Bug#983408: New upstream release 0.0.24

[Marco Amadori]

Package: 0ad
Version: 0.0.23.1-5+b1
Severity: wishlist
Tags: upstream

New tarballs available at: https://play0ad.com/download/source/

-- 
ESC:wq

Bug#843781: An ugly patch that works to /usr/sbin/update-flashplugin-nonfree while the remote server gets an update

[Marco Amadori]

--- update-flashplugin-nonfree  2016-11-12 16:24:55.071017791 +0100
+++ update-flashplugin-nonfree_ugly_patched 2016-11-12
16:25:12.754735526 +0100
@@ -196,6 +196,7 @@
   then
   cp $cachedir/get-upstream-version.pl .
   upstream=`perl get-upstream-version.pl $arch_wget 2>
/dev/null` || true
+   upstream="11.2.202.643"

   if [ "$upstream" = "" ]
   then

-- 
ESC:wq

Bug#731300: Patch link

[Marco Amadori]

This link was provided also in a control message, but I'll report here for
convenience.

https://raw2.github.com/linuxmint/systemd/master/debian/patches/pam-check-runtime-dir-user.patch

-- 
ESC:wq

Bug#698777: python3.3: pyvenv-3.3 fails with easy-install pip

[Marco Amadori]

Package: python3.3
Version: 3.3.0-8
Severity: normal

I cannot use easy_install in a virtualenv freshly created with
pyvenv-3.3, to reproduce:

$ cd /tmp
$ pyvenv-3.3 testenv
$ curl -O http://python-distribute.org/distribute_setup.py
$ . testenv/bin/activate
$ (testenv) python distribute_setup.py
$ (testenv) easy_install pip
Traceback (most recent call last):
  File "/tmp/testenv/bin/easy_install", line 5, in 
from pkg_resources import load_entry_point
ImportError: No module named 'pkg_resources'


-- System Information:
Debian Release: wheezy/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386

Kernel: Linux 3.6-trunk-amd64 (SMP w/8 CPU cores)
Locale: LANG=it_IT.UTF-8, LC_CTYPE=it_IT.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages python3.3 depends on:
ii  libpython3.3-stdlib  3.3.0-8
ii  mime-support 3.52-2
ii  python3.3-minimal3.3.0-8

python3.3 recommends no packages.

Versions of packages python3.3 suggests:
ii  binutils   2.22-7.1
pn  python3.3-doc  

-- no debconf information

-- 
ESC:wq

Bug#695758: pyvenv-3.3 does not play well with distribute

[Marco Amadori]

Package: python3.3
Version: 3.3.0-6
Tags: experimental, fixed-upstream

To reproduce the bug:

$ pyvenv-3.3 /tmp/testenv
$ cd  /tmp/testenv # just to not pollute fs for this test
$ . /tmp/testenv/bin/activate
(testenv) $ curl -O http://python-distribute.org/distribute_setup.py
(testenv) $ python distribute_setup.py
(testenv) $ which easy-install

And you will see an output empty string or your globally installed
easy-install instead of the one from the virtualenv.

The problem is that distribute installs in /tmp/testenv/local/bin instead
of /tmp/testenv/bin and PATH after activate gets /tmp/testenv/bin appended.

So virtualenvs would do not work without a symlink to/from /local/bin

I opened a bug on python bts [0] but it seems to be relative to the debian
configuration. I tried to reproduce it with upstream tarball and default
configuration (./configure, make, make install) and does not happen with it
(it correctly installs in /tmp/testenv/bin)

I also have a hackish workaround here [1] which I hope to trash soon.

[0] http://bugs.python.org/issue16480
[1]
http://ipoveraviancarriers.blogspot.it/2012/11/python-33-and-pyvenv-hackish-solution.html

Thanks.

-- 
ESC:wq

Bug#659112: Please update python-selenium to version 2.18.1

[Marco Amadori]

Package: python-selenium
Severity: wishlist

A new upstream is available.

thanks.
-- 
ESC:x



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

Bug#642335: Subject: RFS: qjoypad

[Marco Amadori]

Dear mentors,

I am looking for a sponsor for my package "qjoypad".

 * Package name: qjoypad
   Version : 4.1.0-1
   Upstream Author : John Toman 
 * URL : http://qjoypad.sourceforge.net
 * License : GPL2+
   Section : x11

It builds those binary packages:

qjoypad- Joypad/Joystick to key/mouse remapper

To access further information about this package, please visit the following 
URL:

  http://mentors.debian.net/package/qjoypad

Alternatively, one can download the package with dget using this command:

  dget -x 
http://mentors.debian.net/debian/pool/main/q/qjoypad/qjoypad_4.1.0-1.dsc

I would be glad if someone uploaded this package for me.

Kind regards,

Marco Amadori
-- 
ESC:wq



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

Bug#642335: I'm working on It

[Marco Amadori]

Hi,
I packaged qjoypad for debian, I will upload a preliminar version soon (on 
alioth/mentors I think) while I'll look for a sponsor for the package.

Stay tuned :-)

-- 
ESC:wq



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

Bug#616143: Clarification request about --control-path interface

[Marco Amadori]

Joy Hess sayd:

> multiarch moves the directory, so all of this could break.
> (Although given current debootstrap symlinking, maybe not.)

I have some questions, sorry for being too naïve there, but I would
like to fully understand if my assumptions are right.

> This could all be future-proofed by running dpkg-query --control-path
> in the chroot.

The manpage of dpkg-query states that:

- * -
-c, --control-path package-name [control-file]
  List  paths  for  control files installed to your system
from package-name.  If control-file is specified
  then only list the path for that control file if it is
present. Warning: this command is semi-public,  it
  should  be used only as a last resort solution, and if
no other interface is available. It might get dep‐
  recated later on if better interfaces or the current
architectural deficiencies have been solved.
- * -

The above "future proof" and "deprecated later" about the same
interface, rise me some doubts.

Maybe it is just the manpage od dpkg-query that should be updated here?

Regarding the code in live-installer, If I understood that matter
right, I should call "dpkg-query --control-path" for each package
which I'm interested in.

I cannot call 'dirname "$(dpkg-query --control-path dpkg | head -1)"'
to have a common path that I could use through previous code (meaning
for all packages)

This because multiarch could give back different (base) paths for
different package architectures.

I got it?

-- 
ESC:wq



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

Bug#591791: Upstart and sysvinit in packages - Policy needed

[Marco Amadori]

On Monday 14 February 2011 21:07:57 Steve Langasek wrote:

> > if [ -e "/etc/init/${NAME}.conf" ] && /sbin/telinit --version >/dev/null
> > 2>&1
> > | grep -qs upstart
> > then
> > exit 0
> > fi
> > 
> > inside the sysvinit init script.
> 
> This does not appear to be consistent with
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=591791#42, in particular:
> 
>   If nothing else, an init script that returns success on 'start' without
>   starting the service would be inconsistent with how we've expected all
>   init scripts to work to date.  (It's not *quite* a policy violation, but
>   near enough to.) And if you argue that nothing should ever call these
> init scripts because everything should use invoke-rc.d, then things using
> this upstart-aware interface will never see the return code anyway, right?

I'm so sorry, in the hurry I put an exit 0 for the wrong psycological reasons, 
you are right.

> Oh, and if you redirect telinit stdout and stderr both to /dev/null, that
> grep is always going to return false.

This is a bad mistake for sure! I apologize to all for being so naïve and 
publishing a bad tested code (I tested an uncommitted previous release, then I 
"improved" and submitted without tests!!)

Again, I apologize.

> Also, I strongly counsel *against* shipping this, or any, upstart job in a
> Debian package until this policy bug is *fixed*.

This seems a good point, having that debhelper prefer upstart jobs over 
sysvinit when both are available.

> > Debhelper seems to not permit to have both, e.g., an upstart and a
> > sysvinit script available in the package, although with a trick like the
> > above mentioned one, the could happily live together in perfect armony,
> > side by side on my wheezy.
 
> Debhelper doesn't permit both because policy does not specify how this is
> supposed to work.  I have already submitted a patch to debhelper (bug
> #577040), which I have asked Joey to sit on until policy is finalized and
> various other key packages have implemented the necessary support (namely,
> sysvinit and lsb-base).

I hope that this discussion could help in clarifing this matter, probably the 
times are mature enough.

-- 
ESC:wq



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

Bug#591791: Upstart and sysvinit in packages - Policy needed

[Marco Amadori]

In data Saturday 12 February 2011 05:55:16, Joachim Breitner ha scritto:

>> I included some patches to have nodm gracefully uses the upstart job.

>> Since those patches permits to have both init scripts in the system, no
>> matter if upstart or sysvinit is used, a little more effort is required
>> to proper build this package.
 
> thanks for your patches. I applied them to the repo besides the patch
> 0005-Exit-from-the-sysvinit-script-to-use-the-upstart-job.patch. Is that
> code taken from some existing package?

I took the idea from discussion on bug #591791 and updated to work also with 
current sysvinit, that does currently not likes the syntax "--version".

I include again this patch to permits other people to follow the discussion 
there, but is just a matter of :

if [ -e "/etc/init/${NAME}.conf" ] && /sbin/telinit --version >/dev/null 2>&1 
| grep -qs upstart
then
exit 0
fi

inside the sysvinit init script.

> I feel like the semantics of both
> upstart jobs and init scripts in one package should be the same across
> packages, so I’d rather see a common solution for that.

Yes, probably something better and simpler than my lines above should emerge 
from the discussion.

>> But I'm stuck as you are (in the git repo) because dh will refuse to add
>> an init.d script if an upstart job is available.
>> Any hints on how to neatly have both init script files on the binary
>> package?

> Maybe you could raise the issue on d-devel?

I CCed the bug in the policy that seems to be the best match for the 
discussion, anyway I'll CC also d-devel to have some help.

Even if all agrees that my code above is right (a thing that I cannot imagine 
to be true in any universe), if it vital that a policy is given for that 
issue.

I think it would be interesting for wheezy to easely permit an Admin to choose 
an alternate init system and to permit to package maintainer to carefully 
provide init script tailored for a particular system of interest.

Debhelper seems to not permit to have both, e.g., an upstart and a sysvinit 
script available in the package, although with a trick like the above 
mentioned one, the could happily live together in perfect armony, side by side 
on my wheezy.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

From f5f51aadd12c3053ad1bc7fee2253a3a5062aa5f Mon Sep 17 00:00:00 2001
From: Marco Amadori 
Date: Fri, 11 Feb 2011 15:45:27 +0100
Subject: [PATCH 5/6] Exit from the sysvinit script to use the upstart job.

---
 debian/changelog |3 ++-
 debian/nodm.init |6 ++
 2 files changed, 8 insertions(+), 1 deletions(-)

diff --git a/debian/changelog b/debian/changelog
index faa6817..c9711ab 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -6,8 +6,9 @@ nodm (0.7-2) UNRELEASED; urgency=low
 
   [ Marco Amadori ]
   * Update Standards-Version to 3.9.1 (no changes required).
+  * Exit from the sysvinit script to use the upstart job.
 
