Bug#1066145: debootstrap: fails with multiversion repositories, including Debian’s

[Thorsten Glaser]

Package: debootstrap
Version: 1.0.134
Severity: important

There exist multiversion Packages files, e.g. they can be created
by dpkg-scanpackages -m but dak also occasionally seems to create
them.

I just had multiple attempts at creating a buildd chroot fail.

One was with debootstrap --variant=buildd where perl-modules-5.38
failed the bootstrap because the Packages file contains two versions
of it (5.38.2-3.2 (the one we want) followed by 5.38.2-3). According
to cbmuser, the arch:all part of debian-ports Packages files is an
identical embedded copy of binaries-all/Packages from dak for sid,
so this is not just a debian-ports or mini-dak problem.

Then I tried --variant=minbase which failed the cowbuilder --create
as well but only after the debootstrap stage: debootstrap installed
libaudit-common 1:3.1.2-2 instead of 1:3.1.2-2.1 so the chroot had
libaudit1’s dependency on libaudit-common unfulfilled, which made a
later apt-get dist-upgrade fail without manually running an apt-get
-f install first (which I thankfully could do in a pbuilder hook
script).

This is a real problem affecting real-world scenarios (such as
buildd maintainers frantically trying to stay on top of t64), and
apparently at least arch:all packages can have multiple versions
show up in the Packages file in Debian sid.

The two failure modes are that debootstrap either results in a
chroot in which packages are installed that don’t meet each other’s
dependencies (which a real-world dpkg call would need --force-depends)
or that debootstrap itself even aborts.

The fix is of course to consider all versions of all packages,
though this will be very hard; apt seems to normally only consider
the latest version (unless another is installed or pinned), so
doing that would be another option, which while not being helpful
in situations of nōn-arch:all packages lagging would fix this bug
just as well.


-- System Information:
Debian Release: stretch/sid
  APT prefers unreleased
  APT policy: (500, 'unreleased'), (500, 'unstable')
Architecture: m68k

Kernel: Linux 6.6.15-m68k
Locale: LANG=C, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/lksh
Init: sysvinit (via /sbin/init)

Versions of packages debootstrap depends on:
ii  wget  1.16.3-3

Versions of packages debootstrap recommends:
pn  arch-test   
ii  debian-archive-keyring  2014.3
ii  gnupg   1.4.19-5
ii  mount   2.26.2-9

Versions of packages debootstrap suggests:
ii  binutils2.25.1-1
pn  squid-deb-proxy-client  
pn  ubuntu-archive-keyring  
ii  xz-utils5.1.1alpha+20120614-2.1
pn  zstd

-- no debconf information

Bug#921815: Versionnumber for closing the bug report

[Thorsten Glaser]

close 921815 1.0.114+deb10u1
thanks

On Mon, 27 Nov 2023, Roland Clobus wrote:

> Do you want to have a more exact version? Which I closed this issue, I used 
> the
> version that is present in Buster. (Version: 1.0.114+deb10u1)
> Going deeper into older versions didn't seen necessary to me.

Yeah, that’s probably fine, thanks.

Bug#921815: Works in buster

[Thorsten Glaser]

reopen 921815
thanks

On Wed, 8 Nov 2023, Roland Clobus wrote:

> While triaging #919659, I found this bug report.
[…]
> It works, debootstrap reports:
> I: Base system installed successfully.

But did you also test whether /proc was not umounted?

bye,
//mirabilos
-- 
Infrastrukturexperte • Qvest Digital AG
Am Dickobskreuz 10, D-53121 Bonn • https://www.qvest-digital.com/
Telephon +49 228 54881-393 • Fax: +49 228 54881-235
HRB AG Bonn 18196 • USt-ID (VAT): DE274355441
Vorstand: Dr. Stefan Barth, Kai Ebenrett, Boris Esser, Alexander Steeg
Vorsitzender Aufsichtsrat: Peter Nöthen

weird interaction (bug?) between two-stage debootstrap and dpkg-divert

[Thorsten Glaser]

Hi,

I ran into this because I have an elaborate chroot script¹ that
handles start-stop-daemon, initctl, policy-rc.d, etc. and can
reproduce this as follows:

I’ve debootstrapped bullseye up to first stage, then entered
the chroot and:

# dpkg-divert --local --quiet --rename --divert /sbin/start-stop-daemon.REAL 
--add /sbin/start-stop-daemon

Then I created a new /sbin/start-stop-daemon. Running
“ls -l /sbin/start-stop-daemon*” confirms both exist and are
executable. Also:

0|I have no name!@tglase:/# dpkg-divert --list
local diversion of /sbin/start-stop-daemon to /sbin/start-stop-daemon.REAL

Then I did:

# /debootstrap/debootstrap --second-stage
[…]
I: Configuring libc-bin...
I: Base system installed successfully.
0|I have no name!@tglase:/# ls -l /sbin/start-stop-daemon*
-rwxr-xr-x 1 root root 94 Nov 27 04:06 /sbin/start-stop-daemon
0|I have no name!@tglase:/# dpkg-divert --list
diversion of /usr/share/man/man1/sh.1.gz to /usr/share/man/man1/sh.distrib.1.gz 
by dash
local diversion of /sbin/start-stop-daemon to /sbin/start-stop-daemon.REAL
diversion of /bin/sh to /bin/sh.distrib by dash

So, the diversion is retained but the *diverted* file is, for
some reason, removed.

Did I stumble upon a bug or is this to be expected?


① 
https://evolvis.org/plugins/scmgit/cgi-bin/gitweb.cgi?p=shellsnippets/shellsnippets.git;a=blob;f=posix/sysadmin/debchroot.sh;hb=HEAD

bye,
//mirabilos
-- 
16:47⎜«mika:#grml» .oO(mira ist einfach gut)  23:22⎜«mikap:#grml»
mirabilos: und dein bootloader ist geil :)23:29⎜«mikap:#grml» und ich
finds saugeil dass ich ein bsd zum booten mit grml hab, das muss ich dann
gleich mal auf usb-stick installieren   -- Michael Prokop über MirOS bsd4grml

Bug#684128: src:debian-installer: allow use of binary units in disk partitioner

[Thorsten Glaser]

Holger Wansing dixit:

>>Could this information (valid unit sufficēs) be added to the dialogue
>>where the size is entered? Screen space should suffice.
>
>Yes, I already thought about if changing the template would make sense here.

Thanks!

Could we also get the size output in both formats? I realise that
will most likely not be a change as simple…

bye,
//mirabilos
-- 
15:41⎜ Somebody write a testsuite for helloworld :-)

Bug#684128: src:debian-installer: allow use of binary units in disk partitioner

[Thorsten Glaser]

reopen 684128
retitle 684128 src:debian-installer: show ISO/IEC 60027-2 units (as well); show 
valid suffixes
found 684128 226
thanks

Holger Wansing dixit:

>So I guess it works as it should.
>
>The (visual) output is still in MB / GB, apart from this a see no issue.

Hrm, the output being only in one format can be a problem, but it’s
not as critical. I would still love for *that* to be fixed, i.e. either
only ISO/IEC 60027-2 units or both those and decimal.

>With a 12.0 netinst image, creating a new partition with a size of 512 MiB
>results in a parition of 536 MB

OK, thanks for confirming. I heard twohundred-something across the room,
which would have been quite off, but didn’t have a chance to visually
inspect myself.

Could this information (valid unit sufficēs) be added to the dialogue
where the size is entered? Screen space should suffice.

I’ll repurpose this bug for that (both input and output changes).

Thanks,
//mirabilos
-- 
  "Using Lynx is like wearing a really good pair of shades: cuts out
   the glare and harmful UV (ultra-vanity), and you feel so-o-o COOL."
 -- Henry Nelson, March 1999

Bug#684128: src:debian-installer: allow use of binary units in disk partitioner

[Thorsten Glaser]

Cyril Brulebois dixit:

>https://www.debian.org/devel/debian-installer/News/2023/20230516.en.html
>documents a fix for #913431, which is a duplicate of this bug report.

Huh.

>> In bookworm d-i, entering 512 MiB seems to be using something
>> entirely different
>
>“Something entirely different” compared to what?

I was a few metres across the room, but I heard something around
twohundred-something, which was definitely no power of two either.

>The second syntax is indeed invalid.

OK, it was only tried as last resort anyway.

>> This still makes d-i unsuitable for most partitioning.
>
>Feel free to use a different tool.

Oh, I personally do, but not everyone is I.

bye,
//mirabilos
-- 
22:20⎜ The crazy that persists in his craziness becomes a master
22:21⎜ And the distance between the craziness and geniality is
only measured by the success 18:35⎜ "Psychotics are consistently
inconsistent. The essence of sanity is to be inconsistently inconsistent

Bug#684128: src:debian-installer: allow use of binary units in disk partitioner

[Thorsten Glaser]

found 684128 226
thanks

Hi,

why is this bug still unfixed?

In bookworm d-i, entering 512 MiB seems to be using something
entirely different, and 512 Mi gives an error “invalid size”.

This still makes d-i unsuitable for most partitioning.

bye,
//mirabilos
-- 
[...] if maybe ext3fs wasn't a better pick, or jfs, or maybe reiserfs, oh but
what about xfs, and if only i had waited until reiser4 was ready... in the be-
ginning, there was ffs, and in the middle, there was ffs, and at the end, there
was still ffs, and the sys admins knew it was good. :)  -- Ted Unangst über *fs

Bug#992034: init choice in Debian installation instructions

[Thorsten Glaser]

Hi,

as the content for the release notes was suggested to be put into the
Wiki (instead?) anyway, how about, to lower translator burden, there
*will* be put a section about this into the installation guide, but one
that is mostly comprised of a link to the Wiki, with a short intro.

@Matthew: that’s most certainly an em dash there; normally, you render
an em dash as U+2014 surrounded by U+200A (hair space) on both sides
(some US-american publishers seem to omit the hair spaces; in Tₑχ/LᴬTᴇΧ
I personally realise them as ⅙em which may be slightly wider than the
hair spaces of some publishers but find this makes it easier to read).
Now I have no idea how this is best done in the sources for the Debian
installation guide…

bye,
//mirabilos
-- 
Gestern Nacht ist mein IRC-Netzwerk explodiert. Ich hatte nicht damit
gerechnet, darum bin ich blutverschmiert… wer konnte ahnen, daß SIE so
reagier’n… gestern Nacht ist mein IRC-Netzwerk explodiert~~~
(as of 2021-06-15 The MirOS Project temporarily reconvenes on OFTC)

Bug#901332: d-i: Offer to shut down / power off instead of reboot at the end

[Thorsten Glaser]

Hi Phil,

>BTW one can preseed this behaviour with 'debian-installer/exit/halt' or
>'debian-installer/exit/poweroff' as mentioned here:
>
>  https://www.debian.org/releases/stable/amd64/apbs04.en.html#preseed-finish

oh, good to know.

