[arch-releng] automated iso builds

newer
[arch-releng] Note about removing...

older
[arch-releng] [PATCH 5/6] Fix doc...

Dieter Plaetinck

9 Feb 2010 9 Feb '10

9:28 p.m.

so i blew the dust of our releng chroots, updated them, installed latest archiso from git and so on. i built this experimental isolinux image (with testing repo enabled!), basically to see if everything still works. http://build.archlinux.org/isos/archlinux-2010.02.09-netinstall-i686.iso through incremental updates, continuous integration and crowd-sourced testing i want to set up automated builds and host all media in the cloud. (if you know what i mean ;-) btw: the image sort of boots, but it can't find /dev/archiso. /me off. to be continued. Dieter

Show replies by date

Gerardo Exequiel Pozzi

9 Feb 9 Feb

9:50 p.m.

On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...

so i blew the dust of our releng chroots, updated them, installed latest archiso from git and so on.

i built this experimental isolinux image (with testing repo enabled!), basically to see if everything still works. http://build.archlinux.org/isos/archlinux-2010.02.09-netinstall-i686.iso

through incremental updates, continuous integration and crowd-sourced testing i want to set up automated builds and host all media in the cloud. (if you know what i mean ;-)

btw: the image sort of boots, but it can't find /dev/archiso.

/me off. to be continued.

Dieter

Good!, the only problem for doing automatic builds is that in _random_ cases squashfs image of "overlay" is created corrupted (from the same data) :( I experimented this many times, and the way to test if all is good is: or doing a boot, or mounting the squashfs image on read some files. I can remember that also tested if the problem persiste with mksquashfs -processors 1 but the problem still persist. -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Gerardo Exequiel Pozzi

9:58 p.m.

On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...

so i blew the dust of our releng chroots, updated them, installed latest archiso from git and so on.

i built this experimental isolinux image (with testing repo enabled!), basically to see if everything still works. http://build.archlinux.org/isos/archlinux-2010.02.09-netinstall-i686.iso

through incremental updates, continuous integration and crowd-sourced testing i want to set up automated builds and host all media in the cloud. (if you know what i mean ;-)

btw: the image sort of boots, but it can't find /dev/archiso.

/me off. to be continued.

Dieter

I tested the image and does not boot. Seems that you have an old enviroment, the initramfs is created with new archiso* hooks but is with mkinitcpio-0.5.... -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

10 Feb 10 Feb

12:35 p.m.

On Tue, 09 Feb 2010 18:58:14 -0300 Gerardo Exequiel Pozzi wrote:

...

Good!, the only problem for doing automatic builds is that in _random_ cases squashfs image of "overlay" is created corrupted (from the same data) :( I experimented this many times, and the way to test if all is good is: or doing a boot, or mounting the squashfs image on read some files. I can remember that also tested if the problem persiste with mksquashfs -processors 1 but the problem still persist.

that's odd. can you investigate? maybe this is a bug in the squashfs tools? in the worst case scenario i add a bit of code to the buildscripts that mounts the image and tests it.

...

I tested the image and does not boot. Seems that you have an old enviroment, the initramfs is created with new archiso* hooks but is with mkinitcpio-0.5....

Hmm. I `pacman -Syu`'d the chroot in which this takes place, so it has the latest updates. should i get the mkinitcpio 0.5.99.4-1 from testing? Dieter

Gerardo Exequiel Pozzi

7:49 p.m.

On 02/10/2010 09:35 AM, Dieter Plaetinck wrote:

...

On Tue, 09 Feb 2010 18:58:14 -0300 Gerardo Exequiel Pozzi wrote:

...
Good!, the only problem for doing automatic builds is that in _random_ cases squashfs image of "overlay" is created corrupted (from the same data) :( I experimented this many times, and the way to test if all is good is: or doing a boot, or mounting the squashfs image on read some files. I can remember that also tested if the problem persiste with mksquashfs -processors 1 but the problem still persist.

that's odd. can you investigate? maybe this is a bug in the squashfs tools? in the worst case scenario i add a bit of code to the buildscripts that mounts the image and tests it.