- -- Marco Amadori   Fri, 11 Feb 2011 15:36:30 +0100
+ -- Marco Amadori   Fri, 11 Feb 2011 15:45:07 +0100
 
 nodm (0.7-1.1) unstable; urgency=low
 
diff --git a/debian/nodm.init b/debian/nodm.init
index 1b8bb6e..15ca887 100644
--- a/debian/nodm.init
+++ b/debian/nodm.init
@@ -18,6 +18,12 @@ NAME=nodm
 PIDDIR=/var/run/
 PIDFILE=${PIDDIR}/${NAME}.pid
 
+if [ -e "/etc/init/${NAME}.conf" ] && /sbin/telinit --version >/dev/null 2>&1 | grep -qs upstart
+then
+	# An upstart job exists and upstart is in use, exit gracefully
+	exit 0
+fi
+
 NODM_ENABLED=no
 NODM_XINIT=/usr/bin/xinit
 NODM_FIRST_VT=7
-- 
1.7.2.3

Bug#603974: Info provided, help needed!

[Marco Amadori]

In data Wednesday 26 January 2011 02:43:12, Colin Watson ha scritto:

> Please review and comment.  I can go ahead and push these patches to
> master if people like them.

Your patches are working on my test environment too. I like them and yes, 
please push them.

They are fixing this bug, many thanks for that quick and neat job.

If I understand also what I read on IRC, this imply that "console-setup-udeb" 
is not required to be available on pool/ directory at all, or maybe just for 
the GTK+ installer, right?

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#603974: Logs of live-installer runs

[Marco Amadori]

Too ease the resolution of this bug, I attached the log of debian installer 
with live-installer with and without a "Depends: console-setup-udeb" in the 
live-installer debian/control file.

In the nodep install I got XKBLAYOUT="us" instead of "it" in 
/etc/default/keyboard and calling setupcon at first reboot didn't fixed the 
keyboard layout.

Having console-setup-udeb as dependency of live-installer produced a correct 
value for XKBLAYOUT (it), still setupcon at first reboot does not fix keyboard 
layout.

Using the live-installer with explicit calls for console-setup-udeb (idea took 
from auto-install postinst) still does not fix the keyboard console but fix it 
after a "setupcon" invocking upon first reboot.

Test system description:

The images were generated with those auto/config files and Makefile attached 
(there for convenience), with a live-build that includes console-setup-udeb 
and with live-installer patched as mentioned and put in config/binary_local-
udebs/ as needed.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



debian-live-nodep-syslog.txt.gz
Description: GNU Zip compressed data
#!/usr/bin/make -f

ISOFILE := $(shell echo live_rb_`date +%F`.iso)
IMAGE := binary-hybrid.iso

default: config build

config:
	lb config

build: $(IMAGE)

$(IMAGE):
	su-to-root -c "lb build"

test: $(IMAGE)
	su-to-root -c "qemu -boot d -cdrom binary.iso -m 256 -net nic,vlan=0 -net user,vlan=0 -net nic,vlan=1 -net tap,vlan=1 -hda hda.qcow2"

clean:
	-su-to-root -c "lb clean"
	-find config -maxdepth 1 -type f -delete

clean_mounts:
	-su-to-root -c "lb chroot_sysfs remove"
	-su-to-root -c "lb chroot_proc remove"
	-su-to-root -c "lb chroot_devpts remove"

set_mounts:
	-su-to-root -c "lb chroot_sysfs install"
	-su-to-root -c "lb chroot_proc install"
	-su-to-root -c "lb chroot_devpts install"

dist_clean: clean
	-su-to-root -c "lb clean --purge"

.PHONY: clean config



config
Description: application/shellscript

Bug#603974: Info provided, help needed!

[Marco Amadori]

This bug seems to rely on two missing bits on two packages (patches included).

the package "live-build" does not include console-setup-udeb in the cdrom 
"pool/" directory and live-installer do not "Depends:" on console-setup-udeb.

So the "/target" system have default "/etc/default/keyboard" file instead of a 
proper one.

Fixing those two above mentioned quirks is not enough it seems, because the 
resulting target, although having a proper configuration file, still needs a 
"setupcon" after reboot in order to have things working.

I see that live-installer calls "setupcon --save-only" in the target, so, by 
reading man page, all should be done.

But strangely it is not the case.

Could anyone from debian installer team, with a deeper understanding of d-i 
than me, suggests a solution there?

The patches included are there just for convenience.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

From d59109c0969f4ef0c469bd0a72b940085e243c8b Mon Sep 17 00:00:00 2001
From: Marco Amadori 
Date: Mon, 24 Jan 2011 11:33:15 +0100
Subject: [PATCH] Fix missing keyboard setup adding console-setup-udeb dependency (Closes: #603974).

---
 debian/changelog |7 ++-
 debian/control   |3 ++-
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/debian/changelog b/debian/changelog
index 9fff7e1..6fa99ad 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,8 +1,13 @@
 live-installer (30) UNRELEASED; urgency=low
 
+  [ Otavio Salvador ]
   * Avoid the build-depends on autotools.
 
- -- Otavio Salvador   Tue, 04 Jan 2011 10:57:25 -0200
+  [ Marco Amadori ]
+  * Fix missing keyboard setup by adding console-setup-udeb
+dependency (Closes: #603974).
+
+ -- Marco Amadori   Mon, 24 Jan 2011 11:30:22 +0100
 
 live-installer (29) unstable; urgency=low
 
diff --git a/debian/control b/debian/control
index ca83145..b6beeb0 100644
--- a/debian/control
+++ b/debian/control
@@ -16,7 +16,8 @@ XC-Package-Type: udeb
 Architecture: any
 Depends:
  ${shlibs:Depends}, archdetect, base-installer (>= 1.105),
- busybox-udeb (>= 1:1.13.3), cdebconf-udeb, created-fstab, mounted-partitions
+ busybox-udeb (>= 1:1.13.3), cdebconf-udeb, created-fstab, mounted-partitions,
+ console-setup-udeb
 Provides: installed-base, kernel-installer
 XB-Installer-Menu-Item: 6500
 Description: Install the system
-- 
1.7.2.3

From dce8c569e953722f2191e8126432a55a396acd79 Mon Sep 17 00:00:00 2001
From: Marco Amadori 
Date: Tue, 16 Nov 2010 17:47:17 +0100
Subject: [PATCH] Do not exclude console-setup-udeb (Helps: #603974)

* it is not in d-i initrd image anymore, and is required to pass keyboard
configuration to /target console-setup package.

Signed-off-by: Marco Amadori 
---
 data/debian-cd/squeeze/exclude-udebs |1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/data/debian-cd/squeeze/exclude-udebs b/data/debian-cd/squeeze/exclude-udebs
index 1faaa9a..ba3ed4f 100644
--- a/data/debian-cd/squeeze/exclude-udebs
+++ b/data/debian-cd/squeeze/exclude-udebs
@@ -132,7 +132,6 @@ auto-install
 libslang2-udeb
 kbd-chooser
 kbd-udeb
-console-setup-*
 # Currently unused
 nbd-client-udeb
 pwgen-udeb
-- 
1.7.2.3

Bug#603974: live-installer should depend and use console-setup-udeb

[Marco Amadori]

In data venerdì 19 novembre 2010 00:47:56, Samuel Thibault ha scritto:

> Marco Amadori, le Fri 19 Nov 2010 00:29:59 +0100, a écrit :
> > As it is now, live-installer fails in providing a proper keyboard
> > configuration in the installed system, in fact it fails in propagating
> > kbd-chooser debconf values.
> > 
> > Something changed in how console-setup-udeb is handled in d-i, and for
> > squeeze i found that depending on it and invoking the postinst would
> > succeed.
 
> I don't understand this. How was console-setup-udeb involved in
> live-installer before?

I didn't checked how it was handled before, maybe something changed in base-
installer, of which live-installer exploits some helpers.

> As I understand your sentences, live-installer
> seems to be using the kbd-chooser udeb, and installing console-setup in
> the installer system.

It does not do that directly, my guessing is that it was automatically cared 
of before.

> If so, then the bug is rather in console-setup,
> not in console-setup-udeb.

I'm pretty sure it isn't the case, I checked what console-setup does in the 
target and it is doing it right, but its debconf database it is empty (I mean 
default values).

>From what I understood kbd-chooser ask keyboard layout question and translate 
its results to console-setup debconf values, then console-setup-udeb 
propagates them to the target system.

auto-install udeb do something similar as my patch does. It depends on 
console-setup-udeb and explicitly calls its postinst.

> Depending on console-setup-udeb is just a
> convoluted way to fix things, as I guess what its postinst does to fix
> this in your case is just doing what console-setup is supposed to be
> doing in the installed system, already...

This is probably true. My patch seems hackish even to me although it works, I 
hope that someone with a deeper understanding of debian-installer internals 
could provide a cleaner fix if possible.

Anyway this fix is better than having an user unable to have its keyboard 
supported at all.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#603975: live-build should provide console-setup-udeb available to debian-installer

[Marco Amadori]

Package: live-build
Severity: serious
Version: any
Owner: debian-l...@lists.debian.org
Tags: patch

live-build for squeeze should provide console-setup-udeb, since it is no more 
available in the inirtd as it was.

Failing in doing that would make impossible for debian-installer to setup the 
keyboard layout in the installed system.

Patch attached (and available in git with a typo in commit message).
-- 
ESC:wq
From 3a559425e4f443ee776ba98105d132e793553e79 Mon Sep 17 00:00:00 2001
From: Marco Amadori 
Date: Tue, 16 Nov 2010 17:47:17 +0100
Subject: [PATCH] Do not exclude console-setup-udeb.

* It is not available in debian-installer's initrd image anymore.

Signed-off-by: Marco Amadori 
---
 data/debian-cd/squeeze/exclude-udebs |1 -
 1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/data/debian-cd/squeeze/exclude-udebs b/data/debian-cd/squeeze/exclude-udebs
index 1faaa9a..ba3ed4f 100644
--- a/data/debian-cd/squeeze/exclude-udebs
+++ b/data/debian-cd/squeeze/exclude-udebs
@@ -132,7 +132,6 @@ auto-install
 libslang2-udeb
 kbd-chooser
 kbd-udeb
-console-setup-*
 # Currently unused
 nbd-client-udeb
 pwgen-udeb
-- 
1.7.2.3

Bug#603974: live-installer should depend and use console-setup-udeb

[Marco Amadori]

Package: live-installer
Severity: serious
Version: 27
Tags: patch
Owner: debian-l...@lists.debian.org

As it is now, live-installer fails in providing a proper keyboard configuration 
in the installed system, in fact it fails in propagating kbd-chooser debconf 
values.

Something changed in how console-setup-udeb is handled in d-i, and for squeeze 
i found that depending on it and invoking the postinst would succeed.

Note that live-build fails to provide console-setup-udeb (fixed in git) 
available in "pool/" and since console-setup-udeb is not available anymore in 
the initrd, to test this patch you need to fix live-build exclude-udeb or 
provide a console-setup-udeb in some way.

Patch attached.

-- 
ESC:wq
From 5c3dfb5cb1d660d6824bc3c4ebdecbbc11bf024c Mon Sep 17 00:00:00 2001
From: Marco Amadori 
Date: Tue, 16 Nov 2010 17:19:26 +0100
Subject: [PATCH] Fix keyboard setup.