>which means that you could specify such a setting on the kernel command
>line when starting debian-installer to get the behaviour you want.

I’m not versed in preseeding… so that means I have to write
debian-installer/exit/poweroff=true to the kernel commandline?

Wow, quite a handful to type (but on the plus side, it can be
added to netboot menus right now, no further change needed, so
it solves half the problem ☺).

Thanks,
//mirabilos
-- 
Gestern Nacht ist mein IRC-Netzwerk explodiert. Ich hatte nicht damit
gerechnet, darum bin ich blutverschmiert… wer konnte ahnen, daß SIE so
reagier’n… gestern Nacht ist mein IRC-Netzwerk explodiert~~~
(as of 2021-06-15 The MirOS Project temporarily reconvenes on OFTC)

Bug#901332: d-i: Offer to shut down / power off instead of reboot at the end

[Thorsten Glaser]

Package: debian-installer
Followup-For: Bug #901332
X-Debbugs-Cc: t...@mirbsd.de

Did anything ever come from this, now that we’re nearing a release?


-- System Information:
Debian Release: 11.0
  APT prefers testing-security
  APT policy: (500, 'testing-security'), (500, 'stable-updates'), (500, 
'oldoldstable'), (500, 'testing'), (500, 'stable'), (500, 'oldstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.10.0-8-amd64 (SMP w/2 CPU threads)
Locale: LANG=C.UTF-8, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE not set
Shell: /bin/sh linked to /bin/lksh
Init: sysvinit (via /sbin/init)

Re: Tentative summary of the AMD/ATI/NVidia issue

[Thorsten Glaser]

Cyril Brulebois dixit:

>Lucas Nussbaum  (2021-04-24):

>> C) Do nothing and document this in the release notes
>
>As said above, I strongly recommend against this.

Right… what about, add another Plan C…

C) When X won’t work, fail gracefully, show a console login

… and dump the above to Plan D?

(Perhaps doing this is a good idea in general, for when a
similar issue pops up in the future?)

bye,
//mirabilos
-- 
22:59⎜ glaub ich termkit is kompliziert | glabe nicht das man
damit schneller arbeitet | reizüberflutung │ wie windows │ alles evil
zuviel bilder │ wie ein spiel | 23:00⎜ die meisten raffen auch
nicht mehr von windows | 23:01⎜ bilderbücher sind ja auch nich
wirklich verbreitet als erwachsenen literatur   ‣ who needs GUIs thus?

Bug#986025: installation-reports: weird characters in network-console

[Thorsten Glaser]

Package: installation-reports
Severity: normal
X-Debbugs-Cc: t...@mirbsd.de

Boot method: CD
Image version: 
https://d-i.debian.org/daily-images/ppc64el/daily/netboot/mini.iso Mar 26 18:04
Date: 2021-03-26 to 2021-03-28 (yes it took three days, the machine was busy)

Machine: qemu-system-ppc64le
Partitions:
root@ppc64el:~# df -Tl
Filesystem Type 1K-blocksUsed Available Use% Mounted on
udev   devtmpfs   2033152   0   2033152   0% /dev
tmpfs  tmpfs   4163205312411008   2% /run
/dev/vda2  ext4  48232140 1726160  44023448   4% /
tmpfs  tmpfs  2081472   0   2081472   0% /dev/shm
tmpfs  tmpfs 5120   0  5120   0% /run/lock
tmpfs  tmpfs  2081472   0   2081472   0% 
/var/cache/pbuilder/build/cow.1965/dev/shm
root@ppc64el:~# fdisk -l
Disk /dev/vda: 48 GiB, 51539607552 bytes, 100663296 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: D3EC9994-AFF1-4A12-BC03-3D2424D74D64

DeviceStart   End  Sectors  Size Type
/dev/vda1  2048 16383143367M PowerPC PReP boot
/dev/vda2 16384  98662399 98646016   47G Linux filesystem
/dev/vda3  98662400 100661247  1998848  976M Linux swap


Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[O]
Configure network:  [O]
Detect media:   [O]
Load installer modules: [O]
Clock/timezone setup:   [O]
User/password setup:[O]
Detect hard drives: [O]
Partition hard drives:  [O]
Install base system:[O]
Install tasks:  [O]
Install boot loader:[O]
Overall install:[E]

Comments/Problems:

Things worked well, except I switched to network-console and
had weird characters instead of line drawing, although both
outside and inside were GNU screen in UTF-8 mode.

I ended up with a UsrMerge system :(


Please make sure that any installation logs that you think would
be useful are attached to this report. Please compress large
files using gzip.


-- Package-specific info:

==
Installer lsb-release:
==
DISTRIB_ID=Debian
DISTRIB_DESCRIPTION="Debian GNU/Linux installer"
DISTRIB_RELEASE="11 (bullseye) - installer build 20210326-04:02:59"
X_INSTALLATION_MEDIUM=netboot

==
Installer hardware-summary:
==
uname -a: Linux ppc64el 5.10.0-5-powerpc64le #1 SMP Debian 5.10.24-1 
(2021-03-19) ppc64le GNU/Linux
lspci -knn: 00:01.0 Ethernet controller [0200]: Red Hat, Inc. Virtio network 
device [1af4:1000]
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:0001]
lspci -knn: Kernel driver in use: virtio-pci
lspci -knn: Kernel modules: virtio_pci
lspci -knn: 00:02.0 SCSI storage controller [0100]: Red Hat, Inc. Virtio SCSI 
[1af4:1004]
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:0008]
lspci -knn: Kernel driver in use: virtio-pci
lspci -knn: Kernel modules: virtio_pci
lspci -knn: 00:03.0 USB controller [0c03]: Red Hat, Inc. QEMU XHCI Host 
Controller [1b36:000d] (rev 01)
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:1100]
lspci -knn: Kernel driver in use: xhci_hcd
lspci -knn: Kernel modules: xhci_pci
lspci -knn: 00:04.0 Communication controller [0780]: Red Hat, Inc. Virtio 
console [1af4:1003]
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:0003]
lspci -knn: Kernel driver in use: virtio-pci
lspci -knn: Kernel modules: virtio_pci
lspci -knn: 00:05.0 SCSI storage controller [0100]: Red Hat, Inc. Virtio block 
device [1af4:1001]
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:0002]
lspci -knn: Kernel driver in use: virtio-pci
lspci -knn: Kernel modules: virtio_pci
lspci -knn: 00:06.0 Unclassified device [00ff]: Red Hat, Inc. Virtio memory 
balloon [1af4:1002]
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:0005]
lspci -knn: Kernel driver in use: virtio-pci
lspci -knn: Kernel modules: virtio_pci
lspci -knn: 00:07.0 Unclassified device [00ff]: Red Hat, Inc. Virtio RNG 
[1af4:1005]
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:0004]
lspci -knn: Kernel driver in use: virtio-pci
lspci -knn: Kernel modules: virtio_pci
lspci -knn: 00:08.0 VGA compatible controller [0300]: Device [1234:] (rev 
02)
lspci -knn: Subsystem: Red Hat, Inc. Device [1af4:1100]
lspci -knn: 00:09.0 SATA controller [0106]: Intel Corporation 82801IR/IO/IH 
(ICH9R/DO/DH) 6 port SATA Controller [AHCI mode] [8086:2922] (rev 02)
lspci -knn: Subsystem: Red Hat, Inc. QEMU Virtual Machine [1af4:1100]
lspci -knn: Kernel driver in use: ahci
lspci -knn: Kernel modules: ahci
usb-list: 
usb-list: Bus 01 Device 01: xHCI Host Controller [1d6b:0002]
usb-list:Level 

Monolithic image questions

[Thorsten Glaser]

Hi,

another case of multiple wonderings, this time moving closer
to “could be merged” land:

• Why are the monolithic images not built by default?

  I’ve been using a local build of them for ages with great
  success; all others are fragile wrt. udeb availability in
  at least netboot scenarios.

• Why are the monolithic images not available for all architectures?

  Conversely making them more useful (perhaps).

• Could network-console be included in the monolithic images?

  This is so much smoother, especially on virtual machines
  (although I found the serial console to be just as smooth)
  and allows copy/paste from the host system more easily.

  I think this is just a #include more.

Thanks in advance,
//mirabilos
-- 
„Cool, /usr/share/doc/mksh/examples/uhr.gz ist ja ein Grund,
mksh auf jedem System zu installieren.“
-- XTaran auf der OpenRheinRuhr, ganz begeistert
(EN: “[…]uhr.gz is a reason to install mksh on every system.”)

Change tasksel defaults

[Thorsten Glaser]

Hi,

I’m wondering how to invoke/change d-i so that the defaults for
tasksel are changed but the selection is still shown.

AIUI tasksel is run in the installed system, so patching it and
including a patched build in a self-built d-i image won’t do.

Specifically, I want to preselect no desktop and not “standard”
system utilities… so, basically, everything off except SSH server
which should be on.

I think I can add desktop=none to the kernel command line for the
installer to achieve the first of the two objectives (please correct
me if I’m wrong), but from looking at tasksel I don’t see how I can
achieve the second.

I can preseed tasksel/first but then the selection isn’t shown.

