Bug#486558: installation-reports: strange (null) in URL for tasksel for automatic install

[Frans Pop]

It still isn't working. I now get an immediate "bad mirror". db_get on
mirror/http/mirror is empty, so mirror_root is still null.

Below a debconf trace with the new code (no default) and two debug log
statements added by me. Boot in "auto" mode.

It clearly shows mirror/http/mirror getting its variable 'mirrors' set,
and just as clearly shows the following db_get not returning anything.

At the very beginning the same happens for mirror/protocol and that does
work, but that still has a default so we just see the preset default being
returned.

Guess this makes the earlier cloned bug to debconf more severe.

Jun 19 09:46:52 debconf: --> SUBST mirror/protocol protocols http, ftp
Jun 19 09:46:52 debconf: Adding [protocols] -> [http, ftp]
Jun 19 09:46:52 debconf: rfc822db_question_set(db,q=mirror/protocol,q=0x8077058)
Jun 19 09:46:52 debconf: <-- 0
Jun 19 09:46:52 debconf: --> INPUT medium mirror/protocol
Jun 19 09:46:52 debconf: Comparing priorities medium (1) with critical (3)
Jun 19 09:46:52 debconf: <-- 30 question skipped
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> GET mirror/protocol
Jun 19 09:46:52 debconf: <-- 0 http
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> GET mirror/country
Jun 19 09:46:52 debconf: <-- 0 
Jun 19 09:46:52 debconf: --> GET debian-installer/country
Jun 19 09:46:52 debconf: <-- 0 US
Jun 19 09:46:52 debconf: --> SET mirror/country US
Jun 19 09:46:52 debconf: rfc822db_question_set(db,q=mirror/country,q=0x80528a0)
Jun 19 09:46:52 debconf: <-- 0 value set
Jun 19 09:46:52 debconf: --> SET mirror/http/countries US
Jun 19 09:46:52 debconf: 
rfc822db_question_set(db,q=mirror/http/countries,q=0x8052300)
Jun 19 09:46:52 debconf: <-- 0 value set
Jun 19 09:46:52 debconf: --> FGET mirror/country seen
Jun 19 09:46:52 debconf: <-- 0 false
Jun 19 09:46:52 debconf: --> FSET mirror/http/countries seen false
Jun 19 09:46:52 debconf: <-- 0 false
Jun 19 09:46:52 debconf: --> INPUT high mirror/http/countries
Jun 19 09:46:52 debconf: Comparing priorities high (2) with critical (3)
Jun 19 09:46:52 debconf: <-- 30 question skipped
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> GET mirror/http/countries
Jun 19 09:46:52 debconf: <-- 0 US
Jun 19 09:46:52 debconf: --> SET mirror/country US
Jun 19 09:46:52 debconf: rfc822db_question_set(db,q=mirror/country,q=0x80528a0)
Jun 19 09:46:52 debconf: <-- 0 value set
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> GET mirror/country
Jun 19 09:46:52 debconf: <-- 0 US
Jun 19 09:46:52 debconf: --> SUBST mirror/http/mirror mirrors 
ftp.us.debian.org, ftp.egr.msu.edu, 
mirrors.kernel.org, debian.lcs.mit.edu, debian.osuosl.org, 
debian.crosslink.net, ftp.gtlib.gatech.edu, 
distro.ibiblio.org, ftp-mirror.internap.com, ftp.c
Jun 19 09:46:52 debconf: Adding [mirrors] -> [ftp.us.debian.org, 
ftp.egr.msu.edu, mirrors.kernel.org, 
debian.lcs.mit.edu, debian.osuosl.org, debian.crosslink.net, 
ftp.gtlib.gatech.edu, distro.ibiblio.org, 
ftp-mirror.internap.com, ftp.cerias.purdue.edu
Jun 19 09:46:52 debconf: 
rfc822db_question_set(db,q=mirror/http/mirror,q=0x8076ed8)
Jun 19 09:46:52 debconf: <-- 0
Jun 19 09:46:52 debconf: --> INPUT high mirror/http/mirror
Jun 19 09:46:52 debconf: Comparing priorities high (2) with critical (3)
Jun 19 09:46:52 debconf: <-- 30 question skipped
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> INPUT high mirror/http/proxy
Jun 19 09:46:52 debconf: Comparing priorities high (2) with critical (3)
Jun 19 09:46:52 debconf: <-- 30 question skipped
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> GET mirror/http/proxy
Jun 19 09:46:52 debconf: <-- 0 
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:52 debconf: <-- 0 ok
Jun 19 09:46:52 debconf: --> GET mirror/http/mirror
Jun 19 09:46:52 debconf: <-- 0 
Jun 19 09:46:52 choose-mirror[2582]: INFO: mirror: 
Jun 19 09:46:52 debconf: --> SET mirror/http/hostname
Jun 19 09:46:52 debconf: 
rfc822db_question_set(db,q=mirror/http/hostname,q=0x80523e0)
Jun 19 09:46:52 debconf: <-- 0 value set
Jun 19 09:46:52 choose-mirror[2582]: INFO: root: (null)
Jun 19 09:46:52 debconf: --> INPUT critical mirror/bad
Jun 19 09:46:52 debconf: <-- 0 question will be asked
Jun 19 09:46:52 debconf: --> GO
Jun 19 09:46:54 debconf: <-- 30 backup
Jun 19 09:46:54 debconf: --> GET debconf/priority
Jun 19 09:46:54 debconf: <-- 0 critical


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

Bug#486892: different behavior from debconf when noninteractively asking a select question with a default not in the select list

[Frans Pop]

This bug is even more severe than it first looked. The Choices list is
completely ignored by cdebconf when a db_get is done immediately after a
db_subst that sets a variable used in Choices (or with a db_input in
between that gets skipped because of priority).

From a trace in #486558:

debconf: --> SUBST mirror/http/mirror mirrors ftp.us.debian.org, 
ftp.egr.msu.edu, [...]
debconf: Adding [mirrors] -> [ftp.us.debian.org, ftp.egr.msu.edu, 
mirrors.kernel.org, [...]
debconf: rfc822db_question_set(db,q=mirror/http/mirror,q=0x8076ed8)
debconf: <-- 0
debconf: --> INPUT high mirror/http/mirror
debconf: Comparing priorities high (2) with critical (3)
debconf: <-- 30 question skipped
debconf: --> GO
debconf: <-- 0 ok
[...]
debconf: --> GET mirror/http/mirror
debconf: <-- 0

I tried if displaying the question but not confirming an answer (by
backing up when the question is displayed) helped, but that does not
change anything.

From template.dat:
Name: mirror/http/mirror
Type: select
Choices: ${mirrors}
Description: [...]

From questions.dat:
Name: mirror/http/mirror
Template: mirror/http/mirror
Owners: d-i
Variables:
 mirrors = ftp.us.debian.org, [...]