* Explicit depends on console-setup-udeb and call its postinst to
  transfer d-i choosed keymap to target system.
---
 debian/changelog   |6 ++
 debian/control |2 +-
 live-installer.d/console-setup |5 +
 3 files changed, 12 insertions(+), 1 deletions(-)
 create mode 100755 live-installer.d/console-setup

diff --git a/debian/changelog b/debian/changelog
index 1d43205..a3317ab 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,9 @@
+live-installer (28) UNRELEASED; urgency=low
+
+  * Fix missing keyboard setup by calling console-setup-udeb postinst.
+
+ -- Marco Amadori   Tue, 16 Nov 2010 17:15:33 +0100
+
 live-installer (27) unstable; urgency=low
 
   * We should depends on created-fstab since it needs to be available to
diff --git a/debian/control b/debian/control
index a18e966..8910e0f 100644
--- a/debian/control
+++ b/debian/control
@@ -17,7 +17,7 @@ Architecture: any
 Depends:
  ${shlibs:Depends}, archdetect, base-installer (>= 1.105),
  busybox-udeb (>= 1:1.13.3), cdebconf-udeb, created-fstab, mounted-partitions,
- squashfs-modules
+ squashfs-modules, console-setup-udeb
 Provides: installed-base, kernel-installer
 XB-Installer-Menu-Item: 6500
 Description: Install the system
diff --git a/live-installer.d/console-setup b/live-installer.d/console-setup
new file mode 100755
index 000..37ffa8e
--- /dev/null
+++ b/live-installer.d/console-setup
@@ -0,0 +1,5 @@
+#!/bin/sh -e
+
+if [ -e /var/lib/dpkg/info/console-setup-udeb.postinst ]; then
+	/var/lib/dpkg/info/console-setup-udeb.postinst
+fi
-- 
1.7.1

Bug#536728: persistent devices name are supported via persistent-subtext boot option

[Marco Amadori]

In data lunedì 27 settembre 2010 13:13:21, Michael Prokop ha scritto:

> Marco, could you please document this feature?

It is already documented in the live-manual:

http://live.debian.net/manual/en/html/persistence.html#id2559143

> I can't find it in manpages/*/live-boot.7 yet, would be great if
> documentation matches implementation. :)

I agree that when accepting code that introduces new features which adds new 
configurations would be great to have those properly documented also in 
manpages.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#595371: zlib1g: experimental version breaks php-pear

[Marco Amadori]

Package: zlib1g
Version: 1:1.2.3.5.dfsg-1
Severity: important
Tags: upstream

This version makes pear segfault on decompression.

>From the gdb trace (below) I suggest it is already known by upstream as I see 
in this mail:

http://mail.madler.net/pipermail/zlib-devel_madler.net/2010-
January/001331.html

So please do not upload zlib1g in unstable yet :-)

# gdb /usr/bin/php
GNU gdb (GDB) 7.1-debian
Copyright (C) 2010 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu".
For bug reporting instructions, please see:
...
Reading symbols from /usr/bin/php...(no debugging symbols found)...done.
(gdb) run -C -q -d include_path=/usr/share/php -d output_buffering=1 
/usr/share/php/pearcmd.php -vvv install Var_Dump
Starting program: /usr/bin/php -C -q -d include_path=/usr/share/php -d 
output_buffering=1 /usr/share/php/pearcmd.php -vvv install Var_Dump
[Thread debugging using libthread_db enabled]

Warning: file_exists(): Unable to find the wrapper "channel" - did you forget 
to enable it when you configured PHP? in PEAR/Downloader/Package.php on line 
1517

Warning: is_file(): Unable to find the wrapper "channel" - did you forget to 
enable it when you configured PHP? in PEAR/Downloader/Package.php on line 1527

Warning: is_file(): Unable to find the wrapper "channel" - did you forget to 
enable it when you configured PHP? in PEAR/Downloader/Package.php on line 1527
Downloading "http://pear.php.net/get/Var_Dump-1.0.4.tgz";
downloading Var_Dump-1.0.4.tgz ...
Starting to download Var_Dump-1.0.4.tgz (17,426 bytes)
..done: 17,426 bytes

Program received signal SIGSEGV, Segmentation fault.
*__GI___libc_free (mem=0x2) at malloc.c:3710
3710malloc.c: No such file or directory.
in malloc.c
(gdb) bt
#0  *__GI___libc_free (mem=0x2) at malloc.c:3710
#1  0x7799a664 in inflateEnd () from /usr/lib/libz.so.1
#2  0x779967a1 in gzclose_r () from /usr/lib/libz.so.1
#3  0x00478458 in ?? ()
#4  0x0063e665 in _php_stream_free ()
#5  0x0063e8f1 in ?? ()
#6  0x0068a6ee in ?? ()
#7  0x00689885 in zend_hash_del_key_or_index ()
#8  0x0068a999 in _zend_list_delete ()
#9  0x005beca0 in zif_fclose ()
#10 0x006cb63a in ?? ()
#11 0x006a29d0 in execute ()
#12 0x0067a6ed in zend_execute_scripts ()
#13 0x00626238 in php_execute_script ()
#14 0x0070aa23 in ?? ()
#15 0x7526cc4d in __libc_start_main (main=, 
argc=, 
ubp_av=, init=, fini=, 
rtld_fini=, stack_end=0x7fffe118) at libc-
start.c:228
#16 0x0042d229 in _start ()
(gdb) q
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#591330: live-snapshot fails to sync at reboot/shotdown

[Marco Amadori]

In data lunedì 2 agosto 2010 10:45:21, Luigi Capriotti ha scritto:

I'll paste relevant lines there for convenience:

> http://live.debian.net/gitweb?p=live-boot.git;a=blob;f=scripts/live;h=a17eb
> 423fca132f0aaf03c8ca6f956766eaef797;hb=HEAD#l1052

echo "export ${snap_type}SNAP="/cow${snap_mount#$rootmnt}":${snapdev}:
${snapfile}" >> /etc/live.conf # for resync on reboot/halt

where snap_type is 

> and parsed with:
 
> http://live.debian.net/gitweb?p=live-boot.git;a=blob;f=bin/live-snapshot;h=
> 08e5fd654adf08609dc9d7a411da83fc58b3f3a1;hb=HEAD#l245

SNAP_COW=$(echo "${SNAP_RESYNC_STRING}" | sed -e 's|
^/root\([^:.]*\).*$|'"${DEF_SNAP_COW}"'\1|')
SNAP_DEV=$(echo "${SNAP_RESYNC_STRING}" | cut -f2 -d ':')
SNAP_MNT=$(echo "${SNAP_RESYNC_STRING}" | cut -f3 -d ':')

${snap_mount#$rootmnt} should then be either "/root/" or "/root/home"


-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#588307: live-initramfs: Unreachable code in Arguments

[Marco Amadori]

In data mercoledì 7 luglio 2010 10:51:57, Colin Watson ha scritto:

> Only the first pattern will be matched.  The second is therefore
> unreachable code.

Agree, a patch is coming on git soon. Thanks for reporting.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#588306: live-initramfs: Unreachable code in do_netmount

[Marco Amadori]

In data mercoledì 7 luglio 2010 10:49:06, Colin Watson ha scritto:

> do_netmount is called in exactly one place:
> 
> if [ ! -z "${NETBOOT}" ] || [ ! -z "${FETCH}" ] || [ ! -z
> "${HTTPFS}" ] || [ ! -z "${FTPFS}" ] then
> if do_netmount
> 
> That function then contains code guarded by this conditional:
> 
> if [ -z "${NETBOOT}" ] && [ -z "${FETCH}" ] && \
>[ -z "${HTTPFS}" ] && [ -z "${FTPFS}" ]
> then

It is "do_netsetup" that contains that code, not "do_netmount", so the rest of 
"do_netsetup" will be executed.

It seems to me that this bug should be tagged invalid for that reason.

> That code can therefore never be called.  When is it supposed to be run?

That code seems to be added when the networking of live-initramfs leveraged 
the initramfs-tools one.

It is quite a time since I didn't personally tested that netboot (cifs or nfs) 
code, did you found that it currently do not work? If that is the case please 
open a new bug report and I'll do a regression test/fix cycle.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#572520: Which games already uses that?

[Marco Amadori]

Hi, 
It seems an interesting project, but I wander which games are using it which 
could be packaged as well.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#571544: Unreproducible but a different bug still there

[Marco Amadori]

Hi,
referring to your (forwarded to http://bugs.debian.org/cgi-
bin/bugreport.cgi?bug=571544) bug, when I try to launch spring-dedicated I got 
an LD issue (which needs to be fixed but is not relative to this bug):

*  *
$ /usr/games/spring-dedicated
/usr/games/spring-dedicated: error while loading shared libraries: 
libspringserver.so: cannot open shared object file: No such file or directory
*  *

if I call it specifying the LD_LIBRARY_PATH, that way:

*  *
$ LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/lib/spring spring-dedicated
If you find any errors, report them to mantis or the forums.

LogOutput initialized.
Spring 0.81.2.0
Build date/time: Feb 22 2010 21:19:02
Available log subsystems: ArchiveScanner, VFS-detail, VFS,
Enabled log subsystems:
Enable or disable log subsystems using the LogSubsystems configuration key
  or the SPRING_LOG_SUBSYSTEMS environment variable (both comma separated).
Using read-only  data directory: /usr/games/
Using read-write data directory: /home/marco/.spring/
Using read-only  data directory: /usr/share/games/spring/
Using read-only  data directory: /usr/lib/spring/
Scanning: /usr/share/games/spring/maps
Scanning: /usr/share/games/spring/base
Scanning: /usr/share/games/spring/mods
usage: spring-dedicated 
*  *

It seems to work properly, so I cannot reproduce (on an amd64 package, version 
0.81.2+dfsg1-1 from unstable) your issue.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#569324: springlobby: update to upstream version 0.63

[Marco Amadori]

On Tuesday 16 February 2010 21:30:24 Jan Dittberner wrote:

> please update springlobby to the latest upstream version

I see a @debian.org address there, I can prepare you a package for 0.63 on 
mentors.debian.net to let you upload it, I'm having some troubles finding a 
sponsor.

An updated spring packages is already available there.

Contact me for sponsorship if you like, but it is better to just join debian-
games team if you are not already in.

> and enable
> bittorrent support.

This is not possible, the libtorrent-rasterbar package depends on openssl 
which conflicts with GPL without a proper license exception.

Look at #552622 and upstream # for more informations:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=552622
http://springlobby.info/issues/

> The attached patch series against the git
> repository linked at the PTS page performs the necessary
> modifications. You just need to do these steps

Thanks for the patches, some of them I already had in my work dir, viceversa I 
completely missed the .git* files. I'll remove them from the upstream tarball  
and I will ask upstream to not include them.

-- 
ESC:wq



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/201002162147.14145.marco.amad...@gmail.com

Bug#544367: Release 3.4.5 is out

[Marco Amadori]

The new news link below:

http://www.bugzilla.org/news/#release345

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#563739: Any hints for fixing this?

[Marco Amadori]

Hi,
thanks for the report, I compiled the debian/copyright file with the help of 
upstream and "licensecheck" command, do you have any suggestion or best 
practice for this task?

-- 
ESC:wq



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

Bug#554153: ITP: spring-kernelpanic -- a fast-paced RTS with no economy for the Spring engine

[Marco Amadori]

Package: wnpp
Version: N/A; reported 2009-11-03
Severity: wishlist

* Package name : spring-kernelpanic
Version : 3.6
Upstream Author :
* URL : http://springrts.com/wiki/Kernel_Panic
* License : Public Domain, GPL-2+
Description : a fast-paced RTS with no economy for the Spring engine
 Kernel Panic is a game about computers. Systems, Hackers, and Networks wage war
 in a matrix of DOOM! The only constraints are time and space; unlike other real
 time strategy games, no resource economy exists in Kernel Panic.
 .
 All units are free in this game, every factory built will be spamming units at
 all times. You can build more factories, but only on pre-defined areas
 (geothermal vents). All that remains is pure strategy and tactics.
 .
 Kernel Panic makes for a frantically fast-paced, action-oriented game, with a
 very unique graphical style.


