branch or anything, I'm assuming
you're working from the master branch, and building within unstable


I have followed the instructions, did the mr thing and have rebuilt 
partman-base from the master branch. I also checked out the 241 tag 
which matches the version that seems to currently be in the archive, 
with the same result. I can say that my resulting partman-base has the 
same depends line as partman-base_241_amd64.udeb from the archive, but 
my locally built partman-base_241_amd64.udeb depends on >= 3.8 of 
libparted-fs-resize0-udeb and libparted2-udeb. The error remains when I 
harcode the same depends in my partman-base udev to be >= 3.7 as in the 
archive.



(e.g. a devel-oriented schroot environment).


The docs don't say anything about schroot, I did the build in my daily 
driver system after installing partman-base's Build-Dependendcies.


What environment am I supposed to use when I want to rebuild a single 
udeb while still using the rest of d-i from the archive?



Starting from a clean Git checkout (meaning reallyclean is a no-op),
the build_netboot target gives me a build/dest/netboot/mini.iso that
lets me start and finish partman successfully.


Looks like your build environment differs to mine. How would you rebuild 
it in case you lost it?



The log then says:
Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring package 
partman-base (dep loop?)
Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' failed.

Why does my image behave differently? Any ideas what might be going
wrong here?


Maybe something to do with the build environment we have no information
about.


I apologize for not knowing enough about d-i to be able to give the 
information that you don't ask for.


I am eager for a solution and would be willing to build a cloud VM that 
I could give anyone access to. Maybe that would help identifying what I 
was doing wrong in building my build environment. As an alternative, I 
am prepared to answer questions about my build environment you might be 
willing to ask.


Greetings
Marc

--
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421



Re: locally built netboot image errors out at partman stage

2026-10-27 Thread Marc Haber

Hi,

On Wed, Jul 15, 2026 at 08:31:54PM +0200, John Paul Adrian Glaubitz wrote:

On Wed, 2026-07-15 at 18:23 +0200, Cyril Brulebois wrote:

Starting from a clean Git checkout (meaning reallyclean is a no-op),
the build_netboot target gives me a build/dest/netboot/mini.iso that
lets me start and finish partman successfully.


All I can say is that I ran into this circular dependency issue as well
when building d-i locally with the udebs partman-hfs and hfsprogs-udeb
injected manually and it only went away after I dropped partman-base from
Depends of partman-hfs.


That's the current state of partman-hfs's master branch. I have rebuilt 
partman-hfs as well (since the version in the archive still depends on 
hfs-modules, hfsprogs-udeb, partman-base, partman-basicfilesystems, 
cdebconf-udeb), my locally built partman-hfs just has 
partman-basicfilesystems, cdebconf-udeb.


My circular dependency abort remains.


I was building inside a freshly generated schroot, so the build environment
was clean in my case.


So you built a clean unstable schroot, and did the entire mr checkout in 
the chroot? That would mean that some package that might be installed on 
my daily driver breaks the build?



Since I did not change anything but updating the d-i
repository with "git pull", I assume this to be a regression in d-i itself
or at least a requirement to implement some changes to udebs if dropped to
the "localudebs" directory.


Maybe it's just partman as being the most complex part of the installer?

Greetings
Marc

--
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread John Paul Adrian Glaubitz
Hi Cyril,

On Wed, 2026-07-15 at 18:23 +0200, Cyril Brulebois wrote:
> Marc Haber  (2026-07-15):
> > when I boot a qemu machine from
> > https://d-i.debian.org/daily-images/amd64/daily/netboot/mini.iso, everything
> > is fine. I can proceed until partman comes up (I abort then, partman is what
> > interests me).
> > 
> > When I check out
> > https://salsa.debian.org/installer-team/debian-installer.git, do make
> > reallyclean, make build_netboot and boot from that mini.iso, at the stage
> > where partman is supposed to come up, I get a red screen saying that
> > partitioning disks failed.
> 
> Since you're not mentioning switching branch or anything, I'm assuming
> you're working from the master branch, and building within unstable
> (e.g. a devel-oriented schroot environment).
> 
> Starting from a clean Git checkout (meaning reallyclean is a no-op),
> the build_netboot target gives me a build/dest/netboot/mini.iso that
> lets me start and finish partman successfully.

All I can say is that I ran into this circular dependency issue as well
when building d-i locally with the udebs partman-hfs and hfsprogs-udeb
injected manually and it only went away after I dropped partman-base from
Depends of partman-hfs.

> > The log then says:
> > Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
> > Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring 
> > package partman-base (dep loop?)
> > Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' failed.
> > 
> > Why does my image behave differently? Any ideas what might be going
> > wrong here?
> 
> Maybe something to do with the build environment we have no information
> about.

