Debian Bug report logs - #987735
allow deployments in non-empty directories

Display info messages

Message #5 received at submit@bugs.debian.org (full text, mbox, reply):

From: Antoine Beaupre <anarcat@debian.org>

To: Debian Bug Tracking System <submit@bugs.debian.org>

Subject: allow deployments in non-empty directories

Date: Wed, 28 Apr 2021 15:17:03 -0400

Package: mmdebstrap
Severity: wishlist

This fails:

# mkdir -p /mnt/var
# mount /dev/sdb /mnt/var
# mmdebstrap buster /mnt
I: automatically chosen mode: root
I: chroot architecture amd64 is equal to the host's architecture
E: /mnt is not empty

While I understand the idea (we don't want to destroy things i
guess?), it seems rather inconvenient if we want to split data across
different partitions which, I suspect, is still a valid use case for
various reasons.

It might be worth adding a commandline flag to allow such behavior, at
least...

-- System Information:
Debian Release: 10.9
  APT prefers stable-debug
  APT policy: (500, 'stable-debug'), (500, 'stable'), (1, 'experimental'), (1, 'unstable'), (1, 'testing')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 5.10.0-0.bpo.3-amd64 (SMP w/4 CPU cores)
Kernel taint flags: TAINT_CPU_OUT_OF_SPEC
Locale: LANG=fr_CA.UTF-8, LC_CTYPE=fr_CA.UTF-8 (charmap=UTF-8), LANGUAGE=fr_CA.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled

Versions of packages mmdebstrap depends on:
ii  apt       1.8.2.2
ii  perl      5.28.1-6+deb10u1
ii  perl-doc  5.28.1-6+deb10u1
ii  python3   3.7.3-1

Versions of packages mmdebstrap recommends:
ii  arch-test            0.15-2+deb10u1
ii  fakechroot           2.19-3.2
ii  fakeroot             1.23-1
ii  gpg                  2.2.12-1+deb10u1
ii  libdistro-info-perl  0.21
ii  mount                2.33.1-0.1
ii  uidmap               1:4.5-1.1

Versions of packages mmdebstrap suggests:
ii  apt [apt-transport-https]  1.8.2.2
pn  apt-transport-tor          <none>
ii  apt-utils                  1.8.2.2
ii  binfmt-support             2.2.0-2
ii  ca-certificates            20200601~deb10u2
ii  debootstrap                1.0.114
ii  distro-info-data           0.41+deb10u3
ii  dpkg-dev                   1.19.7
ii  perl-doc                   5.28.1-6+deb10u1
pn  proot                      <none>
ii  qemu-user                  1:3.1+dfsg-8+deb10u8
ii  qemu-user-static           1:3.1+dfsg-8+deb10u8
pn  squashfs-tools-ng          <none>

Message #8 received at 987735@bugs.debian.org (full text, mbox, reply):

From: Johannes Schauer Marin Rodrigues <josch@debian.org>

To: 987735@bugs.debian.org, Antoine Beaupre <anarcat@debian.org>

Subject: Re: Bug#987735: allow deployments in non-empty directories

Date: Wed, 28 Apr 2021 21:26:48 +0200

[Message part 1 (text/plain, inline)]

Hi,

Quoting Antoine Beaupre (2021-04-28 21:17:03)
> This fails:
> 
> # mkdir -p /mnt/var
> # mount /dev/sdb /mnt/var
> # mmdebstrap buster /mnt
> I: automatically chosen mode: root
> I: chroot architecture amd64 is equal to the host's architecture
> E: /mnt is not empty
> 
> While I understand the idea (we don't want to destroy things i
> guess?), it seems rather inconvenient if we want to split data across
> different partitions which, I suspect, is still a valid use case for
> various reasons.

yes, this is by design to avoid stuff like this from happening:

https://bugs.debian.org/833525

> It might be worth adding a commandline flag to allow such behavior, at
> least...