-- 
ESC:wq



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

Bug#552628: libtorrent-rasterbar: please add a -nossl flavour of the package

[Marco Amadori]

Package: libtorrent-rasterbar-dev
Severity: wishlist

Hi,
I'm packaging springlobby for debian (ITP #550250) which depends on
libtorrent-rasterbar; even if SpringLobby does not use directly openssl the
final binary will end linking it because it links this library.

I asked SpringLobby upstream to change license to include the openssl
exception but the have troubles in asking all contributors about this
license update and since they are not using ssl they asked me to link
springlobby to a ssl free version of libtorrent-rasterbar, which is not
available in debian yet.

Spring is a really modern and beautiful 3D RTS game, would you like to help
me in having it properly available for debian users by providing a
"libtorrent- rasterbar-nossl" binary package please?


Many thanks.

P.S. a "-gnutls" version probably would be better but I think this will
require more work than just extending the packaging, so I'll be fine with
just a "-nossl" flavour.

--
ESC:wq



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

Bug#552622: libtorrent-rasterbar: please add a -nossl flavour of the package

[Marco Amadori]

Source: libtorrent-rasterbar
Version: 0.14.6-1
Severity: wishlist

Hi,
I'm packaging springlobby for debian (ITP #550250) which depends on 
libtorrent-rasterbar; even if SpringLobby does not use directly openssl the 
final binary will end linking it because it links this library.

I asked SpringLobby upstream to change license to include the openssl 
exception but the have troubles in asking all contributors about this license 
update and since they are not using ssl they asked me to link springlobby to a 
ssl free version of libtorrent-rasterbar, which is not available in debian 
yet.

Spring is a really modern and beautiful 3D RTS game, would you like to help me 
in having it properly available for debian users by providing a "libtorrent-
rasterbar-nossl" binary package please?

Many thanks.

P.S. a "-gnutls" version probably would be better but I think this will 
require more work than just extending the packaging, so I'll be fine with just 
a "-nossl" flavour.

-- 
basta vi prego -> http://it.wikipedia.org/wiki/Top-posting
avoid that please! -> http://en.wikipedia.org/wiki/Posting_style#Top-posting
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#552596: libtorrent-rasterbar: please add a -nossl flavour of the package

[Marco Amadori]

Package: libtorrent-rasterbar
Version: 0.14.6-1
Severity: wishlist

Hi,
I'm packaging springlobby for debian (ITP #550250) which depends on 
libtorrent-rasterbar; even if SpringLobby does not use directly openssl the 
final binary will end linking it because it links this library.

I asked SpringLobby upstream to change license to include the openssl 
exception but the have troubles in asking all contributors about this license 
update and since they are not using ssl they asked me to link springlobby to a 
ssl free version of libtorrent-rasterbar, which is not available in debian 
yet.

Spring is a really modern and beautiful 3D RTS game, would you like to help me 
in having it properly available for debian users by providing a "libtorrent-
rasterbar-nossl" binary package please?

Many thanks.

P.S. a "-gnutls" version probably would be better but I think this will 
require more work than just extending the packaging, so I'll be fine with just 
a "-nossl" flavour.

-- 
basta vi prego -> http://it.wikipedia.org/wiki/Top-posting
avoid that please! -> http://en.wikipedia.org/wiki/Posting_style#Top-posting
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#550250: ITP: springlobby -- single/multiplayer lobby for spring rts engine

[Marco Amadori]

Package: wnpp
Severity: wishlist

* Package name : springlobby
  Version  : 0.23
  Upstream Authors : springlobby developers
* URL  : http://springlobby.info
* License  : GPL
  Description  : Single player and multiplayer lobby for Spring, it also 
permits to download maps and mods via a smart 
p2p engine, embeds an IRC-like chat client and many other features.
 
Preliminar packages are present on mentors.debian.net and binaries on 
http://alioth.debian.org/~mammadori-guest

The packages are not in their final release, some copyright files and library 
paths must be addressed but they are working 
and playable.
 

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#367883: Preliminary packages available to test

[Marco Amadori]

Hi all spring enthusiasts,
I'm packaging spring from scratch but I already looked at ubuntu's PPA work, 
I'm in contact with a DD which probably will sponsor those packages.

Preliminar packages are present on mentors.debian.net and binaries on 
http://alioth.debian.org/~mammadori-guest .

The packages are not in their final release, some copyright files and library 
paths must be addressed but they are in a working and playable state.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#522382: live-initramfs: using 'persistent=nofiles' parameter [PATCH]

[Marco Amadori]

On Wednesday 29 July 2009, 12:16:54, Daniel Baumann wrote:

> Rui Miguel Bernardo wrote:
> > Here is a correct patch (I hope).
>
> for the records (it doesn't matter here, but it's a major nuissance in
> general when sending patches): your diff is somewhat cluttered, you
> should only change what you changed on purpose and thus send a minimal
> diff only (e.g. here, the case statement is changed without changing it
> (probably changed the indenting)).
>
> anyway, Marco, can you comment on the 'break' one?

My mistake, this is the correct minimal patch (I'll git it in my repo too):

diff --git a/scripts/live-helpers b/scripts/live-helpers
index 35fb8cf..978aa06 100644
--- a/scripts/live-helpers
+++ b/scripts/live-helpers
@@ -367,7 +367,7 @@ find_cow_device ()
then
# do not mount the device to find for image 
files
# just skip this
-   break
+   continue
fi
 
case "$(get_fstype ${devname})" in


-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#275653: Bug#525012: (no subject)

[Marco Amadori]

On Sunday 26 July 2009, 16:07:22, Jan Engelhardt wrote:

>> The Official Manual example [0] in the section 10.6.1.7 does not work 
>> anymore 
>> because of this.

>> So e.g:
>> # ifconfig eth0:1 inet 0 up

>> does not work anymore

> That's because it is not an interface whatsoever (compare with running
> `ip addr`). ifconfig just displays it in a fashion that makes others
> believe it would be an interface.

You suggest that this bug should be forwarded to debian manual maintainers 
then?

I originally meant that one between the manual and the software is wrong and 
should be fixed.

>> As noted in [1] (which is the same bug of this in the ubuntu's bts) I think
>> that  dhcp3-client's important bug #275653 is caused by this bug too, so I 
>> CC it.

Could then dhcp3-client be fixed if is just the manual which is wrong?

>> [0] http://www.debian.org/doc/manuals/reference/ch-gateway.en.html#s-net-
ifupdown
>> [1] https://bugs.launchpad.net/ubuntu/+source/net-tools/+bug/356800

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#537224: Please package new upstream release 1.4.1

[Marco Amadori]

Package: directfb   
Severity: wishlist

A new directfb stable release is available upstream, would you like to package 
it, please?

Many thanks!
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#41642: Diversions should not be encouraged?

[Marco Amadori]

On  Thu, 21 Aug 2003 17:56:15 -0400, Joey Hess said:

> The problem with this idea is much the same as the problem with
> a dh_alternatives: dpkg-divert and update-alternatives are complicated
> and have many options that might need to be passed in, and adding a
> debhelper command just adds a layer of complexity with no real gain.

> Also of course, use of diversions should not be encouraged..

Could you please explain better this point ?

How a package which should configure or extend another one (think of a company 
custom "config-bind*.deb" package which setup a standard dns service for 
company's servers) should do it in a polite way?

Providing directly other packages configuration files does not work, since they 
wil conflict with other package's files.

Copying files over in postinst is suboptimal since you will lose modification 
in 
upgrades, or duplicate dpkg's work on handling condifuration files.

Diversions in /etc seemed the right thing to do, so both are configuration 
files 
and dpkg will know of the diversion.

What you would suggest, for better addressing both this example use case, and 
the general use cases handled by dpkg-divert ?

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#532620: Please provide the -dbg package for libdirectfb

[Marco Amadori]

Package: directfb   
Severity: wishlist

As most used libraries, I think it could be useful to provide the -dbg 
packages with debuggin symbols also for this package.

Many thanks for paying attention.

P.S. (new upstream is available too)
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#395334: Please update libgegl depends or it will FTBFS

[Marco Amadori]

Hi,
in the gnome-scan-0.6.2 package source found on [0], debian/control should be 
updated to make the package depends on libgegl-0.0-dev (>= 0.0.21), otherwise 
building it could lead to:

 configure output 
checking for GNOME_SCAN... configure: error: Package requirements (gmodule-2.0 
gthread-2.0 glib-2.0 gdk-2.0 >= 2.11 gdk-pixbuf-2.0 gtk+-2.0 >= 2.12 gegl >= 
0.0.21 gconf-2.0 ) were not met:


Requested 'gegl >= 0.0.21' but version of GEGL is 0.0.20
 configure output 

[0] http://sadleder.de/debian/
-- 
ESC:wq



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

Bug#528480: too easy, "patch" included

[Marco Amadori]

I was convinced that the loop was unmounted since it seemed to logical to 
me... I was wrong... the squashfs is still mounted, so there is no surprise 
that /cdrom could not me umounted.

To umount /cdrom those two simple steps must added to live-installer:

# umount /mnt
# /target/sbin/losetup -d /dev/loop0

I'm using "/target/sbin/losetup" since I did not found other losetup in $PATH 
or as busybox applet, but I'm confident that live-installer maintainer will 
probably invoke something better.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#528480: live-installer does not let d-i to eject cdrom

[Marco Amadori]

Package: live-installer

at the end of a successful installation finish-install is not able to umount 
and eject cdrom.

Lates messages are (e.g.):

 * -
May 13 07:13:12 finish-install: info: Running 
/usr/lib/finish-install.d/15cdrom-detect
May 13 07:13:12 cdrom-detect: Unmounting and ejecting '/dev/hdc'
May 13 07:13:12 finish-install: umount: cannot umount /cdrom: Device or 
resource busy
May 13 07:13:12 finish-install: umount: cannot umount /cdrom: Device or 
resource busy
May 13 07:13:12 finish-install: eject: unmount of '/cdrom' failed
May 13 07:13:12 finish-install: info: Running 
/usr/lib/finish-install.d/20final-message
 * -

fuser and lsof showed me nothing, so I tried losetup -a and I saw that 
live-installer forgot to unset the loop device used for the .squashfs:

# /target/sbin/losetup -a
/dev/loop0: [1600]:2822 (/cdrom/live/filesystem.squashfs)

# /target/usr/bin/lsof 2>/dev/null | grep loop
loop0   13671   0   cwd DIR 0,1 0 1 /
loop0   13671   0   rtd DIR 0,1 0 1 /
loop0   13671   0   txt unknown /proc/13671/exe

# ps axw | grep 13671
13671 root  0   SW<  [loop0]

# /target/sbin/losetup -d /dev/loop0
ioctl: LOOP_CLR_FD: Device or resource busy

# /target/bin/fuser /dev/loop0
# /target/bin/fuser -c /dev/loop0

nothing...

I'm stuck there.

Reproducible with qemu.
--
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#528228: A new upstream is available with security fixes

[Marco Amadori]

Package: bugzilla3
Severity: wishlist

A new usptream is available on http://www.bugzilla.org/releases/3.2.3/, it 
would be nice to have in unstable since it fixes some security bugs too.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#525962: libgnutls26 makes apt-transport-https fail with ssl key/cert client authentication

[Marco Amadori]

On Tuesday 28 April 2009, 13:02:29, Simon Josefsson wrote:

>>> What kind of server is
>>> it?

> > It is a debian lenny with apache hosting a custom reprepro repository of
> > some debian packages.

> Apache with mod_ssl or mod_gnutls?