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

[patch] Add support for KuroBox Pro to oldsys-preseed

[Per Andersson]

I added support for KuroBox Pro to oldsys-preseed.

The patch is tested and works. (I'll submit tests/arm
later when I have created them.)

The default IP address is 192.168.11.150, there is no
default gateway and ns but I set it to 192.168.11.1
because that is what I would have expected by default.


-- Per

Index: oldsys-preseed
===
--- oldsys-preseed  (revision 53630)
+++ oldsys-preseed  (working copy)
@@ -174,6 +174,43 @@
fi
umount $path/sda1 || true
rmdir $path/sda1 $path || true
+   elif echo "$machine" | grep -q "^Buffalo/Revogear Kurobox Pro"; 
then
+   check_file /proc/mtd
+   rootfs=$(get_mtdblock "rootfs")
+   if [ -z "$rootfs" ]; then
+   log "Can't find rootfs MTD partition"
+   exit
+   fi
+   path=/tmp/oldsys-preseed
+   mkdir -p $path/rootfs
+   mount -t jffs2 -o ro /dev/mtdblock2 $path/rootfs || true
+   INTERFACE=eth0
+   parse_unix_tree "$path/rootfs"
+   netinfo=$path/rootfs/etc/netinfo
+   if [ -e "$netinfo" ]; then
+   usage=$(cat "$netinfo" | grep "^my_ipaddress" | 
sed 's/.*=//')
+   if [ "$usage" = "dhcp" ]; then
+   NET_CONFIG=dhcp
+   else
+   NET_CONFIG=static
+   IP_ADDRESS=$(cat "$netinfo" | grep 
"^my_ipaddress" | sed 's/.*=//')
+   NETMASK=$(cat "$netinfo" | grep 
"^my_subnetmask" | sed 's/.*=//')
+   GATEWAY=$(cat "$netinfo" | grep 
"^my_dgw" | sed 's/.*=//')
+   fi
+   fi
+   hostinfo=$path/rootfs/etc/host.info
+   if [ -e "$hostinfo" ]; then
+   HOSTNAME=$(cat "$hostinfo" | grep "^hostname" | 
sed 's/.*=//')
+   fi
+   if [ "$NET_CONFIG" != "static" ]; then
+   IPADDRESS=192.168.11.150
+   NETMASK=255.255.255.0
+   GATEWAY=192.168.11.1
+   [ -z "$NAMESERVERS" ] && 
NAMESERVERS=192.168.11.1
+   dhcp_fallback $FILE
+   fi
+   umount $path/rootfs
+   rmdir $path/rootfs $path
else
exit_unknown
fi


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

Make syslinux beep?

[Samuel Thibault]

Hello,

It is a bit hard for blind people to know exactly when they can type
boot options for the installer.  It would be helpful that syslinux beeps
when it is ready to get input.  What do people think about it?

Samuel


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

Re: Make syslinux beep?

[Jan Buchal]

> "ST" == Samuel Thibault <[EMAIL PROTECTED]> writes:

ST> Hello, It is a bit hard for blind people to know exactly when
ST> they can type boot options for the installer. It would be
ST> helpful that syslinux beeps when it is ready to get input. What
ST> do people think about it?
Sure, it is easy good solution.

Best
-- 

Jan Buchal
Tel: (00420) 24 24 86 008
Mob: (00420) 608023021


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

cat "$infofile" | grep "^string" | sed 's/.*=//'

[Geert Stappers]

> + if [ "$usage" = "dhcp" ]; then
> + NET_CONFIG=dhcp
> + else
> + NET_CONFIG=static
> + IP_ADDRESS=$(cat "$netinfo" | grep "^my_ipaddress" | 
> sed 's/.*=//')
> + NETMASK=$(cat "$netinfo" | grep "^my_subnetmask" | sed 
> 's/.*=//')
> + GATEWAY=$(cat "$netinfo" | grep "^my_dgw" | sed 
> 's/.*=//')
> + fi
> + fi
> + hostinfo=$path/rootfs/etc/host.info
> + if [ -e "$hostinfo" ]; then
> + HOSTNAME=$(cat "$hostinfo" | grep "^hostname" | sed 's/.*=//')
> + fi


   cat "$infofile" | grep "^string" | sed 's/.*=//'

can be rewritten as

   grep "^string" "$infofile" | sed 's/.*=//'


with the advantage of 'cat' not being spawned.



Further rewrite to

   sed -n '/^string/s/.*=//' "$infofile" 

even avoids 'grep' being started.



Cheers
Geert Stappers


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

Re: cat "$infofile" | grep "^string" | sed 's/.*=//'

[Geert Stappers]

Op 19-06-2008 om 13:47 schreef Geert Stappers:
> > +   if [ "$usage" = "dhcp" ]; then
> > +   NET_CONFIG=dhcp
> > +   else
> > +   NET_CONFIG=static
> > +   IP_ADDRESS=$(cat "$netinfo" | grep "^my_ipaddress" | 
> > sed 's/.*=//')
> > +   NETMASK=$(cat "$netinfo" | grep "^my_subnetmask" | sed 
> > 's/.*=//')
> > +   GATEWAY=$(cat "$netinfo" | grep "^my_dgw" | sed 
> > 's/.*=//')
> > +   fi
> > +   fi
> > +   hostinfo=$path/rootfs/etc/host.info
> > +   if [ -e "$hostinfo" ]; then
> > +   HOSTNAME=$(cat "$hostinfo" | grep "^hostname" | sed 's/.*=//')
> > +   fi
> 
> 
>cat "$infofile" | grep "^string" | sed 's/.*=//'
> 
> can be rewritten as
> 
>grep "^string" "$infofile" | sed 's/.*=//'
> 
> 
> with the advantage of 'cat' not being spawned.
> 
> 
> 
> Further rewrite to
> 
>sed -n '/^string/s/.*=//' "$infofile" 
Oops, forgot to print
 sed -n '/^string/s/.*=//p' "$infofile" 
> 
> even avoids 'grep' being started.
> 

I don't known how fast (or slow) a KuroBox is.

On a 350MHz Pentium III did I run a speed test

Original version:
real0m8.269s
user0m3.256s
sys 0m4.972s

Without cat:
real0m5.923s
user0m2.160s
sys 0m3.760s

Only sed:
real0m2.692s
user0m1.084s
sys 0m1.608s



Attached the test script

Cheers
Geert Stappers
# execute with sh
infofile=infofile.txt
cat << HERE > $infofile
some line
another line
string=value
more strings
HERE

time for i in $(seq 500)
do
  VAR=$( cat $infofile | grep "^string" | sed 's/.*=//')
done
echo $VAR

time for i in $(seq 500)
do
  VAR=$( grep "^string" $infofile | sed 's/.*=//')
done
echo $VAR

time for i in $(seq 500)
do
  VAR=$( sed -n '/^string/s/.*=//p' $infofile )
done
echo $VAR

Bug#486926: Installed system does not have correct permissions on directories

[Chris Lamb]

Joey Hess wrote:

> > something seems to block the filename output from tar, causing the
> > progress bar to sit at 2% until very near the end - it then shoots up
> > quite quickly.  
> 
> Right, I guess I'd need to use a read loop to get the lines one at a
> time rather than having it buffer. Updated patch attached.