You are probably talking about --skip=check/empty

Alternatively, you can also run mmdebstrap like this:

mmdebstrap buster | tar -C /mnt -xf -

Thanks!

cheers, josch

[signature.asc (application/pgp-signature, inline)]

Message #13 received at 987735@bugs.debian.org (full text, mbox, reply):

From: Antoine Beaupré <anarcat@debian.org>

To: Johannes Schauer Marin Rodrigues <josch@debian.org>, 987735@bugs.debian.org

Subject: Re: Bug#987735: allow deployments in non-empty directories

Date: Wed, 28 Apr 2021 15:34:38 -0400

On 2021-04-28 21:26:48, Johannes Schauer Marin Rodrigues wrote:
> Hi,
>
> Quoting Antoine Beaupre (2021-04-28 21:17:03)
>> This fails:
>> 
>> # mkdir -p /mnt/var
>> # mount /dev/sdb /mnt/var
>> # mmdebstrap buster /mnt
>> I: automatically chosen mode: root
>> I: chroot architecture amd64 is equal to the host's architecture
>> E: /mnt is not empty
>> 
>> While I understand the idea (we don't want to destroy things i
>> guess?), it seems rather inconvenient if we want to split data across
>> different partitions which, I suspect, is still a valid use case for
>> various reasons.
>
> yes, this is by design to avoid stuff like this from happening:
>
> https://bugs.debian.org/833525
>
>> It might be worth adding a commandline flag to allow such behavior, at
>> least...
>
> You are probably talking about --skip=check/empty

Totally missed that. Was that introduced only in bullseye?

> Alternatively, you can also run mmdebstrap like this:
>
> mmdebstrap buster | tar -C /mnt -xf -

Makes sense, thanks!

a.

-- 
See the world as if for the first time; see it through the eyes of a
child, and you will suddenly find that you are free.
                        - Deepak Chopra

Message #16 received at 987735@bugs.debian.org (full text, mbox, reply):

From: Johannes Schauer Marin Rodrigues <josch@debian.org>

To: 987735@bugs.debian.org, Antoine Beaupré <anarcat@debian.org>

Subject: Re: Bug#987735: allow deployments in non-empty directories

Date: Wed, 28 Apr 2021 21:43:22 +0200

[Message part 1 (text/plain, inline)]

Quoting Antoine Beaupré (2021-04-28 21:34:38)
> >> It might be worth adding a commandline flag to allow such behavior, at
> >> least...
> >
> > You are probably talking about --skip=check/empty
> 
> Totally missed that.

The --skip option is semi-intentionally undocumented. The man page is already
very long and I don't want to clutter it with options that are only used in
very few corner cases. Maybe I will start a new document in /usr/share/doc or
something...

> Was that introduced only in bullseye?

The --skip option was introduced in version 0.7.0 and --skip=check/empty exists
since that version as well. Bullseye will ship with 0.7.5, so yes.

[signature.asc (application/pgp-signature, inline)]

Message #21 received at 987735@bugs.debian.org (full text, mbox, reply):

From: Antoine Beaupré <anarcat@debian.org>

To: Johannes Schauer Marin Rodrigues <josch@debian.org>, 987735@bugs.debian.org

Subject: Re: Bug#987735: allow deployments in non-empty directories

Date: Wed, 28 Apr 2021 16:09:04 -0400

Control: found -1 0.4.1-3
Control: fixed 0.7.0-1

On 2021-04-28 21:43:22, Johannes Schauer Marin Rodrigues wrote:
> Quoting Antoine Beaupré (2021-04-28 21:34:38)
>> >> It might be worth adding a commandline flag to allow such behavior, at
>> >> least...
>> >
>> > You are probably talking about --skip=check/empty
>> 
>> Totally missed that.
>
> The --skip option is semi-intentionally undocumented. The man page is already
> very long and I don't want to clutter it with options that are only used in
> very few corner cases. Maybe I will start a new document in /usr/share/doc or
> something...

I think it's fine if manpages are long, as long as they follow the
section conventions. Typically, OPTIONS is long, and then you jump to
EXAMPLES for common patterns or grep through the rest for specific
patterns.

It's not like you're supposed to read the entire thing, and particularly
not the OPTIONS part. :)

>> Was that introduced only in bullseye?
>
> The --skip option was introduced in version 0.7.0 and --skip=check/empty exists
> since that version as well. Bullseye will ship with 0.7.5, so yes.

I see, that's why i missed it. So I guess this is fixed.

A.

-- 
Information is not knowledge. Knowledge is not wisdom.
Wisdom is not truth. Truth is not beauty.
Beauty is not love. Love is not music.
Music is the best.      - Frank Zappa

Message #30 received at 987735-done@bugs.debian.org (full text, mbox, reply):

From: Johannes Schauer Marin Rodrigues <josch@debian.org>

To: 987735-done@bugs.debian.org

Subject: Re: Bug#987735: allow deployments in non-empty directories

Date: Fri, 03 Sep 2021 08:03:06 +0200

[Message part 1 (text/plain, inline)]

> I see, that's why i missed it. So I guess this is fixed.

Okay, closing.

[signature.asc (application/pgp-signature, inline)]

Send a report that this bug log contains spam.