I'm sorry, I meant apache2-mpm-worker 2.2.9-10lenny2:

# ldd /usr/sbin/apache2 | grep ssl
libssl.so.0.9.8 => /usr/lib/libssl.so.0.9.8 (0x7f140f9ec000)
# ldd /usr/sbin/apache2 | grep tls
libgnutls.so.26 => /usr/lib/libgnutls.so.26 (0x7fe71c421000)

> Could you also generate a similar log for gnutls 2.6.x that works?

Done, attached.

> I'm not sure what the log implies, it seems strange to me.  Comparing it
> with the gnutls 2.6.x log might suggest something.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



ssl.debug_working.lzma
Description: application/lzma

Bug#525962: libgnutls26 makes apt-transport-https fail with ssl key/cert client authentication

[Marco Amadori]

On Tuesday 28 April 2009, 12:02:25, you wrote:

> > yes, I attached a log after adding 'Debug::Acquire::https "true";' to the
> > apt- conf.d/ file.

> Unfortunately it doesn't give that many details.  What kind of server is
> it?

It is a debian lenny with apache hosting a custom reprepro repository of some 
debian packages.
 
> Even if apt-transport-https uses curl, it should be possible to include
> a call directly to GnuTLS in the code, like this:

I rebuilt apt-transport-https with your suggestions and attached a more 
verbose debug. (1Mb !)

I hope it could help now.

P.S. I do not see this bug showed under libgnutls26 bug page [0], could it be 
due I tagged it "lenny" ?

[0] http://bugs.debian.org/cgi-bin/pkgreport.cgi?pkg=libgnutls26

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



ssl_enhanced.debug.lzma
Description: application/lzma

Bug#525962: libgnutls26 makes apt-transport-https fail with ssl key/cert client authentication

[Marco Amadori]

On Tuesday 28 April 2009, 10:08:26, you wrote:

> Can you get apt-transport-https to generate debug logs?

yes, I attached a log after adding 'Debug::Acquire::https "true";' to the apt-
conf.d/ file.

> If there isn't
> code in it already, you may need to add something like this:

I cannot add this in a simple way, main method/https.cc just invokes

"curl_global_init(CURL_GLOBAL_SSL) ;"

> It is difficult to debug this further without the information printed by
> the gnutls log.

I hope tah the attached log could be enough.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

Hit http://develop2 lenny Release.gpg
* About to connect() to  port 443 (#0)
*   Trying ... * connected
* Connected to  () port 443 (#0)
Hit http://develop2 lenny Release
* found 1 certificates in /etc/apt/certs/-cacert.pem
Ign http://develop2 lenny/main Packages/DiffIndex
Hit http://develop2 lenny/main Packages
*server certificate verification OK
*common name:  (matched)
*server certificate expiration date OK
*server certificate activation date OK
*certificate public key: RSA
*certificate version: #3
*subject: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*start date: Wed, 22 Apr 2009 13:02:50 GMT
*expire date: Thu, 22 Apr 2010 13:02:50 GMT
*issuer: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*compression: NULL
*cipher: AES-128-CBC
*MAC: SHA1
> GET /debian/dists/lenny/Release.gpg HTTP/1.1
User-Agent: Debian APT-CURL/1.0 (0.7.21)
Host: 
Accept: */*
Cache-Control: max-age=0

* gnutls_handshake() failed: Decryption has failed.
* GnuTLS recv error (-10): The specified session has been invalidated for some 
reason.
* Connection #0 to host  left intact
* Connection #0 seems to be dead!
* Closing connection #0
* About to connect() to  port 443 (#0)
*   Trying ... * connected
* Connected to  () port 443 (#0)
* found 1 certificates in /etc/apt/certs/-cacert.pem
*server certificate verification OK
*common name:  (matched)
*server certificate expiration date OK
*server certificate activation date OK
*certificate public key: RSA
*certificate version: #3
*subject: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*start date: Wed, 22 Apr 2009 13:02:50 GMT
*expire date: Thu, 22 Apr 2010 13:02:50 GMT
*issuer: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*compression: NULL
*cipher: AES-128-CBC
*MAC: SHA1
> GET /debian/dists/lenny/Release HTTP/1.1
User-Agent: Debian APT-CURL/1.0 (0.7.21)
Host: 
Accept: */*
If-Modified-Since: Tue, 28 Apr 2009 09:38:37 GMT
Cache-Control: max-age=0

* gnutls_handshake() failed: Decryption has failed.
* GnuTLS recv error (-10): The specified session has been invalidated for some 
reason.
* Connection #0 to host  left intact
Ign https:// lenny Release
* Connection #0 seems to be dead!
* Closing connection #0
* About to connect() to  port 443 (#0)
*   Trying ... * connected
* Connected to  () port 443 (#0)
* found 1 certificates in /etc/apt/certs/-cacert.pem
*server certificate verification OK
*common name:  (matched)
*server certificate expiration date OK
*server certificate activation date OK
*certificate public key: RSA
*certificate version: #3
*subject: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*start date: Wed, 22 Apr 2009 13:02:50 GMT
*expire date: Thu, 22 Apr 2010 13:02:50 GMT
*issuer: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*compression: NULL
*cipher: AES-128-CBC
*MAC: SHA1
> GET /debian/dists/lenny/main/binary-i386/Packages.diff/Index HTTP/1.1
User-Agent: Debian APT-CURL/1.0 (0.7.21)
Host: 
Accept: */*
If-Modified-Since: Tue, 28 Apr 2009 09:38:38 GMT
Cache-Control: max-age=0

* gnutls_handshake() failed: Decryption has failed.
* GnuTLS recv error (-10): The specified session has been invalidated for some 
reason.
* Connection #0 to host  left intact
* Connection #0 seems to be dead!
* Closing connection #0
* About to connect() to  port 443 (#0)
*   Trying ... * connected
* Connected to  () port 443 (#0)
* found 1 certificates in /etc/apt/certs/-cacert.pem
*server certificate verification OK
*common name:  (matched)
*server certificate expiration date OK
*server certificate activation date OK
*certificate public key: RSA
*certificate version: #3
*subject: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*start date: Wed, 22 Apr 2009 13:02:50 GMT
*expire date: Thu, 22 Apr 2010 13:02:50 GMT
*issuer: C=IT,ST=Italy,L=,O= Certification 
Authority,OU=R&D,CN=,EMAIL=rd@.com
*compression:

Bug#525964: apt-transport-https should raise their dependency to libcurl3-gnutls > 7.18.2-8lenny2 to enable ssl key/cert clients authentication

[Marco Amadori]

Package: apt-transport-https
Tags: lenny

Because of bug #525962 [0], apt-transport-https could not fully work with 
libgnutls26  version 2.4.2-6+lenny. This bug affects lenny release of 
libgnutls, on sid everithing works as expected.

Anyway apt-transport-https should depend on libcurl3-gnutls > 7.18.2-8lenny2 
(e.g. >= 7.19.4-1 which depends on libgnutls26 (>= 2.5.9-0) which works).

Commodity link:
[0] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=525962
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#525962: libgnutls26 makes apt-transport-https fail with ssl key/cert client authentication

[Marco Amadori]

Package: libgnutls26
Severity: important
Version: 2.4.2-6+lenny1
Tags: lenny

When using apt-transport-https to fetch packages from a https web server 
configured with required client authentication (apt options Acquire::
{CaInfo,SslCert,SslKey}), an apt-operation will fail with misleading message:

 * 
# apt-get update
[...]
Err https://debian. lenny/main Packages
  Sub-process bzip2 returned an error code (2)
Fetched 140B in 0s (248B/s)
W: GPG error: https://debian. lenny Release:  The following 
signatures were invalid: NODATA 1 NODATA 2
W: Failed to fetch https://debian./debian/dists/lenny/main/binary-
i386/Packages.bz2  Sub-process bzip2 returned an error code (2)

E: Some index files failed to download, they have been ignored, or old ones 
used instead.
 * 

Upgrading to libgnutls26 to sid's 2.6.5-1 fixes the problem.

If I disable client authentication on the web server, leaving just https 
server authentication (via the cacert), it works even with lenny's version.

This bug should be probably mentioned  on apt-transport-https bug entries too 
since the error reported is misleading and since testing curl via command line 
with --cacert, --cert and --key just works (with the same .pem files specified 
in apt.conf*.)

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#514220: Please support "just leaving a file in a /usr/local path" in order to add a custom certificate

[Marco Amadori]

As Peter Palfrader  noted on Thu, 05 Feb 2009 12:11:53 
+0100:

> Ideally there would just be a place where the admin can dump certs, say
> /etc/ca-certificates/local/ or whatever, then this whole mess wouldn't
> be necessary.)

As noted in the FHS, "/usr" should be considered as read only , so I would 
suggest "/usr/local/ca-certificates" as the above mentioned "whaterver".

So I second this bug entry too, but please use a FHS friendy path in fixing 
this it.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#525796: Please ship libpng's -dbg package

[Marco Amadori]

Source: libpng
Severity:  wishlist

As most important libraries do, it would be useful and nice if libpng could 
ships its file with debugging symbols in a separate package.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#525012: ifconfig fails to initialize a virtual interface with a zero IP address

[Marco Amadori]

package: net-tools
severity: important
version: 1.60-23

ifconfig seem to not be able anymore to "deconfigure" a virtual interface.

The Official Manual example [1] in the section 10.6.1.7 does not work anymore 
because of this.

So e.g:
# ifconfig eth0:1 inet 0 up

does not work anymore

As noted in [1] (which is the same bug of this in the ubuntu's bts) I think 
that  dhcp3-client's important bug #275653 is caused by this bug too, so I CC 
it.

[0] http://www.debian.org/doc/manuals/reference/ch-gateway.en.html#s-net-
ifupdown
[1] https://bugs.launchpad.net/ubuntu/+source/net-tools/+bug/356800
-- 
ESC:wq


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#275653: ifconfig fails to initialize a virtual interface with a zero IP address

[Marco Amadori]

package: net-tools
severity: important
version: 1.60-23

ifconfig seem to not be able anymore to "deconfigure" a virtual interface.

The Official Manual example [1] in the section 10.6.1.7 does not work anymore 
because of this.

So e.g:
# ifconfig eth0:1 inet 0 up

does not work anymore

As noted in [1] (which is the same bug of this in the ubuntu's bts) I think 
that  dhcp3-client's important bug #275653 is caused by this bug too, so I CC 
it.

[0] http://www.debian.org/doc/manuals/reference/ch-gateway.en.html#s-net-
ifupdown
[1] https://bugs.launchpad.net/ubuntu/+source/net-tools/+bug/356800
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#522300: mdadm depends on virtual package 'mail-transport-agent' without an alternative

[Marco Amadori]

package: mdadm
version: 2.6.8-12-gb47dff6-2
severity: normal

"the 'mdadm' package recommends the virtual package 'mail-transport-agent' 
without an alternative, causing aptitude to install any random package 
providing 'mail-transport-agent' (citadel-mta in my case) when 
attempting to install the 'mdamd' package.

Could you please change the recommendings to 'exim4 | sendmail | mail-
transport-agent' ?"

Thanks.

P.S. "" text adapted from #474999 to speed things up.
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

Bug#514916: Not really (was Re: It seems to be related to another package)

[Marco Amadori]

On Thursday 12 February 2009, 20:27:08, rpowell wrote:

> I would like to think that two Debian packages could work to fix
> something at the same time.  :)

Sure, mine was just a request to consider if it could be the case of moving 
the bug to another package to track it better.