I can remember an email from GerBra, which these errors [#1]. But this is unrelated to errors messages about ioctl. The bug is on mksqashfs. I will look how can report it to devs ;) The bad image is created with lots of \x00 just after the header. On mksquashfs the error is detected, but the program always return 0, the only difference is you can see this message: Lseek on destination failed because Bad file descriptor or: Write failed because Bad file descriptor Write on destination failed because Bad file descriptor mksquashfs work/overlay lolcat.sqfs -noappend adding one file inside (about ~100k), I can not reproduce the error. (I tried with different block sizes but is the same)

...

...
I tested the image and does not boot. Seems that you have an old enviroment, the initramfs is created with new archiso* hooks but is with mkinitcpio-0.5....

Hmm. I `pacman -Syu`'d the chroot in which this takes place, so it has the latest updates. should i get the mkinitcpio 0.5.99.4-1 from testing?

Rare, the steps I done (I ran just now, with latest mkinitcpio-0.5.99.5-1): mkarchroot /tmp/pichicho base (without [testing] enabled, to avoid file conflicts between repos) mkarchroot -r bash /tmp/pichicho # enable [testing] on pacman.conf and enable some mirror in mirrorlist pacman -Syu # now pacman will prompt about replaces klibc-* pacman -S git squashfs-tools syslinux devtools cdrkit make cd /tmp git clone git://projects.archlinux.org/archiso.git cd archiso/archiso make install cd ../configs/syslinux-iso/ make net-iso #and wait < 3 minutes. [#1] http://mailman.archlinux.org/pipermail/arch-releng/2009-August/000601.html -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Gerardo Exequiel Pozzi

8:08 p.m.

On 02/10/2010 04:49 PM, Gerardo Exequiel Pozzi wrote:

...

On 02/10/2010 09:35 AM, Dieter Plaetinck wrote:

...
On Tue, 09 Feb 2010 18:58:14 -0300 Gerardo Exequiel Pozzi wrote:

...
Good!, the only problem for doing automatic builds is that in _random_ cases squashfs image of "overlay" is created corrupted (from the same data) :( I experimented this many times, and the way to test if all is good is: or doing a boot, or mounting the squashfs image on read some files. I can remember that also tested if the problem persiste with mksquashfs -processors 1 but the problem still persist. that's odd. can you investigate? maybe this is a bug in the squashfs tools? in the worst case scenario i add a bit of code to the buildscripts that mounts the image and tests it.

I can remember an email from GerBra, which these errors [#1]. But this is unrelated to errors messages about ioctl. The bug is on mksqashfs. I will look how can report it to devs ;) The bad image is created with lots of \x00 just after the header.

On mksquashfs the error is detected, but the program always return 0, the only difference is you can see this message: Lseek on destination failed because Bad file descriptor or: Write failed because Bad file descriptor Write on destination failed because Bad file descriptor

mksquashfs work/overlay lolcat.sqfs -noappend

adding one file inside (about ~100k), I can not reproduce the error. (I tried with different block sizes but is the same) Applying this patch solves the issue :)

http://squashfs.cvs.sourceforge.net/viewvc/squashfs/squashfs/squashfs-tools/mksquashfs.c?r1=1.141&r2=1.142 Opening a bug report in FS ;) -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Gerardo Exequiel Pozzi

8:29 p.m.

On 02/10/2010 05:08 PM, Gerardo Exequiel Pozzi wrote:

...

On 02/10/2010 04:49 PM, Gerardo Exequiel Pozzi wrote:

...
On 02/10/2010 09:35 AM, Dieter Plaetinck wrote:

...
On Tue, 09 Feb 2010 18:58:14 -0300 Gerardo Exequiel Pozzi wrote:

...
Good!, the only problem for doing automatic builds is that in _random_ cases squashfs image of "overlay" is created corrupted (from the same data) :( I experimented this many times, and the way to test if all is good is: or doing a boot, or mounting the squashfs image on read some files. I can remember that also tested if the problem persiste with mksquashfs -processors 1 but the problem still persist. that's odd. can you investigate? maybe this is a bug in the squashfs tools? in the worst case scenario i add a bit of code to the buildscripts that mounts the image and tests it.

I can remember an email from GerBra, which these errors [#1]. But this is unrelated to errors messages about ioctl. The bug is on mksqashfs. I will look how can report it to devs ;) The bad image is created with lots of \x00 just after the header.