That fixes that problem, but now the read command started by "db_progress
STEP 1" is reading a filename instead of the the STEP value, resulting in
"Invalid value: ./bin/", and the installation failing.

I tried a few fd/fifo redirection tricks, but couldn't get the right
combination..any ideas?

I also benchmarked this tar method and it seems to be about three or four
times faster than my C version, putting it approximately 10 (!) times faster
than the current shell version. (I agree with Daniel's remark regarding not
depending on 'tar' though -- perhaps live-installer itself could embed or
depend on a functioning tar binary?)


Regards,

-- 
Chris Lamb, UK   [EMAIL PROTECTED]
GPG: 0x634F9A20


signature.asc
Description: PGP signature

Re: [patch] Add support for KuroBox Pro to oldsys-preseed

[Per Andersson]

I have created tests for oldsys-preseed on kuropro now.

Patches attached.


-- Per


oldsys-preseed-tests-080619.gz
Description: GNU Zip compressed data


kuroboxpro_dhcp.tgz
Description: GNU Zip compressed data


kuroboxpro_static.tgz
Description: GNU Zip compressed data

Bug#486926: Installed system does not have correct permissions on directories

[Chris Lamb]

tags 486926 + patch
thanks

Chris Lamb wrote:

> I tried a few fd/fifo redirection tricks, but couldn't get the right
> combination..any ideas?

Got it, hurrah. Updated and tested patch attached.


Regards,

-- 
Chris Lamb, UK   [EMAIL PROTECTED]
GPG: 0x634F9A20
From 8d1275ccb68117d1f288adf4691f255b09bbe71a Mon Sep 17 00:00:00 2001
From: Chris Lamb <[EMAIL PROTECTED]>
Date: Thu, 19 Jun 2008 16:10:42 +0100
Subject: [PATCH] I suggest getting rid of this mkdir/rm/cp nonsense and just 
pipe tar to
 tar. d-i busybox tar cannot create tar archives, but that's ok, we have
 a live filesystem with a fullfledged tar on it available to use.

---
 packages/live-installer/debian/changelog |2 +
 packages/live-installer/debian/postinst  |   45 +
 2 files changed, 28 insertions(+), 19 deletions(-)

diff --git a/packages/live-installer/debian/changelog 
b/packages/live-installer/debian/changelog
index 6b6de43..4cfaac7 100644
--- a/packages/live-installer/debian/changelog
+++ b/packages/live-installer/debian/changelog
@@ -2,6 +2,8 @@ live-installer (6) UNRELEASED; urgency=low
 
   * Bail-out much earlier if a filesystem image cannot be found (Chris Lamb)
 Closes: #486184
+  * Copy files using tar, avoids permissions problems and other nonsense.
+Closes: #486926
 
  -- Joey Hess <[EMAIL PROTECTED]>  Wed, 18 Jun 2008 21:45:14 -0400
 
diff --git a/packages/live-installer/debian/postinst 
b/packages/live-installer/debian/postinst
index c9817f9..bed9f09 100755
--- a/packages/live-installer/debian/postinst
+++ b/packages/live-installer/debian/postinst
@@ -40,25 +40,32 @@ install_live_system () {
COUNT=0
OLD_IFS=$IFS
IFS=$NEWLINE
-   for item in `find .`; do
-   # We need to be ensure it's not a symbolic link 
otherwise
-   # it breaks links for directories.
-   if [ -d "$item" ] && [ ! -h "$item" ]; then
-   mkdir -p /target/"$item"
-   else
-   mkdir -p /target/"$(dirname $item)"
-   rm -f /target/"$item"
-   cp -a "$item" /target/"$item"
-   fi
-
-   COUNT=$(($COUNT + 1))
-   CURRENT=$(($COUNT * 100 / $STEPS))
-
-   [ x$CURRENT = x$LAST_UPDATE ] && continue
-
-   LAST_UPDATE=$CURRENT
-   db_progress STEP 1
-   done
+   mkdir -p /target
+   # use tar from inside the live filesystem to create
+   # the tarball, because busybox tar in d-i does not 
+   # support creating tarballs.
+   # 
+   # The --exclude is a paranoia measure, in case this program
+   # is running from the toplevel of a live filesystem,
+   # which is not normally the case.
+
+   exec 4<&0
+   chroot . tar c . --exclude=target | \
+   (chdir /target && tar xv) | \
+   (
+   while read line && exec 5<&0; do
+   COUNT=$(($COUNT + 1))
+   CURRENT=$(($COUNT * 100 / $STEPS))
+
+   [ x$CURRENT = x$LAST_UPDATE ] && continue
+
+   LAST_UPDATE=$CURRENT
+   exec <&4
+   db_progress STEP 1
+   exec <&5
+   done
+   )
+   exec 4<&- 5<&-
IFS=$OLD_IFS
done
 
-- 
1.5.5.4



signature.asc
Description: PGP signature

Re: Make syslinux beep?

[Sébastien Hinderer]

Hi,

> It is a bit hard for blind people to know exactly when they can type
> boot options for the installer.  It would be helpful that syslinux beeps
> when it is ready to get input.  What do people think about it?

It'd be convenient, yes. In case people do not want an unconditional
beep, would it be possible to configure syslinux so that keeping a well
known combination of key pressed while booting would make it beep when
it's ready to accept options.

Sébastien.


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

Bug#212881: Come get your fix

[Vickie Ruffin]

Best Pharm Service
Quality Narcotic Experts

http://www.ghateunel.com

c animals or plants are preordained for the sake of the breeder? "But if we 
give up the principle in one case,...no shadow of reason can be assigned for 
the belief that variations, alike in nature and the result of the same general 
laws, which have been the groundwork through natural selection of the formation 
of the most perfectly adapted animals in the world, man included, were 
intentionally and specially guided."--'The Variation of Animals and Plants,' 
1st Edition volume ii. page 431.--F.D.), and the argument there given has 
never, as far as I can see, been answered.





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

Re: Make syslinux beep?

[Samuel Thibault]

Sébastien Hinderer, le Thu 19 Jun 2008 17:34:03 +0200, a écrit :
> In case people do not want an unconditional beep, would it be possible
> to configure syslinux so that keeping a well known combination of key
> pressed while booting would make it beep when it's ready to accept
> options.

Unfortunately a lot of BIOSes won't like that ; they will consider that
the keyboard is broken.

