Re: Put the debian-installer wheezy goals in w.d.o/Releasegoals/DebianInstaller instead of just wheezy goals
Quoting shirish शिरीष (shirisha...@gmail.com): Hi all, Please CC me if somebody responds to the topic. While browsing on/around the Wiki I came to http://wiki.debian.org/DebianInstaller/WheezyGoals and this isn't a nice place to share the wheezy goals. A much better place is/would be http://wiki.debian.org/ReleaseGoals with something like http://wiki.debian.org/ReleaseGoals/DebianInstaller . A change to this URL would have few positives IMHO :- a. It would be consistent b. It is more easily and more likely to be found. c. Possible that more people likely to test and hence have more robust releases. I personnally (ie not speaking for the D-I team...wait for a few other people's advice) am OK with such changeif there's a perceived benefit to it. My first reaction would be how about just linking the D-I release goals page from the general release goals page?but there might be other benefits I'm not aware about when moving the page. Something I'm quite sure, though: these pages do not help that much in getting the needed manpower to actually achieve the goals;.:-) Most of the current top D-I features are here because, suddenly, at one moment or another, someone cared enough to work on them. Ou rproblem in the long term is that, often, that someone later vanished or moves to other interests and we're left with the feature... and its bugs. So, indeed, ATM, what's much more needed is not necessarily new fancy stuff in D-I but more securing and debugging existing stuff (LVM, crypto, network configurationare those that come to mind). signature.asc Description: Digital signature
Bug#623410: installation problems with squeeze
Quoting melbogia (melbo...@gmail.com): We have a number of exact same machines that have a 2GB ATP flash cards in addition to mechanical disks. When I try to install, the installer sees the USB flash card as /dev/sda, but it does not sees the disk, however if I physically remove the flash card then the installer sees all the disks. This is happening with Squeeze. I have a nfsroot environment built on Lenny and it sees all the disks as well as the flash card, all as /dev/sdX. To narrow the problem, could you try the daily built installation images (from http://www.debian.org/devel/debian-installer)? Preferrably the netboot variant. signature.asc Description: Digital signature
SPAMing from my email address
Hi, I deactivated my gmail account while ago but some one use it for sending SPAM mails, at first i apologize for this. I will delete my Gamil account if i can or deactivate it. so i will not send any email with my Gmail account again and if you get any mail from my account it will be a SPAM for sure. from now on my primary mail address will be lxsameer [AT] GNU.org thanks for your consideration
Bug#254727: partman-auto: reusing existing partitions with same method
I've been looking at http://bugs.debian.org/254727 (Creates swap even
when there is already one), and also at
https://bugs.launchpad.net/bugs/746313 (partman should reuse existing
BIOS Boot Partition) which has the same cause but has more serious
consequences.
In general, I'm aware that partman doesn't have very good support for
reusing existing partitions. Much of the problem is in defining good
recipe syntax for it: how do you name the partition you want to reuse?
Partition numbers may be too specific for some users, not every
partition has a UUID, etc. However, in this case, I think we could
define a simple abstraction: if there's already a partition with this
method, I want to reuse it, and encode that as a $reusemethod{ }
internal specifier in partman recipes.
The implementation is a little tricky. You have to exclude reused
partitions from size calculations, but you also have to perform an
action upon applying the recipe. I found that it was easiest to
transform $reusemethod{ } into $reuse{ ID } when decoding the recipe,
and then deal with that when performing the recipe.
I've pushed an initial implementation of this here:
http://git.debian.org/?p=d-i/partman-auto.git;a=shortlog;h=refs/heads/people/cjwatson/reusemethod
and the diff against master is attached. (Of course, we'd also need to
add $reusemethod{ } to a bunch of recipe stanzas.) What do people
think of this?
--
Colin Watson [cjwat...@ubuntu.com]
diff --git a/debian/changelog b/debian/changelog
index ecc8560..9500786 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,6 +1,10 @@
partman-auto (98) UNRELEASED; urgency=low
* Update old comment about partman-auto-lvm and $iflabel.