Thanks in advance,
//mirabilos
-- 
15:41⎜ Somebody write a testsuite for helloworld :-)

Bug#986009: installation-reports: document qemu workarounds and bug in newer d-i image

[Thorsten Glaser]

Package: installation-reports
Severity: normal
Tags: d-i
X-Debbugs-Cc: t...@mirbsd.de

Boot method: CD
Image version: 
https://cdimage.debian.org/cdimage/ports/snapshots/2020-10-12/debian-10.0.0-alpha-NETINST-1.iso
Date: 2021-03-26 to 2021-03-27

Machine: qemu-system-alpha on a buster host
Partitions: 

root@alpha:~# fdisk -l  
|
Disk /dev/sda: 40 GiB, 42949672960 bytes, 83886080 sectors  
|
Disk model: QEMU HARDDISK   
|
Geometry: 255 heads, 2 sectors/track, 164482 cylinders  
|
Units: sectors of 1 * 512 = 512 bytes   
|
Sector size (logical/physical): 512 bytes / 512 bytes   
|
I/O size (minimum/optimal): 512 bytes / 512 bytes   
|
Disklabel type: bsd 
|

|
SliceStart  End  Sectors   Size Type Fsize Bsize Cpg
|
a1 1954 1954   977K ext2 0 0   0
|
b 1955   353517   351563 171.7M ext2 0 0   0
|
c   353518 81884768 81531251  38.9G ext2 0 0   0
|
d 81884769 83886078  2001310 977.2M swap 0 0   0
|
root@alpha:~# df -Tl
|
Filesystem Type 1K-blocksUsed Available Use% Mounted on 
|
udev   devtmpfs   2052568   0   2052568   0% /dev
tmpfs  tmpfs   413304 472412832   1% /run
/dev/sda3  ext4  39860300 1123656  36682092   3% /
tmpfs  tmpfs  2066512   0   2066512   0% /dev/shm
tmpfs  tmpfs 5120   0  5120   0% /run/lock
/dev/sda2  ext2170219   68778 92653  43% /boot
tmpfs  tmpfs   413296   0413296   0% /run/user/0


Base System Installation Checklist:
[O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it

Initial boot:   [O]
Detect network card:[O]
Configure network:  [O]
Detect media:   [O]
Load installer modules: [O]
Clock/timezone setup:   [O]
User/password setup:[O]
Detect hard drives: [O]
Partition hard drives:  [O]
Install base system:[E]
Install tasks:  [O]
Install boot loader:[?]
Overall install:[↓]

Comments/Problems:

- 
https://cdimage.debian.org/cdimage/ports/debian-installer/2021-01-03/alpha/debian-installer-images_20201202+nmu1_alpha.tar.gz
  is insufficient, it lacks the ISO which contains nic-modules
- Tried 
https://cdimage.debian.org/cdimage/ports/snapshots/2021-02-01/debian-10.0.0-alpha-NETINST-1.iso
  first but it contains errors, e.g. no /var/lib/dpkg/status,
  thus used a random older image; turned out to work
- Ran into #985954 so used raw qemu instead of libvirt/virt-manager
- qemu-system-alpha’s SRM firmware ("BIOS") seems to be unable to do
  more than set up the bus and echo back REPL commands instead of executing
  them, so only direct kernel boot is usable, so I copied out the kernel and
  initrd from the ISO and ran:

  qemu-system-alpha -cpu ev68-alpha-cpu -m 2048 -drive 
format=raw,file=/dev/vg-evolvis-wirt/alpha -drive 
format=raw,file=debian-10.0.0-alpha-NETINST-1.iso -nographic -net nic -net user 
-kernel vmlinux-alpha -initrd initrd-alpha.gz -append console=ttyS0

  Installation using the serial console, including screen, worked fine.
- After install, booting from the HDD failed (see above), so I used
  partprobe (kpartx, normally the tool of choice, failed to recognise
  the BSD disklabel used by SRM and thus Linux/Alpha) to map the boot
  partition onto the host and copied kernel and initrd out and boot with:

  qemu-system-alpha -cpu ev68-alpha-cpu -m 4096 -drive 
format=raw,file=/dev/vg-evolvis-wirt/alpha -nographic -net nic -net user 
-kernel ~/vmlinux-5.10.0-5-alpha-generic -initrd 
~/initrd.img-5.10.0-5-alpha-generic -append 'console=ttyS0 
root=UUID=6381e09c-4567-4fff-bbe4-608973308f65'
- After install, getting tons of unaligned messages from dbus:
  [  320.117023] dbus-daemon(366): unaligned trap at 02080bf0: 
f6857c0e 28 18
⚠ Unfortunately there was no way to disable usrmerge in d-i,
  so I’m now left with a system I cannot build on and have
  to resort to debootstrapping me a chroot



Please make sure that any installation logs that you think would
be useful are attached to this report. Please compress large
files using gzip.


-- Package-specific info:

==
Installer lsb-release:
==
DISTRIB_ID=Debian
DISTRIB_DESCRIPTION="Debian GNU/Linux installer"
DISTRIB_RELEASE="11 (bullseye) - installer build 20200315"
X_INSTALLATION_MEDIUM=cdrom

Re: Remove of lilo-installer

[Thorsten Glaser]

John Paul Adrian Glaubitz dixit:

>No idea what the stance on i386 is. But I think Thorsten is a LILO user.

No, I haven’t been using LILO for a long time either.
I _did_ use it about a decade ago, when I got back to Debian
after using only BSD for a while, but even before that, I
used loadlin more, and discovered grub during the BSD-only
time (considering it could do PXE).

Why the removal? Does it not work any more for some reason?
Or is this just about the d-i integration?

bye,
//mirabilos
-- 
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font.   -- Rob Pike in "Notes on Programming in C"

Bug#700633: Debootstrap is very slow. Please use eatmydata to fix this.

[Thorsten Glaser]

Version: 1.0.115

Hi everyone,

I’m appalled that this feature request is still open…

I’ve got a need to do this without patching debootstrap, and have had
success (although I did not time it, but someone else can ;-) with this
sequence of commands (assuming dpkg-deb is available ― which my script
checks earlier):

safe_PATH=/bin:/sbin:/usr/bin:/usr/sbin
# […]
case $TERM in
(Eterm|Eterm-color|ansi|cons25|cons25-debian|cygwin|dumb|hurd|linux|mach|mach-bold|mach-color|mach-gnu|mach-gnu-color|pcansi|rxvt|rxvt-basic|rxvt-m|rxvt-unicode|rxvt-unicode-256color|screen|screen-256color|screen-256color-bce|screen-bce|screen-s|screen-w|screen.xterm-256color|sun|vt100|vt102|vt220|vt52|wsvt25|wsvt25m|xterm|xterm-256color|xterm-color|xterm-debian|xterm-mono|xterm-r5|xterm-r6|xterm-vt220|xterm-xfree86)
# list from ncurses-base (6.1+20181013-2+deb10u1)
;;
(screen.*|screen-*)
# aliases possibly from ncurses-term
TERM=screen ;;
(rxvt.*|rxvt-*)
# let’s hope…
TERM=rxvt ;;
(xterm.*|xterm-*)
# …this works…
TERM=xterm ;;
(linux.*)
# …probably
TERM=linux ;;
(*)
die "Your terminal type '$TERM' is not supported by ncurses-base." \
'Maybe run this script in GNU screen?' ;;
esac
# […]
eatmydata debootstrap --arch=arm64 --include=eatmydata --no-merged-usr \
--force-check-gpg --verbose --foreign buster "$mpt" \
http://deb.debian.org/debian sid
# script specified here as it’s normally what buster symlinks to,
# to achieve compatibility with more host distros
# we need this early
(
set -e
cd "$mpt"
for archive in var/cache/apt/archives/*eatmydata*.deb; do
dpkg-deb --fsys-tarfile "$archive" >a
tar -xkf a
done
rm -f a
) || die 'failure extracting eatmydata early'
# the user can delete this later, from the booted system (or apt will)
cp /usr/bin/qemu-aarch64-static "$mpt/usr/bin/" || die 'cp failed'
echo >&2 'I: second stage bootstrap (under emulation), slooow…'
mount -t tmpfs swap "$mpt/dev/shm" || die 'mount /dev/shm failed'
mount -t proc  proc "$mpt/proc" || die 'mount /proc failed'
mount -t tmpfs swap "$mpt/tmp" || die 'mount /tmp failed'
chroot "$mpt" /usr/bin/env -i LC_ALL=C.UTF-8 HOME=/root PATH="$safe_PATH" \
TERM="$TERM" /usr/bin/eatmydata /debootstrap/debootstrap --second-stage || \
die 'debootstrap (second stage) failed'


Of course, “success” here means it does not error out… ☻ (also, as it’s
run under emulation it’s very slooow anyway…)

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-235
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Dr. Stefan Barth, Kai Ebenrett, Boris Esser, Alexander Steeg

**

Mit der tarent Academy bieten wir auch Trainings und Schulungen in den
Bereichen Softwareentwicklung, Agiles Arbeiten und Zukunftstechnologien an.

Besuchen Sie uns auf www.tarent.de/academy. Wir freuen uns auf Ihren Kontakt.

**

uploading a new version of newt…

[Thorsten Glaser]

… in deep freeze, just shy of the release, is probably
not the best idea, is it? (Especially of things like
libraries which can affect other packages uploaded with
fixes that SHOULD migrate…)

You are not the first, but the package likely affects
everything in d-i due to cdebconf…

Just wondering from my apt-listchanges’ output,
//mirabilos
-- 
“It is inappropriate to require that a time represented as
 seconds since the Epoch precisely represent the number of
 seconds between the referenced time and the Epoch.”
-- IEEE Std 1003.1b-1993 (POSIX) Section B.2.2.2

Bug#930684: pbuilder: creation of build env fails when run inside Docker container

[Thorsten Glaser]

Tobias Junghans dixit:

>> But with {cow|p}builder --login --save-after-login you can
>> upgrade the base to buster inside pbuilder.
>
>It's not about the pbuilder environment itself but the Docker container used 
>for invoking pbuilder/debootstrap. Using a stretch-based Docker container and 
>debootstrapping Buster works fine.

Ah okay. Yes, that’s supported, a difference of only one version,
as long as the host has (at least) the stretch kernel.

As for the other things, the debootstrap people might look at it.

bye,
//mirabilos
-- 
15:41⎜ Somebody write a testsuite for helloworld :-)

Bug#930684: pbuilder: creation of build env fails when run inside Docker container

[Thorsten Glaser]

Tobias Junghans dixit:

>I tried to upgrade my Docker-based pbuilder containers from stretch to

Erm… why do you use chroots inside of chroots? That’s… tricky.

>mount: failed to read mtab: No such file or directory

This might be a container issue.

>mount: /proc: mount(2) system call failed: Too many levels of symbolic links.

Check if /proc outside of pbuilder but inside the container is right.

There also might be a /dev/shm vs. /run/shm issue. I recently had a
failure with these (one’s a mountpoint, the other a symbolic link to
it, either way works but it’s got to be consistent inside and out‐
side of the chroot) with schroot.

>the stretch-based pbuilder container and use it to build packages for buster.

Don’t.

But with {cow|p}builder --login --save-after-login you can
upgrade the base to buster inside pbuilder.

bye,
//mirabilos
-- 
FWIW, I'm quite impressed with mksh interactively. I thought it was much
*much* more bare bones. But it turns out it beats the living hell out of
ksh93 in that respect. I'd even consider it for my daily use if I hadn't
wasted half my life on my zsh setup. :-) -- Frank Terbeck in #!/bin/mksh

Bug#923675: Add related bug #916690 info

[Thorsten Glaser]

Daniel Lange dixit:

> Thorsten Glaser (CC) has produced a prototype early-rng-init-tools (cf.
> https://lists.debian.org/debian-devel/2019/02/msg00327.html) which could be
> extended to try reading entropy off the network when it doesn't have a
> carried-over seed (as in the Debian Installer case).

Sorry, this is deliberately out of scope.

My early-rng-init-tools is exactly for the use case of carrying a random
seed between boots and making it available to the system earlier (as a
stopgap until all bootloaders support passing it to the kernel before
the latter is even run) and *deliberately* does not touch the part where
entropy is collected.