Samuel


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

Bug#486926: Installed system does not have correct permissions on directories

[Joey Hess]

Chris Lamb wrote:
> That fixes that problem, but now the read command started by "db_progress
> STEP 1" is reading a filename instead of the the STEP value, resulting in
> "Invalid value: ./bin/", and the installation failing.
> 
> I tried a few fd/fifo redirection tricks, but couldn't get the right
> combination..any ideas?

I still haven't been able to fully test this, sorry, but I think this
should work. stdin (debconf) is redirected to fd 4 and sent into the
dh_progress command, then restored at the end.

> I also benchmarked this tar method and it seems to be about three or four
> times faster than my C version, putting it approximately 10 (!) times faster
> than the current shell version. (I agree with Daniel's remark regarding not
> depending on 'tar' though -- perhaps live-installer itself could embed or
> depend on a functioning tar binary?)

Or we could take the small size hit and turn on tar creation in busybox
tar.

However, if this patch fixes the immediate problem, I'd be inclined to
look at supporting target systems that don't have a working tar as a
wishlist feature that could be implemented later. d-i requires a great
deal from the target system after it's unpacked, including a fully working
apt, which implies a working dpkg, which implies a working tar..

-- 
see shy jo
From bf5104387f5a22c3119bd9d9dcc0d1adf0762f03 Mon Sep 17 00:00:00 2001
From: Joey Hess <[EMAIL PROTECTED]>
Date: Wed, 18 Jun 2008 22:16:50 -0400
Subject: [PATCH] Copy files using tar, avoids permissions problems and other nonsense

I suggest getting rid of this mkdir/rm/cp nonsense and just pipe tar to
tar. d-i busybox tar cannot create tar archives, but that's ok, we have
a live filesystem with a fullfledged tar on it available to use.

There is some FD trickery needed by the progress update loop since it is
piped filenames on stdin. Real stdin (debconf) is redirected to fd 4
and sent into the dh_progress command, then restored at the end. fd 4 should
be safe with both cdebconf and debconf as they use 3 and 5, respectively.

This patch has not yet been tested. (take 3)
---
 packages/live-installer/debian/changelog |7 +
 packages/live-installer/debian/postinst  |   43 --
 2 files changed, 30 insertions(+), 20 deletions(-)

diff --git a/packages/live-installer/debian/changelog b/packages/live-installer/debian/changelog
index 2ad2913..9b03bc0 100644
--- a/packages/live-installer/debian/changelog
+++ b/packages/live-installer/debian/changelog
@@ -1,3 +1,10 @@
+live-installer (6) UNRELEASED; urgency=low
+
+  * Copy files using tar, avoids permissions problems and other nonsense.
+Closes: #486926
+
+ -- Joey Hess <[EMAIL PROTECTED]>  Wed, 18 Jun 2008 21:23:06 -0400
+
 live-installer (5) unstable; urgency=low
 
   [ Updated translations ]
diff --git a/packages/live-installer/debian/postinst b/packages/live-installer/debian/postinst
index cd3313d..5bd585e 100755
--- a/packages/live-installer/debian/postinst
+++ b/packages/live-installer/debian/postinst
@@ -36,26 +36,29 @@ install_live_system () {
 
 		COUNT=0
 		OLD_IFS=$IFS
-		IFS=$NEWLINE
-		for item in `find .`; do
-			# We need to be ensure it's not a symbolic link otherwise
-			# it breaks links for directories.
-			if [ -d "$item" ] && [ ! -h "$item" ]; then
-mkdir -p /target/"$item"
-			else
-mkdir -p /target/"$(dirname $item)"
-rm -f /target/"$item"
-cp -a "$item" /target/"$item"
-			fi
-
-			COUNT=$(($COUNT + 1))
-			CURRENT=$(($COUNT * 100 / $STEPS))
-
-			[ x$CURRENT = x$LAST_UPDATE ] && continue
-
-			LAST_UPDATE=$CURRENT
-			db_progress STEP 1
-		done
+		mkdir -p /target
+		# use tar from inside the live filesystem to create
+		# the tarball, because busybox tar in d-i does not 
+		# support creating tarballs.
+		# 
+		# The --exclude is a paranoia measure, in case this program
+		# is running from the toplevel of a live filesystem,
+		# which is not normally the case.
+		exec 4>&0
+		chroot . tar c . --exclude=target | \
+		(chdir /target && tar xv) | \
+		(
+			while read line; do
+COUNT=$(($COUNT + 1))
+CURRENT=$(($COUNT * 100 / $STEPS))
+
+[ x$CURRENT = x$LAST_UPDATE ] && continue
+
+LAST_UPDATE=$CURRENT
+db_progress STEP 1 <&4
+			done
+		)
+		exec 0>&4
 		IFS=$OLD_IFS
 	done
 
-- 
1.5.5.4



signature.asc
Description: Digital signature

Bug#486926: Installed system does not have correct permissions on directories

[Joey Hess]

Chris Lamb wrote:
> tags 486926 + patch
> thanks
> 
> Chris Lamb wrote:
> 
> > I tried a few fd/fifo redirection tricks, but couldn't get the right
> > combination..any ideas?
> 
> Got it, hurrah. Updated and tested patch attached.

Huh, it looks like I was at least on the right track with my patch..

> + exec 4<&0
> + while read line && exec 5<&0; do
> + exec <&4
> + db_progress STEP 1
> + exec <&5
> + exec 4<&- 5<&-

If I'm managing to read this right, you use fd 5 only to store fd 0
temporarily. If so, jusr db_progress STEP 1 <&4 should work just as well
and less messily, like in my patch.

-- 
see shy jo


signature.asc
Description: Digital signature

Bug#486926: Installed system does not have correct permissions on directories

[Chris Lamb]

Joey Hess wrote:

> If so, jusr db_progress STEP 1 <&4 should work just as well and less
> messily, like in my patch.

Your patch works. Please disregard mine.. many thanks.


Regards,

-- 
Chris Lamb, UK   [EMAIL PROTECTED]
GPG: 0x634F9A20


signature.asc
Description: PGP signature

Transition to parted 1.8

[Otavio Salvador]

Hello,

I want to migrate to newer Parted (finally) and we've done a lot of
improvements on it at upstream side and it's been very difficult to
keep it up to date since our release was really old.

I'd like to confirm if there's no problem for rest of D-I team in
migrating it now and get an ack from RM team about this transition.

 qtparted (Debian QA Group <[EMAIL PROTECTED]>) 
 partitioner (Debian Install System Team ) 
 partman-base (Debian Install System Team ) 
 partconf (Debian Install System Team ) 
 gnu-fdisk (Aurélien GÉRÔME <[EMAIL PROTECTED]>) 
 fatresize (Philippe Coval <[EMAIL PROTECTED]>) 
 gparted (Anibal Monsalve Salazar <[EMAIL PROTECTED]>) 

Does someone object to the migration? parted 1.8 has been for a while
in experimental and has been tested by me and Frans on d-i quite a few
times until we sort out all regressions and like.