+ * Refactor some of decode_recipe to make it easier to have multiple ways
+to ignore recipe stanzas.
+ * Add $reusemethod internal specifier, which excludes the partition if a
+partition with the same method already exists.
-- Colin Watson cjwat...@debian.org Wed, 20 Apr 2011 11:09:22 +0100
diff --git a/lib/auto-shared.sh b/lib/auto-shared.sh
index 3d9e66a..72573b5 100644
--- a/lib/auto-shared.sh
+++ b/lib/auto-shared.sh
@@ -92,9 +92,22 @@ ensure_primary() {
)
}
-create_primary_partitions() {
+reuse_partitions() {
cd $dev
+ local scheme
+
+ scheme=$scheme_reused
+ foreach_partition '
+ id=$(echo $* | sed -n '\''s/.* \$reuse{ \([^}]*\) }.*/\1/p'\'')
+ if [ -z $id ] || [ ! -d $id ]; then
+ db_progress STOP
+ autopartitioning_failed
+ fi
+ setup_partition $id $*'
+}
+create_primary_partitions() {
+ cd $dev
while [ $free_type = pri/log ] \
echo $scheme | grep -q '\$primary{'; do
pull_primary
diff --git a/lib/recipes.sh b/lib/recipes.sh
index 396f6d9..016eb58 100644
--- a/lib/recipes.sh
+++ b/lib/recipes.sh
@@ -24,10 +24,26 @@ autopartitioning_failed () {
exit 1
}
+find_method () {
+ local num id size type fs path name method found
+ found=
+ open_dialog PARTITIONS
+ while { read_line num id size type fs path name; [ $id ]; }; do
+ [ -f $id/method-old ] || continue
+ method=$(cat $id/method-old)
+ if [ $method = $1 ]; then
+ found=$id
+ fi
+ done
+ close_dialog
+ echo $found
+}
+
unnamed=0
decode_recipe () {
local ignore ram line word min factor max fs iflabel label -
+ local reusemethod method id
ignore=${2:+${2}ignore}
unnamed=$(($unnamed + 1))
ram=$(grep ^Mem: /proc/meminfo | { read x y z; echo $y; }) # in bytes
@@ -109,30 +125,45 @@ decode_recipe () {
# Exclude partitions that have ...ignore set
if [ $ignore ] [ $(echo $line | grep $ignore) ]; then
-:
- else
-# Exclude partitions that are only for a
-# different disk label. The $PWD check
-# avoids problems when running from older
-# versions of partman-auto-lvm, where we
-# weren't in a subdirectory of $DEVICES
-# while decoding the recipe; we preserve it
-# in case of custom code with the same
-# problem.
-iflabel=$(echo $line | sed -n 's/.*\$iflabel{ \([^}]*\) }.*/\1/p')
-if [ $iflabel ]; then
- if [ ${PWD#$DEVICES/} != $PWD ]; then
- open_dialog GET_LABEL_TYPE
- read_line label
- close_dialog
- if [ $iflabel = $label ]; then
- scheme=${scheme:+$scheme$NL}$line
- fi
+line=
+continue
+ fi
+
+ # Exclude partitions that are only for a different
+ # disk label. The $PWD check avoids problems when
+ # running from older versions of partman-auto-lvm,
+ # where we weren't in a subdirectory of $DEVICES
+ # while decoding the recipe; we preserve it in case
+ # of custom code with the same problem.
+ iflabel=$(echo $line | sed -n 's/.*\$iflabel{ \([^}]*\) }.*/\1/p')
+ if [ $iflabel ]; then
+if [ ${PWD#$DEVICES/} = $PWD ]; then
+ line=''
+ continue
+fi
+
+open_dialog GET_LABEL_TYPE
+read_line label
+close_dialog
+if [ $iflabel != $label ]; then
+ line=''
+ continue
+fi
+ fi
+
+ # Exclude partitions where we can reuse an
Re: Bug#254727: partman-auto: reusing existing partitions with same method
On Wed, Apr 20, 2011 at 05:18:52PM +0100, Colin Watson wrote:
I've been looking at http://bugs.debian.org/254727 (Creates swap even
when there is already one), and also at
https://bugs.launchpad.net/bugs/746313 (partman should reuse existing
BIOS Boot Partition) which has the same cause but has more serious
consequences.