FWIW, downloading entropy can be done (we have this in the MirBSD
installer) but has privacy concerns, so it should perhaps be optional.
This is easily done in d-i components, except for the little fact that
busybox wget in d-i lacks https support.

I’ve built myself a locally patched 'monolith' installer with extra
entropy over the network, but that’s site-dependent.

Also, please don’t assume everyone has amd64. The m68k people will,
among others, thank you ;-)

bye,
//mirabilos
--  
When he found out that the m68k port was in a pretty bad shape, he did
not, like many before him, shrug and move on; instead, he took it upon
himself to start compiling things, just so he could compile his shell.
How's that for dedication. -- Wouter, about my Debian/m68k revival

Bug#901332: d-i: Offer to shut down / power off instead of reboot at the end

[Thorsten Glaser]

Ben Hutchings dixit:

>virt-manager (or maybe the underlying libvirt) seems to handle this:
>when you create a VM and provide an installation image, it is
>automatically detached when the VM reboots.

Indeed it does. However, virt-manager also insists on
automatically starting the VM, and I usually also change
something in their settings _before_ the installation,
so I can’t use that feature. (It does offer changing some
settings before initially starting the VM, but that doesn’t
have the final configuration, and is *extremely* laggy up
to unusable if libvirt runs on a different system than
virt-manager.) But…

>> So, please, at the end, where it tells the reboot message, add
>> a third button that shuts down / powers off the system instead
>> of rebooting.
>
>Still, I do agree that this would be useful in general.

… that, yes. In general, this is probably useful.

Thanks,
//mirabilos
-- 
  "Using Lynx is like wearing a really good pair of shades: cuts out
   the glare and harmful UV (ultra-vanity), and you feel so-o-o COOL."
 -- Henry Nelson, March 1999

Bug#901332: d-i: Offer to shut down / power off instead of reboot at the end

[Thorsten Glaser]

Package: debian-installer
Severity: normal

d-i always insists on rebooting at the end. This is made worse
by it telling the user to change boot media, which, for virtual
machines, means reconfiguring the VM.

I *generally* wish to reconfigure the VM after first install
*anyway*.

So, please, at the end, where it tells the reboot message, add
a third button that shuts down / powers off the system instead
of rebooting.

Thanks in advance!

-- System Information:
Debian Release: buster/sid
  APT prefers unreleased
  APT policy: (500, 'unreleased'), (500, 'buildd-unstable'), (500, 'unstable'), 
(100, 'experimental')
Architecture: x32 (x86_64)
Foreign Architectures: i386, amd64

Kernel: Linux 4.15.0-1-amd64 (SMP w/8 CPU cores)
Locale: LANG=C, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8), LANGUAGE=C (charmap=UTF-8)
Shell: /bin/sh linked to /bin/lksh
Init: sysvinit (via /sbin/init)

Bug#860545: debootstrap fails to locate xzcat if the default shell is set to posh

[Thorsten Glaser]

On Tue, 18 Apr 2017, Olliver Schinagl wrote:

> has its shebang set to /bin/sh, it would work with any posix shell, including
> posh.

Do note that posh is *not* a POSIX shell, nor suitable for /bin/sh on
Debian systems at all. It contains tons of bugs inherited from pdksh
which I had to fix in mksh, so I feel qualified to say.

bye,
//mirabilos (current hat: mksh-in-Debian maintainer)
-- 
«MyISAM tables -will- get corrupted eventually. This is a fact of life. »
“mysql is about as much database as ms access” – “MSSQL at least descends
from a database” “it's a rebranded SyBase” “MySQL however was born from a
flatfile and went downhill from there” – “at least jetDB doesn’t claim to
be a database”  ‣‣‣ Please, http://deb.li/mysql and MariaDB, finally die!

Bug#789475: udhcpc: valid rfc1123 hostname recognized as "bad"

[Thorsten Glaser]

Source: busybox
Version: 1:1.22.0-19
Followup-For: Bug #789475
Control: tag 789475 = patch confirmed fixed-upstream d-i

Hi,

I’ve stumbled upon this bug by means of d-i using “bad” as hostname
right now as well and tracked it down to the offending piece of code
myself.

When reading the bugreport messages, I found out that the patch linked –
https://git.busybox.net/busybox/patch/?id=c29021e2a594fb29471c8c7e61ab8f45296622ba
– is indeed correct and will fix this issue.

-- System Information:
Debian Release: 9.0
  APT prefers unreleased
  APT policy: (500, 'unreleased'), (500, 'buildd-unstable'), (500, 'unstable')
Architecture: x32 (x86_64)
Foreign Architectures: i386, amd64

Kernel: Linux 4.9.0-1-amd64 (SMP w/4 CPU cores)
Locale: LANG=C, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/lksh
Init: sysvinit (via /sbin/init)

Bug#851892: To tight permissions on /dev/ptmx

[Thorsten Glaser]

James Clarke dixit:

>#817236 strikes again; the thread there has various workarounds.

Why is this still not fixed in debootstrap, anyway?

Thanks,
//mirabilos
-- 
> Wish I had pine to hand :-( I'll give lynx a try, thanks.

Michael Schmitz on nntp://news.gmane.org/gmane.linux.debian.ports.68k
a.k.a. {news.gmane.org/nntp}#news.gmane.linux.debian.ports.68k in pine

incorrect permissions on /dev/ptmx in newly created chroots break several packages’ builds (was Re: pbuilder: incorrect permissions on /dev/ptmx breaks openpty(); schroot: no access to pseudo-terminal

[Thorsten Glaser]

Ansgar Burchardt dixit:

>That looks like the same issue as I reported in sbuild in #817236 (which

It looks so, yes (Cc’ing now).

>severity maybe should be raised if this is an issue for more packages):

It certainly is! For example, a couple of months ago, cbmuser could
not sponsor my mksh upload because he could not get it to build, while
XTaran could do so successfully, and it took quite a while for cbmuser
to figure out the difference, which led to this explanatory commit:

https://anonscm.debian.org/cgit/collab-maint/mksh.git/diff/?id=aebc8e59054052a35e593718b99454ea94aead12=810e5b3928ea17a9d332475aadd19f6f8e14d245

See also the thread starting at:
http://lists.alioth.debian.org/pipermail/debian-ports-devel/Week-of-Mon-20160530/000185.html

>> … so I assume that something in debootstrap changed so that
>> there is now a symlink created instead of a proper device node.

>there needs to be a mount for /dev/pts to accommodate for the changes in
>debootstrap (and optionally a bind mount for /dev/ptmx if there are
>still old chroots around).

While I use only cowbuilder, I insist support for old chroots
is too important to lose. I’d r̲e̲a̲l̲l̲y̲ prefer if debootstrap
could create the device node again. AIUI it was not dropped
from an act of deliberation but merely as side effect of the
devices tarball cleanup. Other nodes (such as dev/tty) are
still created by it, so I don’t see why dev/ptmx needs to be
an exception if there are good reasons for it to not be.

Thanks,
//mirabilos
-- 
 cool ein Ada Lovelace Google-Doodle. aber zum 197. Geburtstag? Hätten
die nicht noch 3 Jahre warten können?  bis dahin gibts google nicht
mehr  ja, könnte man meinen. wahrscheinlich ist der angekündigte welt-
untergang aus dem maya-kalender die globale abschaltung von google ☺ und darum
müssen die die doodles vorher noch raushauen

Re: Bug#841935: pbuilder: incorrect permissions on /dev/ptmx breaks openpty()

[Thorsten Glaser]

@debootstrap maintainers, look near the end for a tl;dr

James Clarke dixit:

>> # ls -l /dev/ptmx
>> lrwxrwxrwx 1 root root 8 Oct  4 06:43 /dev/ptmx -> pts/ptmx
>> # ls -l /dev/pts/ptmx
>> c- 1 root root 5, 2 Oct 24 14:46 /dev/pts/ptmx

>> For comparison this is from my regular system:
>> $ ls -l /dev/ptmx /dev/pts/ptmx
>> crw-rw-rw- 1 root tty  5, 2 okt 24 17:03 /dev/ptmx
>> c- 1 root root 5, 2 okt 11 12:35 /dev/pts/ptmx

>for std{in,out,err}. I tried many months ago and couldn’t get it to work,
>but would love to be proved wrong.

So, the real problem is the symlink, I guess? This has also
been annoying me for mksh, whose testsuite uses script(1)
since forever (so I’m happy util-linux now uses it too).

I see that in all “old” and “old and upgraded” chroots I
have (sid started during lenny times), I have this setup…

ls: cannot access 'base.cow-sid-i386/dev/pts/ptmx': No such file or directory
crw-rw-rw- 1 root tty 5, 2 Okt 23 00:40 base.cow-sid-i386/dev/ptmx

… whereas all “new” chroots look like…

ls: cannot access 'base.cow-xenial-amd64/dev/pts/ptmx': No such file or 
directory
lrwxrwxrwx 1 root root 8 Mai 10 14:10 base.cow-xenial-amd64/dev/ptmx -> pts/ptmx

… so I assume that something in debootstrap changed so that
there is now a symlink created instead of a proper device node.

Maybe the debootstrap maintainers can comment on this?

Thanks,
//mirabilos
-- 
18:47⎜ well channels… you see, I see everything in the
same window anyway  18:48⎜ i know, you have some kind of
telnet with automatic pong 18:48⎜ haha, yes :D
18:49⎜ though that's more tinyirc – sirc is more comfy

Bug#839046: apt-listchanges: changelogs for tglase-nb.lan.tarent.de

[Thorsten Glaser]

On Mon, 24 Oct 2016, Marco d'Itri wrote:

> On Oct 24, Thorsten Glaser <t.gla...@tarent.de> wrote:
> 
> > This leaves the question: how can we install stretch
> > without usrmerge?

> By using the appropriate debootstrap switch.

In d-i?

> But the real question would be: why would you want to do that?

Umm, perhaps because I prefer my system to behave in a way
consistent with previous and other systems, and because of
personal preference, and because you kinda indicated that
you would *not* force Debian to require a merged /usr and
I kinda wish to cash in that promise.

Thanks,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-235
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Dr. Stefan Barth, Kai Ebenrett, Boris Esser, Alexander Steeg

Bug#839046: apt-listchanges: changelogs for tglase-nb.lan.tarent.de

[Thorsten Glaser]

On Mon, 24 Oct 2016, root wrote:

> debootstrap (1.0.85) unstable; urgency=medium

>   * Enable merged-/usr by default. (Closes: #839046)

This leaves the question: how can we install stretch
without usrmerge?