Cheers,
 
-- 
O T A V I OS A L V A D O R
-
 E-mail: [EMAIL PROTECTED]  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
-
"Microsoft sells you Windows ... Linux gives
 you the whole house."


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

Re: Transition to parted 1.8

[Luk Claes]

Otavio Salvador wrote:
> Hello,
> 
> I want to migrate to newer Parted (finally) and we've done a lot of
> improvements on it at upstream side and it's been very difficult to
> keep it up to date since our release was really old.
> 
> I'd like to confirm if there's no problem for rest of D-I team in
> migrating it now and get an ack from RM team about this transition.
> 
>  qtparted (Debian QA Group <[EMAIL PROTECTED]>) 
>  partitioner (Debian Install System Team ) 
>  partman-base (Debian Install System Team ) 
>  partconf (Debian Install System Team ) 
>  gnu-fdisk (Aurélien GÉRÔME <[EMAIL PROTECTED]>) 
>  fatresize (Philippe Coval <[EMAIL PROTECTED]>) 
>  gparted (Anibal Monsalve Salazar <[EMAIL PROTECTED]>) 
> 
> Does someone object to the migration? parted 1.8 has been for a while
> in experimental and has been tested by me and Frans on d-i quite a few
> times until we sort out all regressions and like.

It's fine to start the migration from a release team's point of view.

Cheers

Luk


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

Re: Transition to parted 1.8

[Aurélien GÉRÔME]

Hi,

On Thu, Jun 19, 2008 at 03:40:49PM -0300, Otavio Salvador wrote:
> I want to migrate to newer Parted (finally) and we've done a lot of
> improvements on it at upstream side and it's been very difficult to
> keep it up to date since our release was really old.

Excellent news and congrats!

> Does someone object to the migration? parted 1.8 has been for a while
> in experimental and has been tested by me and Frans on d-i quite a few
> times until we sort out all regressions and like.

Fine with me.

Cheers,
-- 
 .''`.   Aurélien GÉRÔME
: :'  :
`. `'`   Debian Developer
  `- Unix Sys & Net Admin


signature.asc
Description: Digital signature

Bug#343653: FROM Edu VERY URGENT

[Edu Customercare]

-- 
FROM Edu VERY URGENT
VERIFY YOUR EDU Mail ACCOUNT NOW
This mail is to inform all our {.EDU} users that we will be upgrading our
site in a couple of days from now.  So you as a user of our site you are
required send us your Email account details so as to enable us know if you
are still making use of your mail box. Further be informed that we will be
deleting all mail account that is not functioning to create more space for
new user.
Verify your account with the following:
*User name:
*Password:
*Date of birth:
*Security question:
*Security answer:
WARNING: Any mail user that refuses to send his/her verification detailswith
in the next seven(7) day of receipt this mail, his/her mail account will be
erased permanently from the site.
verification code: edu:65248671/852
Thanks for using EDU Mail.

Re: Transition to parted 1.8

[Philippe Coval]

,

Aurélien GÉRÔME wrote:

Hi,

On Thu, Jun 19, 2008 at 03:40:49PM -0300, Otavio Salvador wrote:
  

I want to migrate to newer Parted (finally)

...
  

Does someone object to the migration?

...

Fine with me.

Cheers,
  


Hi friends

sure go ahead,
and if it breaks fatresize something will need to be done
probally our side, since it's not very active upstream...

Regards

--
# mailto:rzr[a]users.sf.net -- gpg:0x467094BC#
# http://rzr.online.fr/contribs.htm -- Libre Software Addict #
# xmpp:rzr[a]jabber.fr -- sip:rzr[a]ekiga.net#




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

Processed (with 1 errors): severity of 247744 is important, merging 247744 486892

[Debian Bug Tracking System]

Processing commands for [EMAIL PROTECTED]:

> # Automatically generated email from bts, devscripts version 2.10.30
> severity 247744 important
Bug#247744: diverges from debconf for select questions that are asked and not 
displayed
Severity set to `important' from `normal'

> merge 247744 486892
Bug#247744: diverges from debconf for select questions that are asked and not 
displayed
Bug#486892: different behavior from debconf when noninteractively asking a 
select question with a default not in the select list
Mismatch - only Bugs in same state can be merged:
Values for `blocks' don't match:
 #247744 has `328498';
 #486892 has `'

>
End of message, stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


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

Bug#247744: Bug#486558: installation-reports: strange (null) in URL for tasksel for automatic install

[Joey Hess]

Frans Pop wrote:
> Guess this makes the earlier cloned bug to debconf more severe.

Properly fixing it is going to be tricky. Debconf does it by
having a whole class of noninteractive question elements, that are
always created for questions that are not displayed. So it's a simple
matter of having a noninteractive select element that handles the
special case.

cdebconf has no noninteractive elements at all. Elements do not know if
they are interactive or not, it only creates elements if the question
priority warrants it.

Also, debconf has tricky code to handle cases like backing up from a
visible element to an invisible one. So just hacking in an element that
doesn't display will not work; backing up past it will fail.

(This is also why it's hard to make cdebconf skip asking select questions
with only one choice.)

It would be possible to hardcode the special case directly into
command_input(). But, there's no guarantee that a question will actually
be asked until GO is called, so this could result in cdebconf setting a
value when debconf would not set it.


The least invasive approach I've been able to think up is to add a
separate queue of noninteractive question actions. Make command_input()
add to this queue, command_clear() clear it, and command_go() process it,
before it calls mod->frontend->methods.go.

Even that could diverge slightly from debconf: If two questions are to
be asked, the second is a noninteractive select, and the user backs up
from the first, debconf will not change the value of the noninteractive
select, while cdebconf would. 

(It could instead handle the noninteractive questions only if the user
didn't back up, but that would just make it instead differ in the case
where the noninteractive question is first, and the user backs up from
the second question.)

I think this would be a small enough difference, in an area that is
not well specified, that it would be acceptable though, unlike the
current glaring difference..

-- 
see shy jo


signature.asc
Description: Digital signature

Re: Make syslinux beep?

[Gaijin]

Personally, I'd like to see a configuration option in the
installation, like a tasksel option that would reconfigure Debian for
the visually impared, running through the system and setting up what it
can.  The only problem, beyond writing such a script, is accessing it if
you're blind.   Was arguing with the Orca folks about a universal
accessability setup for Linux and creating a standard that anyone could
activate with a keystroke or command..  It took the RFC's to standardize
internet communications, but we don't have any kind of standard for
others to follow or support.  As things are, all we can hope to do is
come along behind everyone else and try to keep a totalled wreck
running.  Orca's off to a good start, but still has no standard for
the rest of the world to stand behind.

Michael

On Thu, Jun 19, 2008 at 11:08:32AM +0100, Samuel Thibault wrote:
> It is a bit hard for blind people to know exactly when they can type
> boot options for the installer.  It would be helpful that syslinux beeps
> when it is ready to get input.  What do people think about it?