In general, I'm aware that partman doesn't have very good support for
reusing existing partitions. Much of the problem is in defining good
recipe syntax for it: how do you name the partition you want to reuse?
Partition numbers may be too specific for some users, not every
partition has a UUID, etc. However, in this case, I think we could
define a simple abstraction: if there's already a partition with this
method, I want to reuse it, and encode that as a $reusemethod{ }
internal specifier in partman recipes.
The implementation is a little tricky. You have to exclude reused
partitions from size calculations, but you also have to perform an
action upon applying the recipe. I found that it was easiest to
transform $reusemethod{ } into $reuse{ ID } when decoding the recipe,
and then deal with that when performing the recipe.
I've pushed an initial implementation of this here:
http://git.debian.org/?p=d-i/partman-auto.git;a=shortlog;h=refs/heads/people/cjwatson/reusemethod
and the diff against master is attached. (Of course, we'd also need to
add $reusemethod{ } to a bunch of recipe stanzas.) What do people
think of this?
What if the other OS using a swap partition is currently suspended
to disk? Can you share the swap partition in that case? Sounds rather
unsafe in some cases.
Of course this doesn't mean you can't let people share them, but doing
it by default would seem hazardous.
--
Len Sorensen
--
To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive:
http://lists.debian.org/20110420171215.gr...@caffeine.csclub.uwaterloo.ca
Bug#623430: install bug report
Quoting 東政一 (fujira...@gmail.com): Package: installation-reports reporter：Ｍ Aｚｕｍａ Boot method: インストーラの起動方法は、CDから Image version: 2011/4/19 ダウンロードした URL 全体は覚えていない Date: 2011/4/19 ｐｍ5 日本時間 Machine: 自作機 motherboard ASUS P4S800D-X Processor: Pentium 4 2.66ghz Memory: 768mbyte Partitions: df -Tl の結果; 生のパーティションテーブルが望ましいです none Output of lspci -nn and lspci -vnn: added file lspci＆”ｌｓｐｃｉｖ” Base System Installation Checklist: [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it Initial boot: [O] Detect network card:[ ] Configure network: [ ] Detect CD: [ E] Load installer modules: [ ] Detect hard drives: [ ] Partition hard drives: [ ] Install base system:[ ] Clock/timezone setup: [ ] User/password setup:[ ] Install tasks: [ ] Install boot loader:[ ] Overall install:[ ] 参考 LENNY 5.08を同じマシンで正常に動作させている。 .../... Hello, I don't see any really obviously useful information in the cited bug report, exceptmaybe the above line which, unfortunately, I can't read. Could you elaborate about this bug report, preferrably in English? Many thanks in advance. signature.asc Description: Digital signature
debian-installer: armhf initial work
Hello, I would like to start merging armhf changes into debian-installer master branch. I wanted to ask if someone feels opposed to that work going into master or have any comment about it. Initial changes are very preeliminar and it might need some fiddling to get it working as it is using debian-ports.org repositories and authentication against debian-ports.org is not supported. I am working in parallel to get debian kernel armhf integration, so kernel packages can be produced (#621032). Best regards, -- Héctor Orón Our Sun unleashes tremendous flares expelling hot gas into the Solar System, which one day will disconnect us. -- Day DVB-T stop working nicely Video flare: http://antwrp.gsfc.nasa.gov/apod/ap100510.html -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/BANLkTi=+2+fvudzks7yyqnnvhcyc6-1...@mail.gmail.com
Re: debian-installer: armhf initial work