Thanks,
//mirabilos
-- 
Yes, I hate users and I want them to suffer.
-- Marco d'Itri on gmane.linux.debian.devel.general

Bug#717298: debian-installer: After installing xfce environment, don't see xfce

[Thorsten Glaser]

Christian PERRIER dixit:

>Well, desktop-chooser is a package that has never been released, so I
>doubt the problem comes from it.

Oh, okay. I only found it because I was searching for which desktop=
command line options are valid for d-i (to tell a coworker), and since
there is no documentation for it I’ve looked into the sources which I
had lying around on my workstation (due to me building a monolith image
for myself locally) already anyway.

>Anyway, your patch is still valid (so, thank you) and I'll apply it to
>desktop-chooser's git in case someone does something with it in the
>future...

OK.

bye,
//mirabilos
-- 
  "Using Lynx is like wearing a really good pair of shades: cuts out
   the glare and harmful UV (ultra-vanity), and you feel so-o-o COOL."
 -- Henry Nelson, March 1999

Bug#717298: debian-installer: After installing xfce environment, don't see xfce

[Thorsten Glaser]

Package: debian-installer
Followup-For: Bug #717298

This is probably because of this…

Choices-C: kde, gnome, lxde, xfcd, sugar
 

… in debian-installer/packages/desktop-chooser/debian/desktop-chooser.templates.

Quick fix attached.

-- System Information:
Debian Release: stretch/sid
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 4.6.0-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=C, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/lksh
Init: sysvinit (via /sbin/init)
>From 049ec0929973c5a8f570496bb8062b58f75f00ff Mon Sep 17 00:00:00 2001
From: mirabilos 
Date: Wed, 10 Aug 2016 16:45:14 +0200
Subject: [PATCH] =?UTF-8?q?fix=20typo:=20xfcd=20=E2=86=92=20xfce?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: mirabilos 
---
 debian/desktop-chooser.templates | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/debian/desktop-chooser.templates b/debian/desktop-chooser.templates
index 62243bf..826a20b 100644
--- a/debian/desktop-chooser.templates
+++ b/debian/desktop-chooser.templates
@@ -13,7 +13,7 @@ Description: for internal use only
 Template: desktop-chooser/suite
 Type: multiselect
 __Choices: The K Desktop Environment, GNU Object Model Environment, Lightweight X11 Desktop Environment, Xfce Desktop Environment, Sugar Learning Platform
-Choices-C: kde, gnome, lxde, xfcd, sugar
+Choices-C: kde, gnome, lxde, xfce, sugar
 Default: gnome
 _Description: Which desktop environments should be installed?
  There are several alternative Desktop environments available in
-- 
2.8.1

Re: Bug#794468: general: no watchdog support in installer kernel

[Thorsten Glaser]

Ben Hutchings dixit:

Which watchdog device is this, that is enabled by the BIOS?  Normally I
expect them to be enabled only by the kernel driver.

It’s a 19″ server from Supermicro; Dominik would’ve to provide details.

install time since the watchdog devices are not enabled by the system
firmware.

Oh well… apparently, the firmware setup screens don’t signal the
watchdog either, so you can’t use that one for five minutes while
the watchdog is enabled. This all points to buggy firmware. Again,
details would have to come from Nik.

bye,
//mirabilos
-- 
Thorsten Glaser
Teckids e.V. – Erkunden, Entdecken, Erfinden.
https://www.teckids.org/


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/pine.bsm.4.64l.1508091336430.29...@herc.mirbsd.org

Re: Debian Installer Jessie RC 1 release

[Thorsten Glaser]

Cyril Brulebois dixit:

 * Due to a change on the linux kernel side, the --- separator is
   now used instead of the historical -- one to separate kernel
   parameters from userland parameters. This makes it possible for
   user-params to do its job, and copy e.g. console=... settings
   where they're expected in the installed system.

Where can I find out what -- and --- on the kernel command line
used to do and do now? Within the circles of people I know, there
has been some confusion about -- recently.

Thanks,
//mirabilos
-- 
This space for rent.


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/pine.bsm.4.64l.1501261007500.32...@herc.mirbsd.org

Bug#768876: unblock: busybox/1:1.22.0-14

[Thorsten Glaser]

On Thu, 27 Nov 2014, Michael Tokarev wrote:

 (The Built-Using field generation is a bit fun here: I asked on IRC
 how people identify which libc is in use, and got various somewhat-
 incpmplete replies (the prob is that on different arches, libc package
 is named differently).  So I invented my own way for busybox, because

You didn’t ask in the #!/bin/mksh channel on IRC (= Freenode) ;-)

So let me add mine:

‣ intimate knowledge of the build system required, so you know
  what precidely is pulled in (reading shlibs:Depends from the
  build of the shared version is almost certainly wrong)

In the mksh case, I have a switch between different libcs to use.
For dietlibc and klibc, the cases are clear:

• libc_pkgname='libklibc-dev linux-libc-dev'

• libc_pkgname=dietlibc-dev

For (e)glibc, or rather, “the default libc”, a little trick:

• x=$(dpkg -S $(readlink -f $($CC -print-file-name=libc.a)))
  libc_pkgname=${x%%: *}

In *all* cases, you also need

• x=$(dpkg -S $(readlink -f $($CC -print-file-name=libgcc.a)))
  libgcc_pkgname=${x%%: *}

Now you need to get from that to the corresponding source packages:

for x in $libc_pkgname $libgcc_pkgname; do
dpkg-query -Wf '${source:Package} (= ${source:Version})\n' $x
done | sort -u | {
srcpkgnames=
while IFS= read -r x; do
test -n $x || continue
test x$x = x (= )  continue
echo Built Using: $x
test -z $srcpkgnames || srcpkgnames=$srcpkgnames, 
srcpkgnames=$srcpkgnames$x
done
echo mksh:B-U=$srcpkgnames builddir/substvars
}

bye,
//mirabilos
-- 
«MyISAM tables -will- get corrupted eventually. This is a fact of life. »
“mysql is about as much database as ms access” – “MSSQL at least descends
from a database” “it's a rebranded SyBase” “MySQL however was born from a
flatfile and went downhill from there” – “at least jetDB doesn’t claim to
be a database”  ‣‣‣ Please, http://deb.li/mysql and MariaDB, finally die!


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/alpine.deb.2.11.1411281305150.10...@tglase.lan.tarent.de

Bug#768876: unblock: busybox/1:1.22.0-14

[Thorsten Glaser]

On Fri, 28 Nov 2014, Michael Tokarev wrote:

  ‣ intimate knowledge of the build system required, so you know
what precidely is pulled in (reading shlibs:Depends from the
build of the shared version is almost certainly wrong)
 
 Why it is wrong?  To be this looks like the most accurate approach

The builds probably differ, especially when using autoconf.

  In *all* cases, you also need
  
  • x=$(dpkg -S $(readlink -f $($CC -print-file-name=libgcc.a)))
libgcc_pkgname=${x%%: *}
 
 Why?

Because some of these functions do end up in the resulting executable.

 This assumes it is gcc an it uses libgcc, so it wont work with,
 say, clang.

True. But either “clang -print-file-name=libgcc.a” will be empty,
or it will report libgcc.a which may or may not be pulled in. But
archive builds are not done using clang _yet_. I’ll revisit this
when that option exists, or so. It’s too fast moving a target at
the moment, anyway.

 Note also that if libgcc is used by shared build, it will
 also be listed in shilbdeps.

Fallacy. I know GCC interna better than I wish to. Most of
libgcc.a is always put into binaries; the shared libgcc is
mostly the libgcc_eh.a equivalent, and -static-libgcc is,
normally, the default for C language builds. (Also, it’s a
good way to keep the GCC version used to build it in the
archive.)

bye,
//mirabilos
-- 
15:41⎜Lo-lan-do:#fusionforge Somebody write a testsuite for helloworld :-)


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/alpine.deb.2.11.1411281602190.10...@tglase.lan.tarent.de

Bug#768876: unblock: busybox/1:1.22.0-14

[Thorsten Glaser]

On Fri, 28 Nov 2014, Michael Tokarev wrote:

 Um.  Maybe we should assume exact versions of software running in
 buildds too?

No, only things that end up in the binaries.

 BTW, how about somethig like gcc -v (I'm not sure it is the right
 option actually) which shows all libs it actually used for linking -

Yes, except that is not parsable, and varies by compiler.

 run it once, find out all actual libs and go from there, translating
 the list to debian package names.  I think _that_ will be the only
 real thing.

I agree. Maybe -Wl,-v or -Wl,-t instead. We always use binutils…
but these options also have their shortcomings.

 Besides, this process should be automated with some helper, something
 like dh_built-using, I dunno.  Or else, due to the fact that it is

No, this requires intimate knowledge of the build system in use;
autoconf will break during the configure phase if you use some
compiler flags (like -Werror) for example, so you have to inject
that flag somewhere.

Also, not everything is always static, and if you build multiple
binary packages, you need to track what is what anyway.

I think that, for B-U, we can’t find a generic solution.

 Oh well.  Do you think I should reopen this bugreport?

I think you probably only should add the libgcc.a provider
and cross fingers for now. B-U are not unimportant, but a
real PITA anyway right now.

bye,
//mirabilos
-- 
[16:04:33] bkix: veni vidi violini
[16:04:45] bkix: ich kam, sah und vergeigte...


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/alpine.deb.2.11.1411281629500.10...@tglase.lan.tarent.de

Re: Bug#762399: console-setup: WARNING: Unknown X keysym permille

[Thorsten Glaser]

Cyril Brulebois dixit:

'd console-setup' says binary diffs and:
 151 files changed, 93 insertions(+), 9257 deletions(-)

The person doing the upload did not do so cleanly.
There is an extra subdirectory fontconfig/ in the
.tgz recently, which has 940'632 bytes in 96 files
in 1.115 (I noticed because I regularily rebuild
console-setup locally, with local patches; my own
builds do not have that directory).

bye,
//mirabilos
-- 
How can you ban language, words? How're words offensive? And why should I
have to tolerate YOUR interpretation? I'm the one using the word. ASK me how
I'm using it, don't TELL me. And if you don't like the way I'm using it, so
what? It's my right. It's my freedom of expression. Without that, we're
nothing but slaves.-- Johnny Rotten


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/pine.bsm.4.64l.1411251015390.28...@herc.mirbsd.org

Bug#762399: console-setup: WARNING: Unknown X keysym permille

[Thorsten Glaser]

On Tue, 25 Nov 2014, Cyril Brulebois wrote:

 Meh, what are debian-bugs-dist@ and debian-boot@ doing in Cc?!

I just hit “reply”, “reply to mails instead of follow up to usenet”,
“reply to all”…

 I'm therefore tempted to upload 1.116 which would match 1.115 without
 the cruft, also known as 1.114 plus the bugfix minus the cruft.