On mksquashfs the error is detected, but the program always return 0, the only difference is you can see this message: Lseek on destination failed because Bad file descriptor or: Write failed because Bad file descriptor Write on destination failed because Bad file descriptor

mksquashfs work/overlay lolcat.sqfs -noappend

adding one file inside (about ~100k), I can not reproduce the error. (I tried with different block sizes but is the same) Applying this patch solves the issue :)

http://squashfs.cvs.sourceforge.net/viewvc/squashfs/squashfs/squashfs-tools/mksquashfs.c?r1=1.141&r2=1.142

Opening a bug report in FS ;)

Done: http://bugs.archlinux.org/task/18277 -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Gerardo Exequiel Pozzi

1 Mar 1 Mar

1:02 a.m.

On 02/10/2010 05:29 PM, Gerardo Exequiel Pozzi wrote:

...

...
http://squashfs.cvs.sourceforge.net/viewvc/squashfs/squashfs/squashfs-tools/mksquashfs.c?r1=1.141&r2=1.142

Opening a bug report in FS ;)

Applying this patch solves the issue :) Done: http://bugs.archlinux.org/task/18277

Applied/fixed in latest squashfs-tools-4.0-3, [ Gracias Ghost1227 :) ] -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

10 Feb 10 Feb

9:09 p.m.

can you bundle your replies/thoughts in 1 mail? makes things easier to follow and reply to. On Wed, 10 Feb 2010 17:29:15 -0300 Gerardo Exequiel Pozzi wrote:

...

...
Applying this patch solves the issue :)

http://squashfs.cvs.sourceforge.net/viewvc/squashfs/squashfs/squashfs-tools/mksquashfs.c?r1=1.141&r2=1.142

Opening a bug report in FS ;)

Done: http://bugs.archlinux.org/task/18277

any ETA on a new squashfs release? maybe we should give them some time instead of patching ourself. although it looks like they don't release very often and this patch is a few months old.

...

Rare, the steps I done (I ran just now, with latest mkinitcpio-0.5.99.5-1):

what do you mean with rare?

...

# enable [testing] on pacman.conf and enable some mirror in mirrorlist pacman -Syu # now pacman will prompt about replaces klibc-*

okay so I need testing. I guess i'll just wait a bit for the new mkinitcpio to appear in core. Dieter

Gerardo Exequiel Pozzi

9:56 p.m.

On 02/10/2010 06:09 PM, Dieter Plaetinck wrote:

...

can you bundle your replies/thoughts in 1 mail? makes things easier to follow and reply to.

On Wed, 10 Feb 2010 17:29:15 -0300 Gerardo Exequiel Pozzi wrote:

...
...
Applying this patch solves the issue :)

http://squashfs.cvs.sourceforge.net/viewvc/squashfs/squashfs/squashfs-tools/mksquashfs.c?r1=1.141&r2=1.142

Opening a bug report in FS ;)

Done: http://bugs.archlinux.org/task/18277

any ETA on a new squashfs release? maybe we should give them some time instead of patching ourself. although it looks like they don't release very often and this patch is a few months old.

What is ETA? The patch does not look bad and solves the issue. Seems that release of squashfs-tools are a bit slow: 4.0 2009/04 3.4 2008/08 3.0 2006/03 More info about bug report here: http://sourceforge.net/mailarchive/message.php?msg_name=87k529zp8f.fsf%40mac...

...

...
Rare, the steps I done (I ran just now, with latest mkinitcpio-0.5.99.5-1):

what do you mean with rare?

weird, but now I understand ...

...

...
# enable [testing] on pacman.conf and enable some mirror in mirrorlist pacman -Syu # now pacman will prompt about replaces klibc-*

okay so I need testing. I guess i'll just wait a bit for the new mkinitcpio to appear in core.

I guessed that you have enabled [testing] at all (on your enviroment and on iso) in reference to this: On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...

i built this experimental isolinux image (with testing repo enabled!),

Latest git of archiso requires "mkinitcpio-0.6" from testing (by error, Aaron merge my "experimental" branch instead of "djgera") [#1] otherwise you need to reset to commit a2f6841c so can build against current [core]. [#1] http://mailman.archlinux.org/pipermail/arch-releng/2010-February/000807.html -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

11 Feb 11 Feb

7:29 a.m.