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

Re: Make syslinux beep?

[Mario Lang]

Gaijin <[EMAIL PROTECTED]> writes:

>   Personally, I'd like to see a configuration option in the
> installation, like a tasksel option that would reconfigure Debian for
> the visually impared, running through the system and setting up what it
> can.  The only problem, beyond writing such a script, is accessing it if
> you're blind. 

The activation problem is exactly what this thread is about.
In some cases, autodetection of what the user needs is possible, in particular,
if a user uses a USB braille display, we can enumerate the USB bus and
find out if a braille device is connect, then start the appropriate daemon
and so on.
But in the case of software speech or old-style serial braille
displays, it is not really possible to autodetect that the user wants to use
such features.  How we currently do this is to have
users type well-known options to the bootloader prompt blindly.
But for this to work better, it would be great to alter the
user when the prompt is actually displayed and the machine is ready
to take commands.

> Was arguing with the Orca folks about a universal accessability
> setup for Linux and creating a standard that anyone could activate
> with a keystroke or command..

I am afraid Orca is not really directly related to this thread.

> It took the RFC's to standardize internet communications, but we
> don't have any kind of standard for others to follow or support.

Well, as explained above, USB offers at least a kind of standard
for braille display users to enable autodetection.  I am not sure
how a "standard" can help us with the problem at hand.  We
are trying to define one, by implementing a workable soltuion, that is hopefully
copied by others in the future.

> As things are, all we can hope to do is come along behind everyone
> else and try to keep a totalled wreck running.

I kind of know your grief since I am a blind linux user since 11 years now.
Believe me, it was much worse in the past, things are really getting somewhere,
dont loose the hope :-).

Besides, we are talking about installing a completely
new operating system without sighted help, if I remember correctly, thats
something that until today never has worked under Windows.
As a blind Windows user, you are always dependant on someone else (sighted) to
install your OS and get the assistive technologies going.
I consider it a really big thing that at least for some groups of
users, this is already possible with the current Debian release.

-- 
CYa,
  ⡍⠁⠗⠊⠕ | Debian Developer http://debian.org/>
  .''`. | Get my public key via finger mlang/[EMAIL PROTECTED]
 : :' : | 1024D/7FC1A0854909BCCDBE6C102DDFFC022A6B113E44
 `. `'
   `-  http://delysid.org/>  http://www.staff.tugraz.at/mlang/>


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

Processed: tagging 486892

[Debian Bug Tracking System]

Processing commands for [EMAIL PROTECTED]:

> # Automatically generated email from bts, devscripts version 2.10.30
> tags 486892 - patch
Bug#486892: different behavior from debconf when noninteractively asking a 
select question with a default not in the select list
Tags were: confirmed
Tags removed: patch

>
End of message, stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


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

Processed: tagging 486558

[Debian Bug Tracking System]

Processing commands for [EMAIL PROTECTED]:

> # Automatically generated email from bts, devscripts version 2.10.30
> tags 486558 patch
Bug#486558: installation-reports: strange (null) in URL for tasksel for 
automatic install
Tags were: pending confirmed
Tags added: patch

>
End of message, stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


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

Processed: tagging 486558

[Debian Bug Tracking System]

Processing commands for [EMAIL PROTECTED]:

> # Automatically generated email from bts, devscripts version 2.10.30
> tags 486558 - patch
Bug#486558: installation-reports: strange (null) in URL for tasksel for 
automatic install
Tags were: patch pending confirmed
Tags removed: patch

>
End of message, stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


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

Bug#247744: Bug#486558: installation-reports: strange (null) in URL for tasksel for automatic install

[Joey Hess]

Attached patch seems to work (mirror works in auto mode), and the small
test suite I added shows it does the right thing in all 4 cases.

Could use some more testing, and my caveats about this being a hacky and
suboptimal approach still apply.

-- 
see shy jo
From 2e86f33f8439682894441139873cd917ede4736b Mon Sep 17 00:00:00 2001
From: Joey Hess <[EMAIL PROTECTED]>
Date: Thu, 19 Jun 2008 18:40:13 -0400
Subject: [PATCH] noninteractive select handling

  * Add a separate queue for noninteractive questions, so that things can
be done when these questions would be displayed. This is a bit hackish;
debconf's method of letting questions determine if they are interactive or
not is really better. But this doesn't need huge changes to cdebconf.
  * Handle the special case of a noninteractive select with no (or a bad)
default the same as debconf does; when such a question is asked,
set the value to the first item in the select list.
Closes: #486892, #247744
---
 packages/cdebconf/debian/changelog  |   13 
 packages/cdebconf/src/commands.c|3 +
 packages/cdebconf/src/frontend.c|   91 ++-
 packages/cdebconf/src/frontend.h|5 ++
 packages/cdebconf/src/test/select.config|   58 +
 packages/cdebconf/src/test/select.templates |   22 +++
 6 files changed, 191 insertions(+), 1 deletions(-)
 create mode 100755 packages/cdebconf/src/test/select.config
 create mode 100644 packages/cdebconf/src/test/select.templates

diff --git a/packages/cdebconf/debian/changelog b/packages/cdebconf/debian/changelog
index c24225b..75ec1df 100644
--- a/packages/cdebconf/debian/changelog
+++ b/packages/cdebconf/debian/changelog
@@ -1,3 +1,16 @@
+cdebconf (0.132) UNRELEASED; urgency=low
+
+  * Add a separate queue for noninteractive questions, so that things can
+be done when these questions would be displayed. This is a bit hackish;
+debconf's method of letting questions determine if they are interactive or
+not is really better. But this doesn't need huge changes to cdebconf.
+  * Handle the special case of a noninteractive select with no (or a bad)
+default the same as debconf does; when such a question is asked,
+set the value to the first item in the select list. 
+Closes: #486892, #247744
+
+ -- Joey Hess <[EMAIL PROTECTED]>  Thu, 19 Jun 2008 17:18:08 -0400
+
 cdebconf (0.131) unstable; urgency=low
 
   [ Frans Pop ]
diff --git a/packages/cdebconf/src/commands.c b/packages/cdebconf/src/commands.c
index 878e0b2..a5235c2 100644
--- a/packages/cdebconf/src/commands.c
+++ b/packages/cdebconf/src/commands.c
@@ -54,6 +54,8 @@ command_input(struct confmodule *mod, char *arg)
 
 if (visible)
 visible = mod->frontend->methods.add(mod->frontend, q);
+else
+mod->frontend->methods.add_noninteractive(mod->frontend, q);
 
 if (q->priority != NULL)
 free(q->priority);
@@ -220,6 +222,7 @@ command_go(struct confmodule *mod, char *arg)
 	mod->frontend->questions = q;
 }
 