Even better ☺

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-235
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Dr. Stefan Barth, Kai Ebenrett, Boris Esser, Alexander Steeg


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/alpine.deb.2.11.1411251752300.19...@tglase.lan.tarent.de

Re: Old-timer installer, task-sysvinit?

[Thorsten Glaser]

Jonas Smedegaard dixit:

 3. The installer CAN be used to install sysvinit via preseeding (and
 admins should know how to use preseeding to install systems...)
[…]
script late in the install process to replace init-related packages 
already installed by debian-installer.

Indeed. You can also just Alt-F2 and run
in-target apt-get --purge -y install sysvinit-core
yourself; the --purge additionally removes the configs of
removed packages, but dependencies of them are *still* there
*and* not eligible for “apt-get --purge autoremove” because
installed by debootstrap.

(I tend to install dselect to clean up junk d-i insists on,
like dictionaries, straight after the installation anyway,
so this is almost not more work for me, but…)

Plus: it installs systemd first, then switches to sysvinit.
Do we have a promise that that will keep being supported?
And it feels a bit backward…

Please Cc me, I am not subscribed here.

Same.

bye,
//mirabilos
-- 
diogenese Beware of ritual lest you forget the meaning behind it.
igli yeah but it means if you really care about something, don't
ritualise it, or you will lose it. don't fetishise it, don't
obsess. or you'll forget why you love it in the first place.


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/pine.bsm.4.64l.1411231217000.24...@herc.mirbsd.org

Re: Old-timer installer, task-sysvinit?

[Thorsten Glaser]

Cyril Brulebois dixit:

bad, because we're not going to change debootstrap, especially not at
this late stage of the release cycle, to behave differently depending
on the target distribution. The current script covers all suites from

This is not necessary. Well, not at first.

The changes to debootstrap I’d “love” to have are:

① add a new --variant called sysvinit, which excludes all the
  systemd bits and explicitly adds sysvinit-core; this variant
  will only be usable with jessie and up, but as it must be
  invoked explicitly will not do anything to older releases.

  This is the change we’d need now.

② change the scripts for --variant minbase, buildd, fakechroot,
  scratchbox to exclude systemd, sysvinit (and, for good measure
  on certain downstreams) upstart. This will not touch either
  --variant sysvinit or base (the default).

  This is a change that could be put into experimental, as it
  is not necessary for either the release or d-i. Then upload
  that to sid after the unfreeze.

③ To be fair, add --variant=upstart based on the sysvinit variant.
  And openrc. But this all will end up in experimental, too.

I’d actually be happy to try and hack these myself and submit
them for inclusion. Is the --exclude not working bug fixed by
now btw?

bye,
//mirabilos
-- 
21:12⎜Vutral sogar bei opensolaris haben die von der community so
ziemlich jeden mist eingebaut │ man sollte unices nich so machen das
desktopuser zuviel intresse kriegen │ das macht die code base kaputt
21:13⎜Vutral:#MirBSD linux war früher auch mal besser :D


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/pine.bsm.4.64l.1411231221540.24...@herc.mirbsd.org

Re: tasksel desktop preseed issues

[Thorsten Glaser]

Steven Chamberlain dixit:

* standard task still seems to bring in a huge amount of dependencies
via gnupg2  gnupg-agent  pinentry-gtk2  xserver-*

Break this by installing pinentry-curses before anything that
brings in gnupg-agent (same apt-get command line works).


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/pine.bsm.4.64l.1411232056250.30...@herc.mirbsd.org

filesystem labels

[Thorsten Glaser]

Hi,

just wondering: why can I tell d-i (manual partitioner)
the filesystem label for an ext4 filesystem, but not for
a swap partition?

mkswap has a -L option, too.

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-235
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Dr. Stefan Barth, Kai Ebenrett, Boris Esser, Alexander Steeg


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
https://lists.debian.org/alpine.deb.2.11.1411191503490.17...@tglase.lan.tarent.de

Bug#705454: mdadm: --examine --scan generates wrong #spares

[Thorsten Glaser]

reassign 705454 mdadm
found 705454 3.2.5-5
retitle 705454 mdadm: --examine --scan generates wrong #spares
thanks

I can reproduce this in a live sid system by creating
the md arrays then running this command:

root@tglase:/ # mdadm --examine --scan
ARRAY /dev/md0 UUID=b4a6dcca:639ab49e:3cbfd189:3a086020
   spares=1
ARRAY /dev/md1 UUID=b1518a63:e8fe3e9f:3cbfd189:3a086020
   spares=2
root@tglase:/ # cat /proc/mdstat
Personalities : [linear] [raid0] [raid1] [raid10] [raid6] [raid5] [raid4] 
[multipath]
md1 : active raid10 sdf2[4](S) sde2[3] sdd2[2] sdc2[1] sda2[0]
  975721472 blocks 512K chunks 2 near-copies [4/4] []

md0 : active raid1 sdf1[4](S) sde1[3] sdd1[2] sdc1[1] sda1[0]
  524224 blocks [4/4] []

unused devices: none

I assume that between the mds an internal counter
is not changed, or something.

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-314
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Boris Esser, Sebastian Mancke


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/alpine.deb.2.02.1304301539520.12...@tglase.lan.tarent.de

Bug#706112: debian-installer: Wheezy installer always install bootloader in /dev/sda

[Thorsten Glaser]

Hi,

just another how-to-reproduce: run d-i on a system with
/boot on RAID 1 and / on LVM on RAID 10 on five discs.

d-i (priority=low) asks whether to install grub2 to the
MBR, I say yes, and it installs to /dev/sda, so I’ll have
to, later, in the installed system, reconfigure it to tell
it to install to all discs.

Why isn’t the normal postinst of the grub-pc run in-target
with debconf pass-through instead? That one WFM…

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-314
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Boris Esser, Sebastian Mancke


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/alpine.deb.2.02.1304261325190.9...@tglase.lan.tarent.de

Bug#705454: d-i puts wrong amount of spares into target mdadm.conf

[Thorsten Glaser]

Package: debian-installer

Hi,

when using d-i to install Debian on a system with RAID,
which we normally preconfigure on the second console in
the shell with a few commands, mdadm.conf is wrong.

We use something like this:

mdadm --create --level=1 -n 4 --spares=1 /dev/md0 /dev/sd{a,b,c,d,e}1
mdadm --create --level=10 -n 4 --spares=1 /dev/md1 /dev/sd{a,b,c,d,e}2

Then put LVM on md1 and use d-i to put /boot on md0
and the rest of the system on LVs.

The mdadm.conf d-i generates for us in the target,
with these settings, have “spares=1” for md0 and
“spares=2” for md1 for some unknown reason, which
makes the mdadm cronjob (rightfully) warn about
missing spares. Changing it to “spares=1” for md1
too using a text editor makes it succeed.

I think the mdadm.conf here is generated by d-i
somehow, as it’s put into the target, but feel
free to reassign this around.

This happens both with d-i from April 2012 and from
April 2013 (though I built the images I used myself
on a sid system from sid sources), so I thought I’d
better report this, in case there’s some mishandling
of systems with more than one md device (TTBOMK grub
doesn’t yet boot from RAID-10, does it?).

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-314
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Boris Esser, Sebastian Mancke


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/alpine.deb.2.02.1304151048100.32...@tglase.lan.tarent.de

Bug#705457: d-i uses DHCP domain for system hostname

[Thorsten Glaser]

Package: debian-installer
Severity: minor

Hi,

when installing a system foo-demo in the datacentre DMZ,
the DHCP server gave back the following information to d-i:

 hostname foo-demo.tarent.de
 dhcp domain tarentex14.tarent.de

Now, d-i gained a new system hostname prompt since Apr 2012,
in that it now asks, separately, for hostname and domain part
of the hostname; the hostname part is correctly pre-shown as
“foo-demo” but the dhcp domain, instead of the domain part of
the hostname (as delivered by DHCP, and as pointed to in the
reverse DNS entry for the Legacy IP in use), for the domain
part of the system hostname.

I don’t want the system to be named foo-demo.tarentex14.tarent.de
though, and I suspect this, i.e. not using the hostname provided
by DHCP or the reverse DNS entry, will surprise many people.

How about only using the DHCP domain as fallback, i.e. when the
hostname does not have a dot and, possibly, no sensible reverse
DNS entry exists?

This is netcfg 1.108, network-preseed 1.58, wide-dhcpv6-client-udeb
20080615-1.1 (though IP was not used here, only Legacy IP).

bye,
//mirabilos
-- 
tarent solutions GmbH
Rochusstraße 2-4, D-53123 Bonn • http://www.tarent.de/
Tel: +49 228 54881-393 • Fax: +49 228 54881-314
HRB 5168 (AG Bonn) • USt-ID (VAT): DE122264941
Geschäftsführer: Boris Esser, Sebastian Mancke


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/alpine.deb.2.02.1304151102270.32...@tglase.lan.tarent.de

Re: Bug#704744: pbuilder: umounts /{dev,run}/shm of the *host* system

[Thorsten Glaser]

Stéphane Glondu dixit:

Why not just do nothing if /dev/shm is a symlink?

A $chroot/run/shm should probably be umounted.

Are there cases where umount_on_exit is called on a symlink that should
be followed? If not, I would just kill the problem directly there, as in

I was thinking about this a bit, and tonight wondered whether/why
things like readlink/realpath (the shell utilities, not the syscalls
or libc functions) do not offer the functionality of resolving symlinks
(should they encounter any) relative to some (emulated) chroot path.

That would probably help…

Cyril Brulebois dixit:

Next time, can you please put the right people in the loop?!

Cc-ing:
 debian-bugs-d...@lists.debian.org

I did a reply-to-all on the mail.

is just plain stupid. Maintainers of the package you're reassigning to
don't get your control mail. Way to communicate!

I did not know that. I think this is a debbugs bug; it’s inconsistent
to require of the bug submitter to manually look up maintainers (for
example, I wasn’t aware debootstrap has anything to do with booting
Debian…) when reassigning, when one doesn’t have to do so normally.

Not sure I agree that's an RC bug in debootstrap. I'd rather call it a

If debootstrap tries to umount a symlink that points to outside
the chroot, I’d call it an issue with debootstrap, independent of…

wishlist in debootstrap to support the new thing pbuilder imposes, and
an RC bug in pbuilder not to depend on a debootstrap version
implementing said improved behaviour.

… this one.

Yes, the maintainers of debootstrap (and possibly cdebootstrap),
pbuilder and cowbuilder should probably talk to each other.
Ideally.