> However, the problem can be fixed solely in live-initramfs, like so:

Thanks for the patch, I think it could be late for Lenny perhaps :-)

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#514916: It seems to be related to another package

[Marco Amadori]

On Wednesday 11 February 2009, 22:10:25, rpowell wrote:

> Package: live-initramfs

> The following patch makes everything work in my environment, and I
> believe will still work in single-NIC setups:

> ***
> /srv/fai/nfsroot/live/filesystem.dir/usr/share/initramfs-tools/scripts/func
>tionsWed Jan  7 06:14:37 2009 ---
> /srv/fai/broken/initramfs-tools_scripts_functions   Wed Feb 11 12:56:23

Thanks for the patch but it applies to a file which belongs 
to "initramfs-tools", not "live-initramfs" which is an extension (or an hook)   
of the former.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#497561: live-initramfs: mount-call in scripts/init-bottom/udev has wrong syntax

[Marco Amadori]

On Monday 15 December 2008, 02:47:49, Daniel Baumann wrote:

> Marco,
>
> could you comment on this one?

live-initramfs currently does not include util-linux-ng "mount" as the 
inspection of "hooks/live" could show. (You could also inspect the generated 
initrd to have a look). It includes the klibc version of mount and this only 
because initramfs-tools does include it.

The incriminated file in "scripts/init-bottom" does not belong to 
live-initramfs so we cannot do anything it seems to resolve this bug other 
than manually patching the generated initrd, but this seems not our duty and 
it is bad and ugly.

Or I missed the point completetly so please someone gently let us know better 
this issue.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




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

Bug#507477: regression: Integrity Checks does not work anymore

[Marco Amadori]

Package: live-initramfs
Version: 1.154.1-1
Severity: important

Booting with "integrity-check" halts the boot process, probably because the 
default "md5sum" binary in that context, is still the busybox one, which does 
not support the "-c" option.

Moreover the "reboot" command tries to redirect to "shutdown" which could not 
be found:

Those are the last console output lines:
 * ---
Begin: Running /scripts/live-bottom ... md5sum: can't open '-c': No such file 
or directory
WARNING: could not determine runlevel - doing soft reboot
  (it's better to use shutdown instead of reboot from the command line)
shutdown: No such file or directory
 * ---

Please comment on the shutdown thing, so a clean fix could be prepared, for 
the md5sum part I think that just calling /usr/bin/md5sum directly could fix 
it.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#506956: dpkg-cross does not handle links to .pc files

[Marco Amadori]

Package: dpkg-cross
Version: 2.3.3

As subject says, if the original package has a .pc symbolic link, the 
resulting -cross package does not provide it, this could result in a 
unsatisfied compilation dependence.

e.g. :

$ dpkg --contents libpng12-dev_1.2.27-2_mipsel.deb | grep pc
-rw-r--r-- root/root   218 2008-10-04 
13:56 ./usr/lib/pkgconfig/libpng12.pc
lrwxrwxrwx root/root 0 2008-10-04 
13:56 ./usr/lib/pkgconfig/libpng.pc -> libpng12.pc

$ dpkg --contents libpng12-dev-mipsel-cross_1.2.27-2_all.deb | grep pc
-rw-r--r-- root/root   235 2008-11-26 
12:25 ./usr/mipsel-linux-gnu/lib/pkgconfig/libpng12.pc

So after installation of the above package example, the target system will do 
not have the required "/usr/mipsel-linux-gnu/lib/pkgconfig/libpng.pc" file 
(symlink) in place.
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#506591: A central place for stacked filesystems

[Marco Amadori]

On Monday 24 November 2008, 21:44:05, Tong Sun wrote:

> > Try to boot with "showmounts" :-)
> > If I understood what you meant, this bug could be closed.

> but the main wish, ie, a central place for stacked filesystems, still
> holds.

If you try the above mentioned options and there are no regressions, 
the "stacked filesystems" will end all under /live which could be a "central 
place for stacked filesystems" IMHO.

Could this satisfy your request?

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#506591: A central place for stacked filesystems

[Marco Amadori]

On Saturday 22 November 2008, 22:20:18, Tong Sun wrote:

> It seems that I can't access the stacked filesystems individually. I.e.,
> the mount point of each individual stacked filesystems is not accessible
> for me.

Try to boot with "showmounts" :-)

If I understood what you meant, this bug could be closed.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#500093: directfb 1.2 series package

[Marco Amadori]

On Friday 21 November 2008, 11:49:27, Fathi BOUDRA wrote:

> Thanks Marco. I have started to work on the packages too. At least, patches
> are not updated yet. Feel free to send your changes. I'll merge and upload
> a package to experimental this week-end.

I attached the debian directory I used to build the package, it came from 
applying in git the patches present on latest sid packages, one by one, then 
I rebased it to 1.2.6. Some patches when away because they were applied 
upstream.

I had to modify just 1 patch, "93_fix_unicode_key_handling" to use the new 
system of invoking the file descriptor "fd".

And obviously some cosmethic patch for the control file and one line addition 
in libdirectfb-extra list.

I didn't made an analisys to see if the patches still makes sense, just a 
brute force update and merge.

P.S. Do not use the included changelog :-)
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



directfb-1.2.6-debian.tar.bz2
Description: application/tbz

Bug#506229: dpkg-cross fails to provide correct .pc for libpthread-stubs0-dev

[Marco Amadori]

On Friday 21 November 2008, 11:07:38, Neil Williams wrote:

> Package: dpkg-cross
> tag 506229 + pending
> thanks

> That change will be in dpkg-cross 2.3.4:

Thanks for the quick fix, Neil.

Really, really nice.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#506229: dpkg-cross fails to provide correct .pc for libpthread-stubs0-dev

[Marco Amadori]

On Wednesday 19 November 2008, 19:31:31, you wrote:

> At first, I couldn't reproduce the bug but I think I have it isolated
> now - the pkgconfig file is in /usr/share/ instead of the more
> usual /usr/lib/

I thought it used a different pattern matching, but this makes perfect sense.

It could be a bug of libpthread-stubs0-dev ?

> That could be a regression - dpkg-cross doesn't see the package as
> useful which is a bug.
> Please let me know which version of dpkg-cross you are currently using.

Sorry for not adding it in the first report I though I used latest from sid, 
anyway here it is:

$ dpkg-cross --help
dpkg-cross version 2.3.3

$ dpkg -l dpkg-cross | grep ii
ii  dpkg-cross2.3.3 tools for cross compiling Debian 
packages
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#500093: Help offer in packagein 1.2.x

[Marco Amadori]

Hi dear DirectFB maintainers,
I  write this mail because I will like to help in speeding up the packageing 
of new directfb upstream (1.2.6 at this mail time).

As now I just started to use git-import-dsc and git-import-orig to take track 
with git with all modification I will made in order to build this package.

I after applied the quilt patches in different commits to then rebase to the 
merged 1.2.6 tree to ease the generation of new quilt patches for new code.

If something I did could be useful, let me know and I can put this git 
repository somewhere, or tell me which could be your preferred collaboration 
way for me to contribute to this task.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#506229: dpkg-cross fails to provide correct .pc for libpthread-stubs0-dev

[Marco Amadori]

Package: dpkg-cross

Hi dear emdebian fellows, dpkg-cross cannot build a useful 
libpthread-stubs0-dev for cross compiling.

Eg:

$ dpkg -L libpthread-stubs0-dev
/.
/usr
/usr/share
/usr/share/pkgconfig
/usr/share/pkgconfig/pthread-stubs.pc
/usr/share/doc
/usr/share/doc/libpthread-stubs0-dev
/usr/share/doc/libpthread-stubs0-dev/README
/usr/share/doc/libpthread-stubs0-dev/copyright
/usr/share/doc/libpthread-stubs0-dev/changelog.Debian.gz

$ dpkg -L libpthread-stubs0-dev-mipsel-cross
/.
/usr
/usr/share
/usr/share/doc
/usr/share/doc/libpthread-stubs0-dev-mipsel-cross
/usr/share/doc/libpthread-stubs0-dev-mipsel-cross/README

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#504713: New upstream release

[Marco Amadori]

Package: jfsutils
Severity: wishlist

At the main sourceforge page is available [0] a new bugfix release tagged 
1.1.13.

Please package it if you could.

Thanks.

[0] http://jfs.sourceforge.net/project/pub/jfsutils-1.1.13.tar.gz
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#502979: please add RTSP_ALLOW_CLIENT_DESTINATION_SETTING to CFLAGS

[Marco Amadori]

Package: livemedia-utils
Severity: wishlist
Tags: patch

Important use cases for liblivemedia live555MediaServer is testing and stream 
quality analisis, the required option mentioned in the subject just permits 
to have the server stream RTP data to a IP different from the RTSP client 
requesting it, allowing also "push" like technology.

To enable it just add -DRTSP_ALLOW_CLIENT_DESTINATION_SETTING in line 8 of 
debian/rules like:

CFLAGS = -Wall -g -DRTSP_ALLOW_CLIENT_DESTINATION_SETTING

Or add "#define RTSP_ALLOW_CLIENT_DESTINATION_SETTING" in the code.

Thanks.
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#502877: lh_binary_debian-installer fails to include local debs

[Marco Amadori]

Package: live-helper
Version: <= 1.0.1-1
Tags: fixed-in-git

in lh_binary_debian-installer @478:

'if Find_files ../config/binary_local-debs/*_"${LH_ARCHITECTURE}".deb'

where we need not to prefix "../" to reach "config" but just use "./"

This is fixed in my live-helper's git @alioth there:
http://git.debian.org/?p=users/mammadori-guest/live-helper.git;a=summary
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#500672: fixed in git

[Marco Amadori]

This should be fixed in my git repository:

http://git.debian.org/?p=users/mammadori-guest/live-initramfs.git;a=commit;h=b34f6425c48d460bd0d93efd76884045b015692c
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#496619: Agreeement

[Marco Amadori]

seconded
-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#481485: Quick and dirty working bad ugly hack

[Marco Amadori]

Until we wait for a new packages, this works for 2.6.25-2 :

$ cd /usr/src
$ su
# grep  CONFIG_XEN modules/ -Rl | xargs -i 
sed -i -e 's/CONFIG_XEN/CONFIG_LUPO/g' {}
# IGNORE_XEN_PRESENCE=1 m-a -tO a-i nvidia -l 2.6.25-2-686

I know it is really ugly and little tested, but seems to work, at least on my 
system.

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#479516: firmware hints?

[Marco Amadori]

> Version: 2.6.25-1

> > The bnx2 driver was included in the 2.6.22-3 kernel
> > (config-2.6.22-3-amd64:CONFIG_BNX2=m)
> > but not anymore in 2.6.24. Still this device is commonly used in Dell/HP
> > server hardware and so
> > quite usefull.

> please do a quick google research next time before reporting
> useless duplicates. kthx

I did a lot of google searches, this time it was not so straightforward even 
for those who knows what dfsg-nonfree and lkml means :-)

> 2.6.25 in unstable has bnx2 ported to request_firmware()
> you'll need separate firwmare too.

The difficult part is finding the firmware blob right now.

A quick google search gave me just headaches. Thanks for the update to  
request_firmware() to Bastian, but a lot of users, could be misoriented by 
the difficulty to make this driver work on sid yet. *

A gentle question on IRC at #debian-kernel served me better, there is some 
code in the source package firmware-nonfree so, (*) a package 
named "firmware-bnx" will be uploaded somewhere in the future to unstable.

Meanwhile it can be found on

# svn co svn://svn.debian.org/kernel/dists/trunk/firmware-nonfree

-- 
ESC:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.