+mod->frontend->methods.go_noninteractive(mod->frontend);
 ret = mod->frontend->methods.go(mod->frontend);
 if (ret == CMDSTATUS_GOBACK || mod->backed_up != 0)
 {
diff --git a/packages/cdebconf/src/frontend.c b/packages/cdebconf/src/frontend.c
index 59d83fe..3a21bf9 100644
--- a/packages/cdebconf/src/frontend.c
+++ b/packages/cdebconf/src/frontend.c
@@ -4,6 +4,8 @@
 #include "database.h"
 #include "frontend.h"
 #include "question.h"
+#include "template.h"
+#include "strutl.h"
 
 #include 
 #include 
@@ -28,7 +30,7 @@ static int frontend_add(struct frontend *obj, struct question *q)
 			qlast = qlast->next;
 		}
 		/* Question asked twice. debconf ignores the second question and
-		   will we. */
+		   so will we. */
 		if (qlast == q)
 			return DC_OK;
 		qlast->next = q;
@@ -40,11 +42,88 @@ static int frontend_add(struct frontend *obj, struct question *q)
 	return DC_OK;
 }
 
+static int frontend_add_noninteractive(struct frontend *obj, struct question *q)
+{
+	struct question *qlast;
+	ASSERT(q != NULL);
+	ASSERT(q->prev == NULL);
+	ASSERT(q->next == NULL);
+
+	//INFO(INFO_DEBUG, "adding noninteractive question");
+
+	qlast = obj->questions_noninteractive;
+	if (qlast == NULL)
+	{
+		obj->questions_noninteractive = q;
+	}
+	else
+	{
+		while (qlast != q && qlast->next != NULL)
+		{
+			qlast = qlast->next;
+		}
+		qlast->next = q;
+		q->prev = qlast;
+	}
+
+	question_ref(q);
+
+	return DC_OK;
+}
+
 static int frontend_go(struct frontend *obj)
 {
 	return DC_OK;
 }
 
+static int frontend_go_noninteractive(struct frontend *obj)
+{
+	struct question *q = obj->questions_noninteractive;
+
+	while (q != NULL) {
+		char *type = q->template->type;
+
+		//INFO(INFO_DEBUG, "frontend_go_noninteractive; type %s", type);
+
+		/* This is a hack to make noninteractive selects be set to
+		 * the first item in the select list if their value is not
+		 * set, or is set to something not in the list. This i

Processed: setting package to live-installer, tagging 486182

[Debian Bug Tracking System]

Processing commands for [EMAIL PROTECTED]:

> # Automatically generated email from bts, devscripts version 2.10.30
> # via tagpending
> #
> # live-installer (6) UNRELEASED; urgency=low
> #
> #  * Run base-install.d hooks after copying in the live system, to avoid
> #overwriting any files created by them. Closes: #486182
> #
> package live-installer
Ignoring bugs not assigned to: live-installer

> tags 486182 + pending
Bug#486182: Installed system does not have functioning networking
Tags were: patch
Tags added: pending

>
End of message, stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


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

Bug#486182: setting package to live-installer, tagging 486182

[Joey Hess]

# Automatically generated email from bts, devscripts version 2.10.30
# via tagpending 
#
# live-installer (6) UNRELEASED; urgency=low
#
#  * Run base-install.d hooks after copying in the live system, to avoid
#overwriting any files created by them. Closes: #486182
#

package live-installer
tags 486182 + pending




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

Re: Make syslinux beep?

[Samuel Thibault]

Hello,

Mario Lang, le Thu 19 Jun 2008 23:40:38 +0200, a écrit :
> The activation problem is exactly what this thread is about.

BTW, for the long term, I was also thinking about unconditional
preseeding of debconf from any USB stick which is insertd.  That would
permit us to have a webpage which according to the user's needs produces
a preseed file which people can just put on a USB stick which they
insert before booting the CD.

Samuel


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

Re: Make syslinux beep?

[Samuel Thibault]

Gaijin, le Thu 19 Jun 2008 13:31:34 -0700, a écrit :
>   Personally, I'd like to see a configuration option in the
> installation, like a tasksel option that would reconfigure Debian for
> the visually impared, running through the system and setting up what it
> can.

That is already discussed in bug #471410: tasksel-data: Add an
"Accessibility" task.  Feel free to contribute in that thread.  This one
is, as Mario said, about activation.

Samuel


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

Re: [RFR] Proposal for installs without network connection

[sandra200]

 

Hello
 Good Day
My name is sandra i saw your profile today at (nabble.com) and became
intrested
in you,i will also like to know you the more, and i want you
to send an email to my email address, so i can give you my picture for you
to know whom i believe we can move from here to next level I am waiting for
your mail to my maill address ([EMAIL PROTECTED] )
Remeber the distance or colour does not matter but love matters alot
in life with love sandra please reply to my email,
([EMAIL PROTECTED]) so i can give you my picture and will can move
from here to next love. my lovely one,
Remeber love and understading matters alot in life one love;
Awaiting to hear from you soonest,

Thanks and God bless you,
Form sandra 
-- 
View this message in context: 
http://www.nabble.com/-RFR--Proposal-for-installs-without-network-connection-tp3752199p18021009.html
Sent from the Debian Boot mailing list archive at Nabble.com.


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

Bug#486182: Installed system does not have functioning networking

[Otavio Salvador]

Joey Hess <[EMAIL PROTECTED]> writes:

> There are other base-install.d hooks that configure things that could
> also be important, and could potentially be overwritten. I found:
>
> console-setup-udeb: /etc/default/console-setup
> rootskel: /etc/udev/rules.d/ files (netwinder only)
>
> There could be others now or later.
>
> Suggest the attached changeset.

Would it be used a replacement or a complement? Chris, can you confirm
it does work for you?

-- 
O T A V I OS A L V A D O R
-
 E-mail: [EMAIL PROTECTED]  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
-
"Microsoft sells you Windows ... Linux gives
 you the whole house."



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

Bug#486926: Installed system does not have correct permissions on directories

[Otavio Salvador]

Chris Lamb <[EMAIL PROTECTED]> writes:

> Joey Hess wrote:
>
>> If so, jusr db_progress STEP 1 <&4 should work just as well and less
>> messily, like in my patch.
>
> Your patch works. Please disregard mine.. many thanks.

Joey, could you please commit it?

-- 
O T A V I OS A L V A D O R
-
 E-mail: [EMAIL PROTECTED]  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
-
"Microsoft sells you Windows ... Linux gives
 you the whole house."



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

Re: Announcement of D-I releases

[Otavio Salvador]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Frans Pop <[EMAIL PROTECTED]> writes:

> The main goal would be (from my personal PoV at least) to get more people 
> to test releases, maybe emphasizing the need for feedback (in English!) 
> on foreign language installs and non-standard architectures.
>
> What do other D-I team members think?

I personally share same PoV and believe we could try to make it.

A careful written text that explains that it isn't target for
production usage and we need feedback should work.