On Wed, 10 Feb 2010 18:56:58 -0300 Gerardo Exequiel Pozzi wrote:

...

...
On Wed, 10 Feb 2010 17:29:15 -0300 Gerardo Exequiel Pozzi wrote:

...
Done: http://bugs.archlinux.org/task/18277

any ETA on a new squashfs release? maybe we should give them some time instead of patching ourself. although it looks like they don't release very often and this patch is a few months old.

What is ETA? The patch does not look bad and solves the issue. Seems that release of squashfs-tools are a bit slow: 4.0 2009/04 3.4 2008/08 3.0 2006/03

More info about bug report here: http://sourceforge.net/mailarchive/message.php?msg_name=87k529zp8f.fsf%40mac...

allright. fine by me then. hope Tpowa accepts it :) (eta = estimated time of arrival)

...

I guessed that you have enabled [testing] at all (on your enviroment and on iso) in reference to this:

On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...
i built this experimental isolinux image (with testing repo enabled!),

Latest git of archiso requires "mkinitcpio-0.6" from testing (by error, Aaron merge my "experimental" branch instead of "djgera") [#1] otherwise you need to reset to commit a2f6841c so can build against current [core].

I meant that the iso is built from testing+core packages. The environment (build-chroot) is core only. Frankly I'm not sure if there's much point in doing this. Maybe it's better to only use core for both the chroot and the actual iso's Dieter

Gerardo Exequiel Pozzi

7:18 p.m.

On 02/11/2010 04:29 AM, Dieter Plaetinck wrote:

...

On Wed, 10 Feb 2010 18:56:58 -0300 Gerardo Exequiel Pozzi wrote:

...
I guessed that you have enabled [testing] at all (on your enviroment and on iso) in reference to this:

On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...
i built this experimental isolinux image (with testing repo enabled!),

Latest git of archiso requires "mkinitcpio-0.6" from testing (by error, Aaron merge my "experimental" branch instead of "djgera") [#1] otherwise you need to reset to commit a2f6841c so can build against current [core].

I meant that the iso is built from testing+core packages. The environment (build-chroot) is core only. Frankly I'm not sure if there's much point in doing this. Maybe it's better to only use core for both the chroot and the actual iso's

Dieter

Yes, the original idea was work on 2010.02 with current [core]. mkinitcpio-0.6 is in [testing] now, I guess in few days will be moved to [core]. Maybe 2010.02 will be based on it, that works fine. -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

17 Feb 17 Feb

9:50 p.m.

On Tue, 09 Feb 2010 18:58:14 -0300 Gerardo Exequiel Pozzi wrote:

...

On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...
so i blew the dust of our releng chroots, updated them, installed latest archiso from git and so on.

i built this experimental isolinux image (with testing repo enabled!), basically to see if everything still works. http://build.archlinux.org/isos/archlinux-2010.02.09-netinstall-i686.iso

through incremental updates, continuous integration and crowd-sourced testing i want to set up automated builds and host all media in the cloud. (if you know what i mean ;-)

btw: the image sort of boots, but it can't find /dev/archiso.

/me off. to be continued.

Dieter

I tested the image and does not boot. Seems that you have an old enviroment, the initramfs is created with new archiso* hooks but is with mkinitcpio-0.5....

fwiw. I built a new image http://build.archlinux.org/isos/archlinux-2010.02.17-netinstall-i686.iso I updated the chroot again (mkinitcpio 0.6.2-1), built the image without having testing enabled, but the problem is the same as last time. (/dev/archiso did not appear in 30s) Dieter

Gerardo Exequiel Pozzi

10:17 p.m.

On 02/17/2010 06:50 PM, Dieter Plaetinck wrote:

...