But right now, we have a bug that breaks unrelated software,
by umounting my /run/shm every time I create a chroot, and
thus deleting every piece of data that was put there. In this
particular instance, I’m the user who doesn’t really care
about where the bug is, or whose fault it is – this is one
of the major problems with a system based on packages of
separate maintainers…

bye,
//mirabilos
-- 
Sorry,  I’m annoyed today and you came by as an Arch user. These are the
perfect victims for any crime against humanity, like  systemd,  feminism
or social democracy.
-- Christoph Lohmann on d...@suckless.org


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/pine.bsm.4.64l.1304081205520.22...@herc.mirbsd.org

Re: Bug#704744: pbuilder: umounts /{dev,run}/shm of the *host* system

[Thorsten Glaser]

Cyril Brulebois dixit:

Certainly not an RC one. Faulty setups can lead to suboptimal
behaviours. That's one such case. Lowering severity accordingly (even
if as I said, important is probably too high on the debootstrap side).

Excuse me? Running debootstrap umounts /run/shm and you
call this a “faulty setup”?

Unbelievable.

bye,
//mirabilos
-- 
Support mksh as /bin/sh and RoQA dash NOW!
‣ src:bash (257 (276) bugs: 0 RC, 180 (194) IN, 77 (82) MW, 0 (0) FP)
‣ src:dash (84 (99) bugs: 3 RC, 39 (44) IN, 42 (52) MW, 0 FP)
‣ src:mksh (1 bug: 0 RC, 0 IN, 1 MW, 0 FP, 1 gift)
http://qa.debian.org/data/bts/graphs/d/dash.png is pretty red, innit?


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/pine.bsm.4.64l.1304081539360.24...@herc.mirbsd.org

Bug#635370: busybox: integer overflow in expression on big endian

[Thorsten Glaser]

Source: busybox
Version: 1.18.5-1
Priority: wishlist

include/archive.h:17: warning: integer overflow in expression

Only shown on big endian architectures. Probably best to replace
 17 XZ_MAGIC1a  = ((0xfd * 256 + '7') * 256 + 'z') * 256 + 'X',
with XZ_MAGIC1a = 0xFD377A58UL directly instead.

The little endian case is probably fine because 0xFD has bit7 on, 'X' off.



-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/pine.bsm.4.64l.1107251513380.6...@herc.mirbsd.org

Re: debian-installer: kernel lagging for sh4, m68k, alpha

[Thorsten Glaser]

Hector Oron dixit:

If you have kernel flavours, then you probably want to have
subarch support for m68k (similar to armel).

I don’t know. Do we have kernel flavours? The -atari, -amiga,
etc. are like -i486 and -i686-pae kernels.

If you need any update, please send a patch with updated
information, I'll be happy to apply any changes to help out
m68k porting effort.

Patch against what? I’ve got about no experience with d-i…
we currently have stock 2.6.39-1 from Debian unstable working
just fine, if that helps, but no *-di (source or binary) pak-
kage… other d-i deps are available.

bye,
//mirabilos
-- 
22:59⎜Vutral glaub ich termkit is kompliziert | glabe nicht das man
damit schneller arbeitet | reizüberflutung │ wie windows │ alles evil
zuviel bilder │ wie ein spiel | 23:00⎜Vutral die meisten raffen auch
nicht mehr von windows | 23:01⎜Vutral bilderbücher sind ja auch nich
wirklich verbreitet als erwachsenen literatur   ‣ who needs GUIs thus?


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/pine.bsm.4.64l.1106071840580.20...@herc.mirbsd.org

Re: debian-installer: kernel lagging for sh4, m68k, alpha

[Thorsten Glaser]

Hector Oron dixit:

(CC me on replies as I might not be subscribed to this list)

Done.

  On SH4, m68k and alpha architecture, as those are not Debian
Official candidates for release, debian-installer is lagging kernel
behind.

d-i/installer/doc/devel/kerntabl:
arch   kernel udebs build/config
[…]
m68k   2.6.26 (-7)  2.6.26-1 # unoficial

Ben Hutchings has uploaded 2.6.38-5 (ABI 2) today, which
includes all “needed” patches for m68k support. This means
that I will probably upload debs to unstable end of the
week, it takes a while to compile six kernels.

The packages are linux-image-2.6.38-2-{amiga,atari,…}
versioned as *_2.6.38-5_m68k.deb (I don’t know which
of these two versions you need). As I’ve cross-built
and tested 2.6.38-4 plus the patches that got applied
earlier, I’m fairly certain they will work better than
anything we’ve ever had before, so I’d like to ask you
to update the information in that file.

If you need any udebs to be recompiled (I only keep track
of regular debs, but upload the udebs built anyway so a
lot of them should be pretty up to date by now), please
contact me (either directly or via debian-68k@) with the
names of the _source_ packages in Debian, and I’ll see
what I can do. (m68k currently still has no (automated)
buildd running, but we’re working on that too.)

bye,
//mirabilos
-- 
If Harry Potter gets a splitting headache in his scar
when he’s near Tom Riddle (aka Voldemort),
does Tom get pain in the arse when Harry is near him?
-- me, wondering why it’s not Jerry Potter………


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/pine.bsm.4.64l.1105081855100.28...@herc.mirbsd.org

Bug#588324: Please include cttyhack and setsid

[Thorsten Glaser]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA384

This is really useful and may be needed in a lot of situations:

+1 for cttyhack

This can help in some cases:

+1 for setsid

//mirabilos
- -- 
FWIW, I'm quite impressed with mksh interactively. I thought it was much
*much* more bare bones. But it turns out it beats the living hell out of
ksh93 in that respect. I'd even consider it for my daily use if I hadn't
wasted half my life on my zsh setup. :-) -- Frank Terbeck in #!/bin/mksh
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.7 (MirBSD)
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=16VV
-END PGP SIGNATURE-



-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: 
http://lists.debian.org/pine.bsm.4.64l.1104021044160.17...@herc.mirbsd.org

Re: debian-installer: kernel lagging for sh4, m68k, alpha

[Thorsten Glaser]

Hector Oron dixit:

m68k   2.6.26 (-7)  2.6.26-1 # unoficial

I’ve got working 2.6.32-26+m68k.2 which I uploaded to unreleased on
debian-ports.org yesterday, and 2.6.32-27+m68k.3 is almost built.
The kernels for m68k sort of need patches to build which, at the
current time, will probably not be added to the sid (2.6.32-*)
kernel to keep it stable. (Also, I’m building them with gcc-4.4
right now, because we got that into better shape than gcc-4.3.)

Can you use kernels from d-p.o unreleased for the installer?
Stephen said that d-i kernels were traditionally cross-built
manually anyway, IIRC.

Preliminary debdiff (from dpkg-buildpackage -S before building it)
at: 
https://www.freewrt.org/~tg/debs68k/notyet/linux-2.6/linux-2.6_2.6.32-27+m68k.3.debdiff
(will move to debs68k/dists/sid/main/PS-ok-unrel/linux-2.6/ once tested)

I’m currently at
│Note: Writing set_irq_chip.9
│Warn: AUTHOR sect.: no personblurb|contrib for Thomas Gleixner
│Note: AUTHOR sect.: see see http://docbook.sf.net/el/contrib
│Note: AUTHOR sect.: see see http://docbook.sf.net/el/personblurb
│Warn: AUTHOR sect.: no personblurb|contrib for Ingo Molnar
│Note: AUTHOR sect.: see see http://docbook.sf.net/el/contrib
│Note: AUTHOR sect.: see see http://docbook.sf.net/el/personblurb
in the compilation process (doing an arch+indep native build for
the first time, previously I built the indep portion on i386), so
the binary part is finished but not yet packaged or tested. The
difference to 2.6.32-26+m68k.2 is that RTC_GENERIC is enabled on
atari for ARAnyM, and it’s resynched with 2.6.32-27, so I expect
it to work, but you can never be sure… (if I don’t manage it until
tonight or maybe tomorrow morning, it will take a few days more due
to a dentist visit *shudders*).

Since the package is for d-p.o unreleased, it’s nothing that should
be applied as-is. Specifically, either the compiler switch to gcc-4.4
(will be uploaded today) should be made m68k specific, or the config
needs CONFIG_ARCH_WANT_FRAME_POINTERS and CONFIG_FRAME_POINTER to
work around a gcc-4.3 bug. The patches… fixing them should be applied
right now, I think; enabling them is questionable as some touch MI
stuff, such as msdosfs. They’re needed (well, a subset of them) for
ARAnyM machine support though (without, it boots but has no ethernet,
which is kind of bad) which is vital to development.

bye,
//mirabilos
-- 
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or only God invents algorithms, we merely copy them.
If you don't believe in God, just consider God as Nature if you won't deny
existence.  -- Coywolf Qi Hunt


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/pine.bsm.4.64l.1011071248500...@herc.mirbsd.org

Re: Bootable USB images

[Thorsten Glaser]

Ferenc Wagner dixit:

In case it's not widely known, let me quote isolinux.doc:

Starting in version 3.72, ISOLINUX supports a hybrid mode which can
be booted from either CD-ROM or from a device which BIOS considers a
hard disk or ZIP disk, e.g. a USB key or similar.

That’s one of the methods; Robert’s new GRUB change, as well as my
“manifold-boot” method which also works with GRUB2, are others. We
looked at them in Grml; the ISOHYBRID variant used to not work with
stock Debian packaging though that may have changed, mail m...@d.o
for details. It basically depends on whether you want SYSLINUX or
GRUB as loader (I think GRUB offers more flexibility which can be
desirable on d-i).

bye,
//mirabilos
-- 
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font.   -- Rob Pike in Notes on Programming in C


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: using d-i to install kfreebsd-i386 from usb

[Thorsten Glaser]

Luca Favatella dixit:

I'll probably test this solution (dd iso to usb) when possible.

Well, the ISO image has to be specially prepared for it.

At the *very* least, you’ll need to
• make sure there is GRUB2 *inside* the ISO on /boot/grub
• git clone git://git.grml.org/grml-live.git
• take the *same* version of GRUB2 on your local computer
• (
echo 4 63 | mksh grml-live/scripts/bootgrub.mksh -A
dd if=/dev/zero bs=512 count=3 2/dev/null
grub-mkimage biosdisk iso9660
  ) | dd of=«isoname» conv=notrunc

Side note: the output of “grub-mkimage biosdisk iso9660” must not
be larger than 30720 bytes when using this method. (By omitting
padding and possibly its first sector, we can probably get 32 KiB.)

The Grml ISOs (prepared with the grml-live script) already do this.

Source for bootgrub.mksh, in case someone wonders, is at
https://www.mirbsd.org/cvs.cgi/src/sys/arch/i386/stand/bootxx/

bye,
//mirabilos
-- 
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font.   -- Rob Pike in Notes on Programming in C


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: using d-i to install kfreebsd-i386 from usb

[Thorsten Glaser]

Luca Favatella dixit:

I was thinking there was no need to prepare a special iso.
Thanks for your clarification.

