We believe that your issue has been closed by the upload of
Version 0.33 of grml-debootstrap from Michael Prokop <mika@grml.org>.
The explanation is attached below

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Format: 1.8
Date: Mon, 15 Feb 2010 22:32:33 +0100
Source: grml-debootstrap
Binary: grml-debootstrap
Architecture: source all
Version: 0.33
Distribution: unstable
Urgency: low
Maintainer: Michael Prokop <mika@grml.org>
Changed-By: Michael Prokop <mika@grml.org>
Description: 
 grml-debootstrap - wrapper around debootstrap for installing plain Debian via grml
Changes: 
 grml-debootstrap (0.33) unstable; urgency=low
 .
   [ Michael Prokop ]
   * Do not provide the 'install bootloader into partition' feature in
     the interactive menu.
   * Install to directory: normalise the path to an absolute directory
     name.
   * remove cached debs to reduce diskpace during bootstrap (based on
     patch by gebi, thanks).
   * Fix sed usage for fall back to old behaviour in MBR handling.
   * Bump Standards-Version to 3.8.4 (no further changes).
 .
   [ Ulrich Dangel ]
   * Support and enable per default --keyring option. [Closes: issue746]
   * Do not run MAKEDEV inside the chroot.
   * Adjusted umount handling in bailout() to work with set -e
   * Always execute install_policy_rcd in chroot [Closes: issue778]
   * Install signal handler to always cleanup chroot.
   * Replace /bin/true with true in chroot-script
   * Remove duplicate : in getopt definition. Caused wrong parameter
     handling.
   * Adjusted Filename comment in cmdlineopts.clp
   * Remove trailing whitespace from debian/rules
   * Switched order of including cmdlineopts.clp, first check local
     directory
   * Only set PARTITION=1 if target is a block device.
   * Apply patch by Tong Sun <suntong@cpan.org> which addresses
     grml-debootstrap's parameter handling. Thanks!
   * Added unit tests for cmddlineopts.clp.
 .
   [ Michael Gebetsroither ]
   * introduce --pre-scripts to be executed before chroot-scripts
Checksums-Sha1: 
 ded403c41e1e8ccfb932df4fc5c4a48414aae5e5 983 grml-debootstrap_0.33.dsc
 0c720e2e0db5fbb42938aa3cbcca0752702ace1f 137388 grml-debootstrap_0.33.tar.gz
 3e1aaac8fce076771c8e771fae50d6764f932fea 113132 grml-debootstrap_0.33_all.deb
Checksums-Sha256: 
 970aa4491a9fa6c034189da94c0cd7c3316d6f1988bec82b7b3668dc96e6a7c0 983 grml-debootstrap_0.33.dsc
 f66b20bf28b24e41c45b67dc5db032cd5a534526b942613c6ed02968371acd44 137388 grml-debootstrap_0.33.tar.gz
 22e5f76e4da16c0e0cce20e6bb9b05bd45792d298c3a18f644faddc9b2da10f2 113132 grml-debootstrap_0.33_all.deb
Files: 
 ed668dee44f7a3ded91642707051a5b8 983 grml optional grml-debootstrap_0.33.dsc
 8f9e594bce80f7a690fd5f935b991da2 137388 grml optional grml-debootstrap_0.33.tar.gz
 4bb1b0e5565395aa4d2127f792a95bf5 113132 grml optional grml-debootstrap_0.33_all.deb

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)

iD8DBQFLeb3S2N9T+zficugRAts0AJ9fkqpsgRlVeNU6CTnahl2+N1tNAgCfWPYB
ptT3RK5DytWnuqd4sbZX4ws=
=WJUU
-----END PGP SIGNATURE-----

On Monday 23 November 2009 05:45:27 Tong Sun wrote:

> Thanks Ulrich. I believe that clean up for install_policy_rcd is always
> executed as well, right?

What do you mean with cleanup? As its not longer a stage you do not have to 
clean the stage files.
If you meant with cleanup the policy-rc.d file, yes it will be deleted.

Ulrich

On Sun, Nov 22, 2009 at 9:31 AM, Ulrich Dangel <bts@bts.grml.org> wrote:

>
> Ulrich Dangel <mru@grml.org> added the comment:
>
> Fixed in git. Have a look at
>
> http://git.grml.org/?p=grml-debootstrap.git;a=commitdiff;h=10bd5c915d6e0fd7630a6d76b21d9872a9665702
>

Thanks Ulrich. I believe that clean up for install_policy_rcd is always
executed as well, right?

Fixed in git. Have a look at
http://git.grml.org/?p=grml-debootstrap.git;a=commitdiff;h=10bd5c915d6e0fd7630a6d76b21d9872a9665702

On Sat, Nov 21, 2009 at 2:45 PM, Ulrich Dangel <bts@bts.grml.org> wrote:

>
> > If I run the grml-debootstrap a 2nd time, will the install_policy_rcd
> stage
> > be invoked or not?
>
> Ah got it, you want that install_policy_rcd is *always* executed, even if
> it
> was executed the first time to prevent the start of any service in the
> second
> run?
>

Yes, exactly.

 People might specify different package files on the following runs, which
might contain service packages.


thanks.

On Saturday 21 November 2009 20:20:26 Tong Sun wrote:

Hi,

> If I run the grml-debootstrap a 2nd time, will the install_policy_rcd stage
> be invoked or not?

Ah got it, you want that install_policy_rcd is *always* executed, even if it 
was executed the first time to prevent the start of any service in the second 
run?

regards,
Ulrich

On Saturday 21 November 2009 20:20:26 Tong Sun wrote:
> Tong Sun <suntong001@gmail.com> added the comment:
> 
> On Sat, Nov 21, 2009 at 2:14 PM, Ulrich Dangel <bts@bts.grml.org> wrote:
> > Ulrich Dangel <mru@grml.org> added the comment:
> >
> > On Saturday 21 November 2009 19:53:42 Tong Sun wrote:
> > > Shouldn't the install_policy_rcd stage be cleared on successfully
> > > finish the script?
> >
> > It is cleaned in the finalize function.
> 
> What do you mean by "cleaned"?

rm -f /usr/sbin/policy-rc.d

What do you mean by cleared?

> If I run the grml-debootstrap a 2nd time, will the install_policy_rcd stage
> be invoked or not?

No, as /etc/debootstrap/stages/install_policy_rcd contains the string done

Ulrich

On Sat, Nov 21, 2009 at 2:14 PM, Ulrich Dangel <bts@bts.grml.org> wrote:

>
> Ulrich Dangel <mru@grml.org> added the comment:
>
> On Saturday 21 November 2009 19:53:42 Tong Sun wrote:
>
> > Shouldn't the install_policy_rcd stage be cleared on successfully finish
> >  the script?
>
> It is cleaned in the finalize function.
>

What do you mean by "cleaned"?

If I run the grml-debootstrap a 2nd time, will the install_policy_rcd stage
be invoked or not?

thanks

On Saturday 21 November 2009 19:53:42 Tong Sun wrote:

> Shouldn't the install_policy_rcd stage be cleared on successfully finish
>  the script?

It is cleaned in the finalize function. Or did you mean something else? If 
not i will close this issue.

regards
Ulrich

Hi,

Shouldn't the install_policy_rcd stage be cleared on successfully finish the
script?

thanks