On Tue, 09 Feb 2010 18:58:14 -0300 Gerardo Exequiel Pozzi wrote:

...
On 02/09/2010 06:28 PM, Dieter Plaetinck wrote:

...
so i blew the dust of our releng chroots, updated them, installed latest archiso from git and so on.

i built this experimental isolinux image (with testing repo enabled!), basically to see if everything still works. http://build.archlinux.org/isos/archlinux-2010.02.09-netinstall-i686.iso

through incremental updates, continuous integration and crowd-sourced testing i want to set up automated builds and host all media in the cloud. (if you know what i mean ;-)

btw: the image sort of boots, but it can't find /dev/archiso.

/me off. to be continued.

Dieter

I tested the image and does not boot. Seems that you have an old enviroment, the initramfs is created with new archiso* hooks but is with mkinitcpio-0.5....

fwiw. I built a new image http://build.archlinux.org/isos/archlinux-2010.02.17-netinstall-i686.iso I updated the chroot again (mkinitcpio 0.6.2-1), built the image without having testing enabled, but the problem is the same as last time. (/dev/archiso did not appear in 30s)

Dieter

Hi, I downloaded and tested. Again the same problem archiso.img is build with old mkinitcpio (not in repos anymore) and new archiso hook. Are you sure that your enviroment is up-to-date? Also root-image.sqfs is build with out-of-date packages. Mirror is up-to-date? -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

18 Feb 18 Feb

8:36 a.m.

On Wed, 17 Feb 2010 19:17:36 -0300 Gerardo Exequiel Pozzi wrote:

...

Hi, I downloaded and tested. Again the same problem archiso.img is build with old mkinitcpio (not in repos anymore) and new archiso hook. Are you sure that your enviroment is up-to-date? Also root-image.sqfs is build with out-of-date packages. Mirror is up-to-date?

Yeah i checked with pacman -Qi in the chroot it was 0.6.2-1. maybe a stale work dir or something i don't know. when i have time i'll try to figure out what went wrong. Dieter

Gerardo Exequiel Pozzi

11:53 a.m.

On 02/18/2010 05:36 AM, Dieter Plaetinck wrote:

...

On Wed, 17 Feb 2010 19:17:36 -0300 Gerardo Exequiel Pozzi wrote:

...
Hi, I downloaded and tested. Again the same problem archiso.img is build with old mkinitcpio (not in repos anymore) and new archiso hook. Are you sure that your enviroment is up-to-date? Also root-image.sqfs is build with out-of-date packages. Mirror is up-to-date?

Yeah i checked with pacman -Qi in the chroot it was 0.6.2-1. maybe a stale work dir or something i don't know. when i have time i'll try to figure out what went wrong.

Dieter

OK Sure, do a "make clean" before do anything, when new packages are on repo ""external change"", because this is not detected by Makefile. The purpose when I rewrote the Makefile, was just for """in the working hour""" avoid unnecessary rebuilds, if none was changed inside ""internal change"" the working space and making it more subdivided in few jobs. More or less like it was my idea at the time. I do not know how to describe in English words. I hope you understand what I mean. :P -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

9:02 p.m.

On Wed, 17 Feb 2010 19:17:36 -0300 Gerardo Exequiel Pozzi wrote:

...

On 02/17/2010 06:50 PM, Dieter Plaetinck wrote:

...

...
fwiw. I built a new image http://build.archlinux.org/isos/archlinux-2010.02.17-netinstall-i686.iso I updated the chroot again (mkinitcpio 0.6.2-1), built the image without having testing enabled, but the problem is the same as last time. (/dev/archiso did not appear in 30s)

Dieter

Hi, I downloaded and tested. Again the same problem archiso.img is build with old mkinitcpio (not in repos anymore) and new archiso hook. Are you sure that your enviroment is up-to-date? Also root-image.sqfs is build with out-of-date packages. Mirror is up-to-date?

okay. the problem was an old work dir. i cleaned it up and rebuilt images and now they boot. new core and netinstall iso's at http://build.archlinux.org/isos/ i tried running `aif -p automatic -c /usr/share/aif/examples/fancy-install-on-sda` but ran into the following issues: 1) all filesystems get created but after the supposedly mounting of them, when i check with df, i see: df: /mnt/boot: no such file or directory df: /mnt/home: no such file or directory a bit later it shows them all of the same size, ie they are all pointing to the same filesystem (/mnt, which is the root LV) 2) transient resolver errors 3) booting fails (no bootable media found) i guess some stuff has changed with device naming, how cryptsetup works or whatever. anyway, that's about all the time i have for today. until next time. Dieter