-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#474090: New upstream release

[Marco Amadori]

Package: sdcc
Version: 2.7.0-1
Severity: wishlist

A new upstream (2.8.0) has been released, so this wishlist bug to inform the 
maintainer about that.

--
ESC:wq



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#461877: A new upstream is available with nice feature.

[Marco Amadori]

Package: python-pexpect
Severity: wishlist

On upstream site, http://pexpect.sourceforge.net/, pexpect 2.3 is available. 

It supports to "spawn" directly on file descriptor "int"s, and this is a 
really nice feature to have in debian, to drive directly e.g. serial lines.

-- 
ESC:wq



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#457177: [Yaird-devel] Bug#457177: Bug#457177: Bug#457177: keep yaird out of Testing

[Marco Amadori]

Alle sabato 22 dicembre 2007, Sven Luther ha scritto:

> > Unfortunatly, I have to agree from a release team POV (i.e. speaking
> > with my Release Manager hat on) with maks on the general status of the
> > package, especially as maks spoke with his kernel arch maintainer hat
> > on (so his remarks shouldn't be lightly waived away).
>
> Andreas, face it, Max speaks as a strong supporter of initramfs-tools,
> the concurent of yaird, and has shown real antagonism and agressivity
> toward yaird since the begining of the initrd->initramfs migration.
> [..others  personal issues..] 

False, I remember those days too, Max was only techically concerned about the 
migration. Nothing personal.

Keep the discussion technical please.

Yaird is nice to have in sid, I personally contributed 2 patches upstream, 
evms and dmraid (the latter unmerged), but for our users is better to keep it 
out from lenny, at least until those bugs could be fixed.

-- 
ESC:wq


signature.asc
Description: This is a digitally signed message part.

Bug#453511: Please provide "MPEG2TransportStreamIndexer" binary

[Marco Amadori]

package: livemedia-utils
severity: wishlist

The subject mentioned "MPEG2TransportStreamIndexer" binary is available in 
the "-dev" package of this suite as compressed source code, it would be nice 
to have built in this package.

Rationale: since it is mandatory to enable "tricky play" RTSP streaming server 
operations like seek, fastforward and rewind, it could be very useful to have 
it on a (streaming) production environment without manually build it.

-- 
ESC:wq


signature.asc
Description: This is a digitally signed message part.

Bug#453449: New upstream available

[Marco Amadori]

package: livemedia-utils
severity: wishlist

Upstream released a lot of new releases of the software since debian's 
packaging, some of them are important bugfixes.

Could you please update the package?

-- 
ESC:wq


signature.asc
Description: This is a digitally signed message part.

Bug#444077: libftdi: does not build API docs

[Marco Amadori]

Package: libftdi
Severity: wishlist
Tags: patch

I saw that the package does not build dep on doxygen and it does not build the 
API html docs, could this behaviour be reversed?

A stupid patch for that is included here below:


$ diff -urNa libftdi-0.10 libftdi-0.10.new
diff -urNa libftdi-0.10/debian/control libftdi-0.10.new/debian/control
--- libftdi-0.10/debian/control 2007-09-26 00:04:52.0 +0200
+++ libftdi-0.10.new/debian/control 2007-09-25 23:58:44.829165079 +0200
@@ -2,7 +2,7 @@
 Section: libs
 Priority: optional
 Maintainer: Aurelien Jarno <[EMAIL PROTECTED]>
-Build-Depends: debhelper (>= 4.0.0), autotools-dev, libtool, automake1.4, 
autoconf, libusb-dev (>= 2:0.1.11), pkg-config (>= 0.18)
+Build-Depends: debhelper (>= 4.0.0), autotools-dev, libtool, automake1.4, 
autoconf, libusb-dev (>= 2:0.1.11), pkg-config (>= 0.18), doxygen
 Standards-Version: 3.7.2

 Package: libftdi0
diff -urNa libftdi-0.10/debian/libftdi-dev.docs 
libftdi-0.10.new/debian/libftdi-dev.docs
--- libftdi-0.10/debian/libftdi-dev.docs1970-01-01 01:00:00.0 
+0100
+++ libftdi-0.10.new/debian/libftdi-dev.docs2007-09-26 00:02:51.343213114 
+0200
@@ -0,0 +1 @@
+doc/html

-- 
ESC:wq


libftdi-0.10_addedd-docs-build.diff.bz2
Description: BZip2 compressed data


signature.asc
Description: This is a digitally signed message part.

Bug#403405: 2.6.19 seems to fix it

[Marco Amadori]

This seems to be fixed "upstream", along with other issues regarding ieee1394 
on linux 2.6.19, according to [0]:

-
kernel 2.6.19 release notes
From: Stefan Richter:
 Date: 2006-12-03 3:00 am
eth1394: memory alignment problem fixed
ieee1394: recursive locking fixed
ohci1394: initializes before most other device drivers to aid kernel debugging 
with firescope
raw1394: kernel freeze in address range mapping functions fixed
-

Since I also experimented a kernel panic because of this bug, I think that the 
proper severity should be re raised, since it surely "breaks the system up".

Could this code merged on 2.6.18 or the whole 2.6.19 shall be let entering 
unstable and then etch ? 
(And since bug# are still high, maybe the shortcoming 2.6.20 could be 
considered for future inclusion in unstable too... joking!)

[0] http://www.linux1394.org
-- 
ESC:wq


pgp1DgUyPvUO2.pgp
Description: PGP signature

Bug#399834: debmirror: Debmirror does not delete their mktemp files in /tmp

[Marco Amadori]

Package: debmirror
Version: 20060907.1
Severity: normal


As subject says, debmirror leaves in my /tmp a lot of temporary files,
that it should remove when it finishes using them.

For example, invoking debmirror like this:
"debmirror -arch=i386,amd64 \
-section=main,contrib,non-free,main/debian-installer --method=rsync \
--root=:debian --host=ftp.de.debian.org --dist=sid,etch --getcontents \
/srv/mirror/debian"

Creates in /tmp a lot of files used as include-list for rsync, but it
forgots to delete them:

$ ls /tmp
1wDeRv2qdT  Cpvhyq0ej1  
Svy6K1AfYh  x8M3IdYZ41
3aW3nVdFeR  eAvyVsYfS2  kMslFBF2Wq
T3oZmNmW2P  xlZnqCHNw9
52HNUwrrFY  eKZSOLFMfi  
TJDFWZOYao  xOlSxi0xmI
6aYqeH6CUa  FNK6A9kyCn  lXcb2BReya
TsvHLXAtcj  z3V9uCsPRR
6NnR5xaHWk  G5ltITFrgl  OjakOd6T5c
UxSjDBRIOj  Z5ki6CIfPD
8myt3sbBdC  gpg-U0xnqy  olYSGAK4CI
z5ldCOlMrk  9wI0X45oqa  
VXl5NlsNHn  zkSAQzHVGn
AkBpZDTsuA  gs9AnVE3qq  rDZLnpxX3j
w8HwqhPMkr  ZQS9owwpdU
aV0L0N0pSY  HO7TLMVX53  
wFUISgzG5h  ZwEZFm9OUr
AVaNGhi9yb  JFeGORiPL0  RJmokVbjQK
WNhaFD7sSS
BIuC9ato3F  JuXVCt7q0z  RtZk5k53Pk
wRAVnkl535
Ch2L3wHAIX  
Wu15hyD8Xk

-- System Information:
Debian Release: 4.0
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.18-2-amd64
Locale: LANG=it_IT.UTF-8, LC_CTYPE=it_IT.UTF-8 (charmap=UTF-8)

Versions of packages debmirror depends on:
ii  bzip2 1.0.3-6high-quality block-sorting file co
ii  libcompress-zlib-perl 1.42-1 Perl module for creation and manip
ii  libdigest-sha1-perl   2.11-1 NIST SHA-1 message digest algorith
ii  liblockfile-simple-perl   0.2.5-7Simple advisory file locking
ii  libwww-perl   5.805-1WWW client/server library for Perl
ii  perl [libdigest-md5-perl] 5.8.8-6.1  Larry Wall's Practical Extraction 
ii  perl-modules [libnet-perl]5.8.8-6.1  Core Perl modules
ii  rsync 2.6.9-2fast remote file copy program (lik

Versions of packages debmirror recommends:
ii  gnupg 1.4.5-2GNU privacy guard - a free PGP rep
ii  patch 2.5.9-4Apply a diff file to an original

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#397320: casper: allow commandline-device to show up late

[Marco Amadori]

Thanks for the patch, included in the next upload, by the way, busybox seems 
to do not like "seq".

See you soon :-)

-- 
ESC:wq


pgpXmRnIoftqW.pgp
Description: PGP signature

Bug#394686: [PATCH] fix casper-getty setup

[Marco Amadori]

Alle 15:41, domenica 22 ottobre 2006, Alex Owen ha scritto:

> The attached patch fixes the casper-getty setup.
> There were reports that it did not work on cdrom boots. I don't think
> the code in 1.71+debian1 could have worked at all net or cd boot. I
> have tested the new code by booting a cdrom image using qemu. This
> test shows the code works.

A similar patch was already lying in bzr repository, thanks anyway for that, 
will be fixed in the next upload.

P.S. Since the repository is public, as a suggestion, look in here before 
sending patches to avoid duplicated efforts.
-- 
ESC:wq


pgp8EuybtDJUy.pgp
Description: PGP signature

Bug#394308: casper: Wait for usb device to be detected

[Marco Amadori]

Alle 18:17, venerdì 20 ottobre 2006, Bas Wijnen ha scritto:

> When booting from a usb device, some time is needed before the device is
> actually visible for the system.  Casper seems to solve this by trying to
> detect the device 20 times.  However, if there is no slow part in the
> detection (which happens if there is no other device to mount, for
> example), those 20 detections are done real fast, and all finished before
> the device shows up.  The attached patch adds a "sleep 2" in the loop, so
> it takes at least 40 seconds.  That should be enough for everyone. ;-)

This problem comes from a bad reorganization of code I did in the past when 
adding netboot support, a "sleep 1" were here but I moved it where it makes 
no sense, sorry about that.

> The patch also fixes a bug which led to non-wrapped systems being unable to
> boot (that is, the test would always fail).  [ A || B ] is interpreted by
> bash as two commands, the second of which will only be executed if the
> first fails. In this case, it always will, because the [ does not have a
> matching ].  To do OR in test, -o should be used.  I don't think anyone
> would encounter this though, due to the rareness of non-wrapped live
> systems.

Thanks for this insight, that part of code was a bad hack I need to polish 
soon anyway, your patch is included in bzr repo and the next upload will 
close this bug too.

Thanks for your time and support.

-- 
ESC:wq


pgpqe71w9n3bT.pgp
Description: PGP signature

Bug#394554: casper: Boot from non-CD non-usb devices

[Marco Amadori]

Alle 23:01, sabato 21 ottobre 2006, Bas Wijnen ha scritto:

> The current code checks for CD or usb, and only boots from those devices. 
> I need to boot a removable IDE drive (cold-pluggable), which didn't work
> due to explicit checks.  The attached patch removes them, and thus allows
> booting from any medium.

I already did this in bzr in a different way, anyway really thanks for this 
patch and for feedback.

This issue will be fixed in the next upload.

-- 
ESC:wq


pgpcgEu8bMBPL.pgp
Description: PGP signature

Bug#393573: Fixed upstream

[Marco Amadori]

As reported here [0] this bug seems to be fixed upstream in the new beta 
driver 1.0-9625 [1] and 1.0-9626. 