On Wed, Apr 20, 2011 at 08:29:19PM +0200, Hector Oron wrote: Hello, I would like to start merging armhf changes into debian-installer master branch. I wanted to ask if someone feels opposed to that work going into master or have any comment about it. I think this makes sense - now is the right time to do this, right after a release. This may be disruptive, but I think it is a valuable addition to D-I and I think people will understand. It might be nice to get some information and updates about your progress so people can test and help out if needed. Regards, Jeremiah -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110420205742.GA24293@localhost
Bug#621400: debian-installer: keymap alias
Christian PERRIER, le Mon 11 Apr 2011 06:43:15 +0200, a écrit : But isn't console-setup using the console-keymaps-at/keymap value if already set? Actually it uses debian-installer/keymap. Anyway, even if we completely drop console-data udeb, we can still keep a keymap alias, which we'll be pointing to whatever is needed to properly preseed c-s, can't we? Perhaps, but which kind of value should it contain? console-data-ish, or xkb-ish, or something else? Currently, the basic console-setup keymap configuration item is xkb-ish, in keyboard-configuration/xkb-keymap Samuel -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20110421001127.gy10...@const.famille.thibault.fr
Debian installer build: failed or old builds
Debian installer build overview --- Failed or old builds: * FAILED BUILD: armel Apr 19 22:14 joey@box build_orion5x_network-console http://people.debian.org/~joeyh/d-i/armel/images/daily/build_orion5x_network-console.log * FAILED BUILD: armel Apr 19 22:30 joey@box build_iop32x_netboot http://people.debian.org/~joeyh/d-i/armel/images/daily/build_iop32x_netboot.log * FAILED BUILD: armel Apr 19 22:31 joey@box build_versatile_netboot http://people.debian.org/~joeyh/d-i/armel/images/daily/build_versatile_netboot.log * FAILED BUILD: armel Apr 19 22:47 joey@box build_ads_cf http://people.debian.org/~joeyh/d-i/armel/images/daily/build_ads_cf.log * FAILED BUILD: armel Apr 20 02:25 joey@box build_ixp4xx_network-console http://people.debian.org/~joeyh/d-i/armel/images/daily/build_ixp4xx_network-console.log * FAILED BUILD: armel Apr 20 08:48 joey@box build_orion5x_network-console http://people.debian.org/~joeyh/d-i/armel/images/daily/build_orion5x_network-console.log * FAILED BUILD: armel Apr 20 10:09 joey@box build_ads_cf http://people.debian.org/~joeyh/d-i/armel/images/daily/build_ads_cf.log * OLD BUILD:powerpc Apr 19 00:11 buildd@praetorius build_powerpc_cdrom http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc_cdrom.log * OLD BUILD:powerpc Apr 19 00:15 buildd@praetorius build_powerpc_netboot http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc_netboot.log * OLD BUILD:powerpc Apr 19 00:19 buildd@praetorius build_powerpc_netboot-gtk http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc_netboot-gtk.log * OLD BUILD:powerpc Apr 19 00:21 buildd@praetorius build_powerpc_hd-media http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc_hd-media.log * OLD BUILD:powerpc Apr 19 00:24 buildd@praetorius build_powerpc64_cdrom http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc64_cdrom.log * OLD BUILD:powerpc Apr 19 00:26 buildd@praetorius build_powerpc64_netboot http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc64_netboot.log * OLD BUILD:powerpc Apr 19 00:31 buildd@praetorius build_powerpc64_netboot-gtk http://d-i.debian.org/daily-images/powerpc/daily/build_powerpc64_netboot-gtk.log * FAILED BUILD: amd64 Apr 18 10:32 debian-cd@pettersson 3sids390 http://cdbuilder.debian.org/cdimage-log/3sids390 * FAILED BUILD: amd64 Apr 18 10:32 debian-cd@pettersson sids390 http://cdbuilder.debian.org/cdimage-log/sids390 * FAILED BUILD: amd64 Apr 18 10:32 debian-cd@pettersson Bsids390 http://cdbuilder.debian.org/cdimage-log/Bsids390 * FAILED BUILD: amd64 Apr 18 10:32 debian-cd@pettersson 4sids390 http://cdbuilder.debian.org/cdimage-log/4sids390 Totals: 129 builds (11 failed, 7 old) -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/e1qcify-0001vs...@ravel.debian.org