I was building inside a freshly generated schroot, so the build environment
was clean in my case. Since I did not change anything but updating the d-i
repository with "git pull", I assume this to be a regression in d-i itself
or at least a requirement to implement some changes to udebs if dropped to
the "localudebs" directory.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread Cyril Brulebois
Hi,

Marc Haber  (2026-07-15):
> when I boot a qemu machine from
> https://d-i.debian.org/daily-images/amd64/daily/netboot/mini.iso, everything
> is fine. I can proceed until partman comes up (I abort then, partman is what
> interests me).
> 
> When I check out
> https://salsa.debian.org/installer-team/debian-installer.git, do make
> reallyclean, make build_netboot and boot from that mini.iso, at the stage
> where partman is supposed to come up, I get a red screen saying that
> partitioning disks failed.

Since you're not mentioning switching branch or anything, I'm assuming
you're working from the master branch, and building within unstable
(e.g. a devel-oriented schroot environment).

Starting from a clean Git checkout (meaning reallyclean is a no-op),
the build_netboot target gives me a build/dest/netboot/mini.iso that
lets me start and finish partman successfully.

> The log then says:
> Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
> Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring 
> package partman-base (dep loop?)
> Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' failed.
> 
> Why does my image behave differently? Any ideas what might be going
> wrong here?

Maybe something to do with the build environment we have no information
about.


Cheers,
-- 
Cyril Brulebois ([email protected])
D-I release manager -- Release team member -- Freelance Consultant


signature.asc
Description: PGP signature


Re: locally built netboot image errors out at partman stage

2026-07-15 Thread Pascal Hambourg

On 15/07/2026 at 15:13, John Paul Adrian Glaubitz wrote:

On Wed, 2026-07-15 at 12:51 +0200, Pascal Hambourg wrote:

On 15/07/2026 at 11:09, Marc Haber wrote:


Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring
package partman-base (dep loop?)
Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base'
failed.


Could it be another unforeseen side effect [1] of my SET_FLAG change
which introduced a loop dependency between partman-base and
partman-partitioning ? But then, why would it affect only your local build ?


Yes, it must have been some change that was introduced later. I don't know
why it affects local builds only, but it assume there were no tests with
local builds prior this change was committed, were there?


I do not know about others, but I have not tested with any local build. 
I only may have tested with the mini.iso generated by salsa CI.


If circular dependencies in udebs are an issue, then IMO partman-base 
udeb should be split in two: one udeb providing the common resources 
which other partman-* udebs depend on, and one providing the main-menu 
entry which depends on other partman-* udebs.





Re: locally built netboot image errors out at partman stage

2026-07-15 Thread Marc Haber

On Wed, Jul 15, 2026 at 03:12:33PM +0200, John Paul Adrian Glaubitz wrote:

On Wed, 2026-07-15 at 12:22 +0200, Marc Haber wrote:

On Wed, Jul 15, 2026 at 11:22:20AM +0200, John Paul Adrian Glaubitz wrote:
> I have had this issue as well when using partman-hfs in a local build and I
> fixed it by dropping partman-base from Depends in debian/control for the
> partman-hfs package. There seems to be an implicit dependency now so that
> the explicit dependency on partman-base is not necessary.

That didn't work for me. For the image to build, I also had to drop
hfs-modules and hfsprogs-udeb from partman-hfs's depends, but in the end
the error persists.


I didn't say you need to patch partman-hfs, but whatever partman-* package
that are shipping locally.


The issue also happens when I build a mini.iso without my locally 
patched partman-base. Which patches do you suggest doing to make it 
build?


Greetings
Marc

--
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread Marc Haber

On Wed, Jul 15, 2026 at 03:13:42PM +0200, John Paul Adrian Glaubitz wrote:

On Wed, 2026-07-15 at 12:51 +0200, Pascal Hambourg wrote:

On 15/07/2026 at 11:09, Marc Haber wrote:
>
> Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
> Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring
> package partman-base (dep loop?)
> Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base'
> failed.

Could it be another unforeseen side effect [1] of my SET_FLAG change
which introduced a loop dependency between partman-base and
partman-partitioning ? But then, why would it affect only your local build ?


Yes, it must have been some change that was introduced later. I don't know
why it affects local builds only, but it assume there were no tests with
local builds prior this change was committed, were there?


Is it documented how the daily mini.isos that get uploaded to the web 
page get built?


Greetings
Marc

--
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread John Paul Adrian Glaubitz
On Wed, 2026-07-15 at 12:51 +0200, Pascal Hambourg wrote:
> On 15/07/2026 at 11:09, Marc Haber wrote:
> > 
> > Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
> > Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring 
> > package partman-base (dep loop?)
> > Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' 
> > failed.
> 
> Could it be another unforeseen side effect [1] of my SET_FLAG change 
> which introduced a loop dependency between partman-base and 
> partman-partitioning ? But then, why would it affect only your local build ?