So the driver in experimental could migrate to unstable, and by the way I'm 
using that driver since it hit experimental without issues on AMD64 (and 
compiz works too).

[0] http://www.nvnews.net/vbulletin/showthread.php?t=78322
[1] http://www.nzone.com/object/nzone_downloads_rel70betadriver.html.

-- 
ESC:wq


pgpq1yxz2Hxn5.pgp
Description: PGP signature

Bug#392799: Having the need to have "/proc" mounted in postinst halts installations in chroots, hence in debian-live.

[Marco Amadori]

Alle 17:41, sabato 14 ottobre 2006, Tollef Fog Heen ha scritto:
> > [.. ] asking to have
> > /proc installed have only meanings of "current system" installs, not a
> > generic one. So this bug is valid for all packages that requires /proc to
> > finish the installation.

> Then you should build in a UML or Xen instance or something similar
> which is as close to a generic system as possible.

This is a nice advice and will permits to spot "bug" similar to that one, 
thanks I will look at this issue after etch.

> > If I dig maintainers scripts and provide a patch, could that be
> > integrated asap or will just be a useless effort by me?

> I don't think you can, given that the way it currently works is by
> design. 

This sentence is exactly clear to me, that means that such a patch will not be 
considered?

> If we can't determine whether port 80 is in use or not, we 
> assume it is.

So if /proc is not mounted you could just assume that port 80 is free and do 
not fail the installation of package will be possible?

-- 
ESC:wq


pgpdymwpOoJzO.pgp
Description: PGP signature

Bug#392799: Having the need to have "/proc" mounted in postinst halts installations in chroots, hence in debian-live.

[Marco Amadori]

Alle 21:54, venerdì 13 ottobre 2006, Tollef Fog Heen ha scritto:
> Seriously, there is no reason not to mount /proc
> in any and all chroots.

Ok, I resign this "battle".

But I think that it is not a clean way to proceed in a "create live 
environment" sense of a package, /proc filesystem means the "current hardware 
and kernel system" pack of informations and not a "generic debian system" 
informations nor an "hardwareless" or "kernelless" pack of information, so 
even if apache2 team wont fix this bug, asking to have /proc installed have 
only meanings of "current system" installs, not a generic one. So this bug is 
valid for all packages that requires /proc to finish the installation.

This issue is not blocking debian-live, I admit, but is blocking a "clean and 
nice" way to do system administration, and since the fix could be easy, why 
not provide it?

If I dig maintainers scripts and provide a patch, could that be integrated 
asap or will just be a useless effort by me?

-- 
ESC:wq


pgpNCcmeL77DI.pgp
Description: PGP signature

Bug#392799: Having the need to have "/proc" mounted in postinst halts installations in chroots, hence in debian-live.

[Marco Amadori]

> I can tell 
> you that it's not rocket science to get /proc mounted in your target
> chroot while installing packages and then, y'know, unmount it when
> you're done.

In fact we did an hack for having /proc mounted in debian-live, but the bug 
regards also installing apache in chroots without /proc mounted, and I cannot 
see why having /proc mounted should be a requirements, others package 
maintainers have patched issues like this to have their packages building in 
automated building tools like pbuilder or similar, I think this could be an 
easy to add feature.

> If you're concerned about the build script dying and leaving things
> mounted in the chroot, I've heard talk of shell functions like "trap"...
> I'm sure that if a moron like me can figure out how it works, a
> sufficiently clever person can probably do so as well.

Thanks for the suggestion, we planned our hack in a similar manner to avoid 
unmountable proc.

But I think this bug instead of closed should be put on whislist at least, 
since it is not resolved but just put in a "I do not agree on that" form.

-- 
ESC:wq


pgp0fWImy71Ay.pgp
Description: PGP signature

Bug#392800: Having the need to have "/proc" mounted in postinst halts installations in chroots, hence in debian-live.

[Marco Amadori]

Package: amavis

Hello,
as subject says, in postinst the looking for presency of mounted "/proc" could 
be a nice improvements for this package, otherwise could not be installed in 
chroots without /proc mounted, and since debian-live should not mount /proc 
because could leave with a chroot where proc could not be umounted, here it 
is why I filed this bug.

Furthermore this is a nice package that should be available in standard 
debian-live images, and etch is approaching :-)

Thanks for your patience, best regards.
-- 
ESC:wq


pgppQfcP32XrE.pgp
Description: PGP signature

Bug#392799: Having the need to have "/proc" mounted in postinst halts installations in chroots, hence in debian-live.

[Marco Amadori]

Package: apache2

Hello,
as subject says, in postinst the looking for presency of mounted "/proc" could 
be a nice improvements for this package, otherwise could not be installed in 
chroots without /proc mounted, and since debian-live should not mount /proc 
because could leave with a chroot where proc could not be umounted, here it 
is why I filed this bug.

Furthermore this is a nice package that should be available in standard 
debian-live images, and etch is approaching :-)

Thanks for your patience, best regards.
-- 
ESC:wq


pgp1LNJlNi6kb.pgp
Description: PGP signature

Bug#392798: Having the need to have "/proc" mounted in postinst halts installations in chroots, hence in debian-live.

[Marco Amadori]

Package: console-common

Hello,
as subject says, in postinst the looking for presency of mounted "/proc" could 
be a nice improvements for this package, otherwise could not be installed in 
chroots without /proc mounted, and since debian-live should not mount /proc 
because could leave with a chroot where proc could not be umounted, here it 
is why I filed this bug.

Furthermore this is a basic package that should be available in standard 
debian-live images to correctly configure console keyboards, and etch is 
approaching :-)

Thanks for your patience, best regards.
-- 
ESC:wq


pgpAenyw5MVtf.pgp
Description: PGP signature

Bug#389822: previous "patch" is needed but does not fix the bug

[Marco Amadori]

Just as hint:

I saw two suspicious "while" in postinst at "validate_string_db_input" and 
at "validate_monitor_frequency_db_input".


-- 
ESC:wq


pgp4M5eRQ3LQa.pgp
Description: PGP signature

Bug#390078: (no subject)

[Marco Amadori]

reassign xserver-xorg
merge 389822
-- 
ESC:wq


pgpTFhglE0rwn.pgp
Description: PGP signature

Bug#389822: something like this should fix it

[Marco Amadori]

at line 1905 of /var/lib/dpkg/info/xserver-xorg.config try to patch in a 
similar way:

if ! db_get debian-installer/keymap; then
if [ -z "$first_time_keymap" ]; then
  debug_report_status "db_get debian-installer/keymap"
else
  first_time_keymap="set"
  DI_KEYMAP="us"
fi
  else
DI_KEYMAP="${RET##mac-usb-}"
DI_KEYMAP="${DI_KEYMAP##-latin1}"
  fi

This bug is simply reproducible by installing xorg in a fresh chroot (and it 
affects make-live so we noticed it).

-- 
ESC:wq


pgpTlo4qAJyQt.pgp
Description: PGP signature

Bug#106399: grub-net

[Marco Amadori]

On Tue, 8 Jun 2004 15:23:59 +0200 Frédéric said :

> I'm often use GRUB's net boot images for installating Debian on cards
> without CD nor floppies, and after multiple (re)builds (on Woody), I've
> patched the GRUB Debian package to add a separate package (grub-net)
> which contains only net boot images. Due to possible conflicts between
> network drivers (I've experienced some), I build one boot image for each
> driver, one in Etherboot format, another for PXE...  

Yes please give us that patches, maybe against latest unstable packages, could 
be really handy and helpful to have here on the BTS.

-- 
ESC:wq


pgp2MflJWATwB.pgp
Description: PGP signature

Bug#348761: That could be really useful for Debian-Live

[Marco Amadori]

Hi, I'm one of the developers of the Debian-Live [0] team, and I saw that this 
bug has no reply from the maintainer so far.

Since debian live uses syslinux to boot, the gfxboot patches could give, as in 
latest ubuntu live cds, the ease for unexperienced users to select and choose 
proper language, locales and keyboard from the very beginning of the boot 
process.

Feedback would be really appreciated.

[0] http://live.debian.net
-- 
ESC:wq


pgp0ojfEwiSp6.pgp
Description: PGP signature

Bug#376305: Temporary workaround

[Marco Amadori]

As temporary workaround, until linux-kbuild exit NEW queue and hits sid, 
someone could leave below an url of a package to dl?

-- 
ESC:wq


pgphdQX8SKyS2.pgp
Description: PGP signature

Bug#376538: zope-psycopgda: Please add support (and "depends:") for zope 2.9.

[Marco Amadori]

Package: zope-psycopgda
Version: 1.1.21-7
Severity: minor


I saw that zope-psycopgda needs zope2.8 or zope2.7, but it should work
also for zope 2.9 although dzhanlde will not show it for 2.9 instances.
It is possible to have this product also for version 2.9 of zope?

Thanks.

-- System Information:
Debian Release: testing/unstable
  APT prefers unstable
  APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.16-2-amd64-k8-smp
Locale: LANG=it_IT.UTF-8, LC_CTYPE=it_IT.UTF-8 (charmap=UTF-8)

Versions of packages zope-psycopgda depends on:
ii  python-psycopg1.1.21-7   Python module for PostgreSQL [dumm
ii  zope-common   0.5.24 common settings and scripts for zo
ii  zope2.8   2.8.7-1Open Source Web Application Server

zope-psycopgda recommends no packages.

-- no debconf information


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Bug#371083: Package description warn about a false problem, fixed long ago.

[Marco Amadori]

Alle 19:54, mercoledì 7 giugno 2006, hai scritto:

> On Wed, Jun 07, 2006 at 07:00:33PM +0200, Marco Amadori wrote:
> > A simple use could provide the proof, it is what I did before filing this
> > bug, create some files with different uid/gid, generate an ext2 file,
> > loop mount and look the contents.
>
> Confirmed, thanks very much. The description will be fixed along with my
> next upload.

Thanks to you, good and fast job!

-- 
ESC:wq


pgpBaX613dYfg.pgp
Description: PGP signature

Bug#371083: Package description warn about a false problem, fixed long ago.

[Marco Amadori]

Alle 18:39, mercoledì 7 giugno 2006, Franz Pletz ha scritto:
> On Wed, Jun 07, 2006 at 11:02:43AM +0200, Marco Amadori wrote:
> > The package description warns about:
> >
> > "Warning! `genext2fs' has been designed for embedded systems. As
> >  such, it will generate a filesystem for single-user usage: all
> >  files/directories/etc. will belong to UID/GID 0."
> >
> > This is no longer the case, so this writing should be removed, tnx a lot.
>
> Thanks for your report. I did some quick investigations but wasn't able
> to find any proof of this neither on the mailing list nor in the NEWS
> file. Can you proof this somehow? (CVS commit, mailing list post etc.)

A simple use could provide the proof, it is what I did before filing this bug, 
create some files with different uid/gid, generate an ext2 file, loop mount 
and look the contents.

-- 
ESC:wq


pgpgrh89onHll.pgp
Description: PGP signature

Bug#371083: Package description warn about a false problem, fixed long ago.

[Marco Amadori]

Package: genext2fs
Version: 1.4-rc1-1
Severity: minor

The package description warns about:

"Warning! `genext2fs' has been designed for embedded systems. As
 such, it will generate a filesystem for single-user usage: all
 files/directories/etc. will belong to UID/GID 0."

This is no longer the case, so this writing should be removed, tnx a lot.
-- 
ESC:wq


pgpULo87Bqig4.pgp
Description: PGP signature