Dieter Plaetinck

2 Mar 2 Mar

10:50 p.m.

new images @ http://build.archlinux.org/isos/ Using the latest stuff. (mkinitcpio 0.6.3-1 etc) I found out a nasty issue that caused fs'es to be mounted in the wrong order, thus breaking installations. http://projects.archlinux.org/aif.git/commit/?h=experimental if you want to test, be sure to install the latest aif from git experimental branch. ( you can use the aur git package http://aur.archlinux.org/packages.php?ID=22101) in a while i will build a new aif package/new isos with new aif. Dieter

Dieter Plaetinck

5 Mar 5 Mar

3:46 p.m.

x86_64 builds (untested!) http://build.archlinux.org/isos/ i also wrote a changelog and new instructions. next builds will contain a much needed aif update. Dieter

Gerardo Exequiel Pozzi

4:09 p.m.

On 03/05/2010 12:46 PM, Dieter Plaetinck wrote:

...

x86_64 builds (untested!) http://build.archlinux.org/isos/ i also wrote a changelog and new instructions.

next builds will contain a much needed aif update.

Dieter

Great! :) I guess for next test isos should be announced in forum, so can get more feedback from people. Not all users reads this list. Also will be great, but need some coordination with Aaron, to git pull archiso2dual into archiso [#1]. Then you can build dual isos, and at least, put netinstall-dual (220MB) for download. [#1] http://mailman.archlinux.org/pipermail/arch-releng/2010-March/000889.html Good luck! -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Gerardo Exequiel Pozzi

4:44 p.m.

On 03/05/2010 12:46 PM, Dieter Plaetinck wrote:

...

x86_64 builds (untested!) http://build.archlinux.org/isos/ i also wrote a changelog and new instructions.

next builds will contain a much needed aif update.

Dieter

Note that archlinux-2010.03.04-core-x86_64.iso and archlinux-2010.03.02-core-i686.iso are builded with an old archiso git (archiso hook is 3 commits old from current master), also syslinux menu is old. Ensure for next time doing a make clean (syslinux-iso), git pull and make install (archiso). Thanks! -- Gerardo Exequiel Pozzi ( djgera ) http://www.djgera.com.ar KeyID: 0x1B8C330D Key fingerprint = 0CAA D5D4 CD85 4434 A219 76ED 39AB 221B 1B8C 330D

Dieter Plaetinck

6:04 p.m.

Gerardo Exequiel Pozzi wrote:

...

Great! :)

I guess for next test isos should be announced in forum, so can get more feedback from people. Not all users reads this list.

I first want to hear from Aaron/other people who are into traffic/mirror stuff how we will handle the distribution of the images (i.e. we don't want the entire Arch community to get the images from build.archlinux.org). we're currently discussing this in the (private) arch-dev mailing list.

...

Also will be great, but need some coordination with Aaron, to git pull archiso2dual into archiso [#1]. Then you can build dual isos, and at least, put netinstall-dual (220MB) for download.

[#1] http://mailman.archlinux.org/pipermail/arch-releng/2010-March/000889.html

Well, I'm just the installer guy, I have no knowledge of this stuff. I read the dual arch thread http://mailman.archlinux.org/pipermail/arch-releng/2010-February/000836.html and ticket http://bugs.archlinux.org/task/10753 but could not see much testing/feedback. If you get Thomas' blessing or Aarons, it's fine for me. Since your dual-netinstall image is only 220MB, I think the "people will complain about downloadsize" is void and we can ship only the dual-arch images. OTOH since it's trivial to release both dual-arch and single-arch images, we can do that aswel.

...

Note that archlinux-2010.03.04-core-x86_64.iso and archlinux-2010.03.02-core-i686.iso are builded with an old archiso git (archiso hook is 3 commits old from current master), also syslinux menu is old. Ensure for next time doing a make clean (syslinux-iso), git pull and make install (archiso).

will do. Dieter

5165

Age (days ago)

5189

Last active (days ago)

List overview

Download

21 comments

2 participants

participants (2)

Dieter Plaetinck
Gerardo Exequiel Pozzi