Yes, it must have been some change that was introduced later. I don't know
why it affects local builds only, but it assume there were no tests with
local builds prior this change was committed, were there?

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread John Paul Adrian Glaubitz
Hi Marc,

On Wed, 2026-07-15 at 12:22 +0200, Marc Haber wrote:
> On Wed, Jul 15, 2026 at 11:22:20AM +0200, John Paul Adrian Glaubitz wrote:
> > I have had this issue as well when using partman-hfs in a local build and I
> > fixed it by dropping partman-base from Depends in debian/control for the
> > partman-hfs package. There seems to be an implicit dependency now so that
> > the explicit dependency on partman-base is not necessary.
> 
> That didn't work for me. For the image to build, I also had to drop 
> hfs-modules and hfsprogs-udeb from partman-hfs's depends, but in the end 
> the error persists.

I didn't say you need to patch partman-hfs, but whatever partman-* package
that are shipping locally. It was just partman-hfs in my case as I always
have to build debian-installer with partman-hfs/hfsprogs-udeb added locally
because Debian doesn't consider Apple's APSL license dfsg-compliant which
is why hfsprogs was moved to non-free at some point.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread Pascal Hambourg

On 15/07/2026 at 11:09, Marc Haber wrote:


Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring 
package partman-base (dep loop?)
Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' 
failed.


Could it be another unforeseen side effect [1] of my SET_FLAG change 
which introduced a loop dependency between partman-base and 
partman-partitioning ? But then, why would it affect only your local build ?


[1] See threads beginning with





Re: locally built netboot image errors out at partman stage

2026-07-15 Thread Marc Haber

Hi,

On Wed, Jul 15, 2026 at 11:22:20AM +0200, John Paul Adrian Glaubitz wrote:

I have had this issue as well when using partman-hfs in a local build and I
fixed it by dropping partman-base from Depends in debian/control for the
partman-hfs package. There seems to be an implicit dependency now so that
the explicit dependency on partman-base is not necessary.


That didn't work for me. For the image to build, I also had to drop 
hfs-modules and hfsprogs-udeb from partman-hfs's depends, but in the end 
the error persists.


Greetings
Marc

--
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421



Re: locally built netboot image errors out at partman stage

2026-07-15 Thread John Paul Adrian Glaubitz
Hi Marc,

On Wed, 2026-07-15 at 11:09 +0200, Marc Haber wrote:
> when I boot a qemu machine from 
> https://d-i.debian.org/daily-images/amd64/daily/netboot/mini.iso, 
> everything is fine. I can proceed until partman comes up (I abort then, 
> partman is what interests me).
> 
> When I check out 
> https://salsa.debian.org/installer-team/debian-installer.git, do make 
> reallyclean, make build_netboot and boot from that mini.iso, at the 
> stage where partman is supposed to come up, I get a red screen saying 
> that partitioning disks failed.
> 
> The log then says:
> Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
> Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring 
> package partman-base (dep loop?)
> Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' failed.
> 
> Why does my image behave differently? Any ideas what might be going 
> wrong here?

I have had this issue as well when using partman-hfs in a local build and I
fixed it by dropping partman-base from Depends in debian/control for the
partman-hfs package. There seems to be an implicit dependency now so that
the explicit dependency on partman-base is not necessary.

Adrian

-- 
 .''`.  John Paul Adrian Glaubitz
: :' :  Debian Developer
`. `'   Physicist
  `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913



locally built netboot image errors out at partman stage

2026-07-15 Thread Marc Haber

Hi,

when I boot a qemu machine from 
https://d-i.debian.org/daily-images/amd64/daily/netboot/mini.iso, 
everything is fine. I can proceed until partman comes up (I abort then, 
partman is what interests me).


When I check out 
https://salsa.debian.org/installer-team/debian-installer.git, do make 
reallyclean, make build_netboot and boot from that mini.iso, at the 
stage where partman is supposed to come up, I get a red screen saying 
that partitioning disks failed.


The log then says:
Jul 15 09:03:01 main-menu[259]: INFO: Menu item 'partman-base' selected
Jul 15 09:03:01 main-menu[259]: WARNING **: Deep recursion configuring package 
partman-base (dep loop?)
Jul 15 09:03:01 main-menu[259]: WARNING **: Menu item 'partman-base' failed.

Why does my image behave differently? Any ideas what might be going 
wrong here?


Greetings
Marc


--
-
Marc Haber | "I don't trust Computers. They | Mailadresse im Header
Leimen, Germany|  lose things."Winona Ryder | Fon: *49 6224 1600402
Nordisch by Nature |  How to make an American Quilt | Fax: *49 6224 1600421