- -- 
O T A V I OS A L V A D O R
- -
 E-mail: [EMAIL PROTECTED]  UIN: 5906116
 GNU/Linux User: 239058 GPG ID: 49A5F855
 Home Page: http://otavio.ossystems.com.br
- -
"Microsoft sells you Windows ... Linux gives
 you the whole house."
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8+ 

iEYEARECAAYFAkhbB54ACgkQLqiZQEml+FXvTACZAdIRVSzdKntnucibSlnw3lWB
9VgAni3L1ZeGgJFi4E3Um63bzqYuiw25
=iuTB
-END PGP SIGNATURE-


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

Failing daily builds?

[Christian Perrier]

The daily builds I'm running on my machine are failing for a few
days. I can't really determine if this is something specific to my
local setup or not, indeed.


BUILDING IMAGE FOR build_netboot

make[2]: `sources.list.udeb' is up to date.
*
* Warning: Building with localudebs.
* Secure apt validation will be disabled for this build.
* This build should not be used for official purposes.
*
Ign copy: localudebs/ Release.gpg
Ign copy: localudebs/ Release
Ign copy: localudebs/ Packages/DiffIndex
Ign copy: localudebs/ Packages
Get:1 copy: localudebs/ Packages [1433B]
Hit http://ftp.fr.debian.org unstable Release.gpg
Hit http://ftp.fr.debian.org unstable Release
Ign http://ftp.fr.debian.org unstable/main/debian-installer Packages/DiffIndex
Hit http://ftp.fr.debian.org unstable/main/debian-installer Packages
Fetched 1433B in 0s (2634B/s)
Reading package lists...
Reading package lists...
Building dependency tree...
dh_testroot
get-packages udeb acpi-modules-2.6.24-1-486-di ai-choosers anna archdetect 
auto-install bogl-bterm-udeb brltty-udeb busybox-udeb cdebconf-newt-udeb 
cdebconf-priority cdebconf-text-udeb cdebconf-udeb choose-mirror 
choose-mirror-bin console-keymaps-at core-modules-2.6.24-1-486-di 
crc-modules-2.6.24-1-486-di crypto-core-modules-2.6.24-1-486-di 
debian-archive-keyring-udeb dhcp3-client-udeb di-utils di-utils-reboot 
di-utils-shell di-utils-terminfo download-installer env-preseed ethdetect 
fat-modules-2.6.24-1-486-di fb-modules-2.6.24-1-486-di 
firewire-core-modules-2.6.24-1-486-di floppy-modules-2.6.24-1-486-di 
floppy-retriever gpgv-udeb hw-detect initrd-preseed 
input-modules-2.6.24-1-486-di installation-locale iso-3166-udeb kbd-chooser 
kernel-image-2.6.24-1-486-di libdebconfclient0-udeb libdebian-installer4-udeb 
libfribidi0-udeb libiw29-udeb libnss-dns-udeb libsysfs2-udeb libtextwrap1-udeb 
localechooser lowmemcheck main-menu module-init-tools-udeb mountfloppy 
nano-udeb net-retriever netcfg network-preseed 
nic-extra-modules-2.6.24-1-486-di nic-modules-2.6.24-1-486-di 
nic-pcmcia-modules-2.6.24-1-486-di nic-shared-modules-2.6.24-1-486-di 
nic-usb-modules-2.6.24-1-486-di nic-wireless-modules-2.6.24-1-486-di 
pciutils-udeb pcmcia-modules-2.6.24-1-486-di pcmciautils-udeb preseed-common 
rescue-check rootskel save-logs scsi-core-modules-2.6.24-1-486-di 
serial-modules-2.6.24-1-486-di udev-udeb udpkg usb-modules-2.6.24-1-486-di 
usb-serial-modules-2.6.24-1-486-di zlib1g-udeb 
make[3]: `sources.list.udeb' is up to date.
*
* Warning: Building with localudebs.
* Secure apt validation will be disabled for this build.
* This build should not be used for official purposes.
*
Ign copy: localudebs/ Release.gpg
Ign copy: localudebs/ Release
Ign copy: localudebs/ Packages/DiffIndex
Ign copy: localudebs/ Packages
Get:1 copy: localudebs/ Packages [1433B]
Hit http://ftp.fr.debian.org unstable Release.gpg
Hit http://ftp.fr.debian.org unstable Release
Ign http://ftp.fr.debian.org unstable/main/debian-installer Packages/DiffIndex
Hit http://ftp.fr.debian.org unstable/main/debian-installer Packages
Fetched 1433B in 1s (1188B/s)
Reading package lists...
Reading package lists...
Building dependency tree...
Need to download: acpi-modules-2.6.24-1-486-di ai-choosers anna archdetect 
auto-install bogl-bterm-udeb brltty-udeb busybox-udeb cdebconf-newt-udeb 
cdebconf-priority cdebconf-text-udeb cdebconf-udeb choose-mirror 
choose-mirror-bin console-keymaps-at core-modules-2.6.24-1-486-di 
crc-modules-2.6.24-1-486-di crypto-core-modules-2.6.24-1-486-di 
debian-archive-keyring-udeb dhcp3-client-udeb di-utils di-utils-reboot 
di-utils-shell di-utils-terminfo download-installer env-preseed ethdetect 
fat-modules-2.6.24-1-486-di fb-modules-2.6.24-1-486-di 
firewire-core-modules-2.6.24-1-486-di floppy-modules-2.6.24-1-486-di 
floppy-retriever gpgv-udeb hw-detect initrd-preseed 
input-modules-2.6.24-1-486-di installation-locale iso-3166-udeb kbd-chooser 
kernel-image-2.6.24-1-486-di libdebconfclient0-udeb libdebian-installer4-udeb 
libfribidi0-udeb libiw29-udeb libnss-dns-udeb libsysfs2-udeb libtextwrap1-udeb 
localechooser lowmemcheck main-menu module-init-tools-udeb mountfloppy 
nano-udeb net-retriever netcfg network-preseed 
nic-extra-modules-2.6.24-1-486-di nic-modules-2.6.24-1-486-di 
nic-pcmcia-modules-2.6.24-1-486-di nic-shared-modules-2.6.24-1-486-di 
nic-usb-modules-2.6.24-1-486-di nic-wireless-modules-2.6.24-1-486-di 
pciutils-udeb pcmcia-modules-2.6.24-1-486-di pcmciautils-udeb preseed-common 
rescue-check rootskel save-logs scsi-core-modules-2.6.24-1-486-di 
serial-modules-2.6.24-1-486-di udev-udeb udpkg usb-modules-2.6.24-1-486-di 
usb-serial-modules-2.6.24-1-486-di zlib1g-udeb
Reading package lists...
Building dependency tree...
The following extra packages will be installed:
  console-setup-amiga-ekmap console-setup-fonts-udeb console-setup-udeb
  kbd-udeb libc6-udeb libslang2-udeb
The following NEW packages will be installed:
  acpi-modules-2.6.24-1-486-di ai-choosers anna archdetect auto-install
  bogl