Well, no. ISO 9660 filesystem images are usually booted from CD-ROM or
similar media using the El Torito standard. However, when dd’d onto a
hard disc, CF/SD card, USB stick, etc. they are actually “superfloppies”.

The standard (I have Ecma 119 (= ISO 9660), SUSP, RRIP and El Torito here
as PDFs, they’re “freeware”) however specifies that the first 32 KiB of
an ISO 9660 filesystem image are empty and available for, for example,
bootloaders. The instructions I gave you will put the MirBSD first-stage
bootloader and GRUB 2 in there, it’s small enough that fits. Grml addi-
tionally pads the image and creates a partition table, so that you can
use the rest of the stick (AFTER dd’ing to it) as storage space, since
ISO 9660 filesystems are usually read-only. MirBSD uses ldbsd.com instead
of GRUB 2, which only differs by having a different load address, entry
point and stack segment/pointer, but also adds not only the partition
table but also a Sun disklabel and sparc bootloader in there (that one
was tricky, but works very well), as our CDs are usually bi-arch.

Hope that helps.

bye,
//mirabilos
-- 
«MyISAM tables -will- get corrupted eventually. This is a fact of life. »
“mysql is about as much database as ms access” – “MSSQL at least descends
from a database” “it's a rebranded SyBase” “MySQL however was born from a
flatfile and went downhill from there” – “at least jetDB doesn’t claim to
be a database”  -- Tonnerre, psychoschlumpf and myself in #nosec


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: using d-i to install kfreebsd-i386 from usb

[Thorsten Glaser]

Luca Favatella dixit:

Is there a way to use d-i to install kfreebsd-i386 from usb?

I don't see any technical reason why not. It probably depends
on the bootloader used.

MirBSD uses the manifold-boot method with MirBSD's loader.
Grml uses the manifold-boot method with GNU GRUB 2.
Others use ISOHYBRID with ISOLINUX.

With these techniques, you can dd(1) an ISO 9660 filesystem
image onto a USB stick and have it bootable immediately.

I'm pretty sure that the FreeBSD kernel is capable of handling
such a situation. It's just an oversized floppy with a filesy-
stem supported, although unusual, after all.

Note that GNU GRUB-legacy isn't capable of handling it, for
SYSLINUX you need a FAT16 filesystem (so you'd have to have
separate images for ISO and USB stick, which doubles the need
to produce, test and distribute), and other loaders also are
or may be problematic. I wouldn't know about the FreeBSD loa-
der, but from what I read, GRUB2 is already up to booting a
Debian GNU/kFreeBSD by itself.

I'll gladly provide assistance if desired.

bye,
//mirabilos
-- 
16:47⎜«mika:#grml» .oO(mira ist einfach gut)  23:22⎜«mikap:#grml»
mirabilos: und dein bootloader ist geil :)23:29⎜«mikap:#grml» und ich
finds saugeil dass ich ein bsd zum booten mit grml hab, das muss ich dann
gleich mal auf usb-stick installieren   -- Michael Prokop über MirOS bsd4grml


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

Luca Favatella dixit:

* a screenshot of the problem I get in the (monolithic) image built
  using makefs

Ok, I see now (forwarded to workplace):

I think you really should RTFM. There are quite some options:

 -f free-files
   Ensure that a minimum of free-files free files (inodes) exist in
   the image. An optional '%' suffix may be provided to indicate that
   free-files indicates a percentage of the calculated image size.

 -o fs-options
   avgfpdir  Expected number of files per directory.
   density   Bytes per inode.

cf. http://www.mirbsd.org/htman/i386/man8/makefs.htm

I’d say your best bet is to use -f without ‘%’ – you should know beforehand
though roughly how many new inodes (symlinks, devices (probably not, due to
devfs), temp files) you’ll need; they’re allocated in groups, so a guessti-
mate is okay.

//mirabilos
-- 
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font.   -- Rob Pike in Notes on Programming in C


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

On Wed, 5 Aug 2009, Luca Favatella wrote:

 It looks like there are no regressions in the produced monolithic and
 netboot images.

Cool!

Should I proceed to upload makefs as-is to Debian? Is there someone
here willing to sponsor? Otherwise, Mika Prokop from grml.org has
volunteered to do it if I need a sponsor. (I can only maintain it,
but NEW must go through a DD.)

bye,
//mirabilos
-- 
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font.   -- Rob Pike in Notes on Programming in C


-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

Luca Favatella dixit:

Tested. Please see

Great ☺

http://svn.debian.org/viewsvn/d-i/branches/d-i/kfreebsd/installer/build/Makefile?r1=59940r2=59943pathrev=59943

You forgot “diff_format=u” otherwise I see nothing…

Why didn’t you set minfree to 0 from the default of 5% as
it doesn’t make any sense in initrd?

bye,
//mirabilos
-- 
  Using Lynx is like wearing a really good pair of shades: cuts out
   the glare and harmful UV (ultra-vanity), and you feel so-o-o COOL.
 -- Henry Nelson, March 1999


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

Aurelien Jarno dixit:

I will have a look at the package and upload it if everything is fine.

Thanks.

I am a bit concerned by the name though. I understand it comes from the

NetBSD® world, actually. It handles 4.2FFS, UFS2 and ISO 9660 (plus
SUSP, RRIP, El Torito). Support for ext2fs is possible to be added.

But still, renaming an upstream utility………?

The package short description should make that clear, maybe.
We have:
Description: create a file system image from a directory tree
Maybe better would be:
Description: create a cd9660 or ffs file system image from a directory tree

bye,
//mirabilos
-- 
[...] if maybe ext3fs wasn't a better pick, or jfs, or maybe reiserfs, oh but
what about xfs, and if only i had waited until reiser4 was ready... in the be-
ginning, there was ffs, and in the middle, there was ffs, and at the end, there
was still ffs, and the sys admins knew it was good. :)  -- Ted Unangst über *fs


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

Luca Favatella dixit:

Can you please explain me better the difference between -f and
minfree (the man page is not clear). I read:

Sure. -f ensures, for makefs(8), that the filesystem has enough
free inodes, whereas minfree is a tunefs(8) parametre you might
know from ext2fs, which measures the amount of available space
which is SUBTRACTED from the total space shown by df(1) and re-
served for the superuser.

Clear now?

bye,
//mirabilos
-- 
Solange man keine schmutzigen Tricks macht, und ich meine *wirklich*
schmutzige Tricks, wie bei einer doppelt verketteten Liste beide
Pointer XORen und in nur einem Word speichern, funktioniert Boehm ganz
hervorragend.   -- Andreas Bogk über boehm-gc in d.a.s.r


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

Luca Favatella dixit:

* the patch against current kfreebsd d-i svn

Try something like this:

IMGSZ?= 20m
define mkfs.ufs1
  fs=$$(mktemp -d) ; \
  cp -a $(TREE)/* $${fs}/ ; \
  tmp=$$(mktemp -d) ; \
  makefs -t ffs -s ${IMGSZ} -o minfree=0 $${tmp}/ufs $${fs}/ ; \
  mv $${tmp}/ufs
endef
define mkfs.ufs2
  fs=$$(mktemp -d) ; \
  cp -a $(TREE)/* $${fs}/ ; \
  tmp=$$(mktemp -d) ; \
  makefs -t ffs -s ${IMGSZ} -o minfree=0,version=2 $${tmp}/ufs $${fs}/ ; \
  mv $${tmp}/ufs
endef
 
Afterwards adjust the IMGSZ to be just a little more than
what's actually needed.

Why do you copy ${TREE}/* to a temporary $$fs/ first, anyway?
You can run it directly on ${TREE} I'd say:

define mkfs.ufs1
  sh -c 'makefs -t ffs -s ${IMGSZ} -o minfree=0,version=1 $$0 ${TREE}'
endef

Again, untested.

//mirabilos
-- 
“It is inappropriate to require that a time represented as
 seconds since the Epoch precisely represent the number of
 seconds between the referenced time and the Epoch.”
-- IEEE Std 1003.1b-1993 (POSIX) Section B.2.2.2


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: test of makefs in kfreebsd d-i

[Thorsten Glaser]

Luca Favatella dixit:

I'm attaching
* a screenshot of the problem I get in the (monolithic) image built using 
makefs

Please attach screenshots in ASCII. I mostly read my eMails either
via ssh or on an 80486DLC laptop from 1993 with not enough RAM for
XFree86®, so I cannot view any images.

How can I keep some free inodes?

I don't quite know if that's the problem you meant, but:
http://www.NetBSD.org/cgi-bin/query-pr-single.pl?number=33925
The size of 4.2FFS images cannot currently be calculated, you
must use the -s option.

How can I explicitly select ufs1 (not ufs2) in makefs? (the man page
is not so clear)

makefs -t ffs target.img dir

4.2FFS is the default.

bye,
//mirabilos, who will answer more if he gets sent an ASCII screenshot
-- 
“It is inappropriate to require that a time represented as
 seconds since the Epoch precisely represent the number of
 seconds between the referenced time and the Epoch.”
-- IEEE Std 1003.1b-1993 (POSIX) Section B.2.2.2


--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#538171: ITP: makefs -- create a file system image from a directory tree

[Thorsten Glaser]

Package: wnpp
Severity: wishlist
Owner: Thorsten Glaser t...@mirbsd.de

* Package name: makefs
  Version : 20090724
  Upstream Author : The MirOS Project miros-disc...@mirbsd.org
* URL : http://www.mirbsd.org/cvs.cgi/src/usr.sbin/makefs/
* License : 4-clause BSD
  Programming Lang: C
  Description : create a file system image from a directory tree


NetBSD® makefs(8) creates a file system image from a directory tree
without the need for superuser privileges. The MirBSD version fixes
ECMA 119, SUSP and RRIP (Rock Ridge) compliance.

Supported target filesystem types are:
   cd9660   ISO 9660 (ECMA 119) compatible filesystem images, with
Rock Ridge, El Torito, and other features
   ffs  4.2FFS, the BSD Fast Filesystem, also known as UFS1;
support for UFS2 is currently untested

The images created can be of a fixed (predefined) size, given on the
command line, or sized automatically. Permission bits are taken from
the source directory tree but may be overridden using an mtree file.


This package is created to aid Luca Favatella in porting d-i to
Debian GNU/kFreeBSD. It also serves purpose as a genisoimage re-
placement, if people should then so wish. (Which is why the MirBSD
version was ported, as it conforms to ECMA 119, RRIP, SUSP and the
El Torito standard, which the NetBSD® version doesn’t.)

I fully intend to maintain this package in MirBSD (“upstream”) as
well as Debian (in my capacity as DM); I also try to coordinate
pushing all patches to the “up-upstream” (NetBSD®).



-- 
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org