- Arch-releng - lists.archlinux.org

[arch-releng] Providing meta information for iso releases
by Pierre Schmitz 28 Jan '11

28 Jan '11

Hi engineers, while working on my archportal project I wondered if it wouldn't be nice to have some meta information about our install images. The purpose of this would be automaticall generating the download page, README files, atom feeds etc.. What do you think about this idea in general? 1) Which information could we provide? * release date/time * release name * kernel version * changelog * maybe some readme/instructions * for each iso image: * file name * short description (e.g. netinstall iso image) * description * architectures (e.g. arch=(i686 x86_64) * size * checksums (md5, sha1) * signature (when package signing available) 2) Which format to use? The easiest was is probably to create something similar to the db files for packages. That means several text files and directories within a tar file. For example some file with common information about this release in the root dir and specific information for each iso within subdir named like the iso file names. 3) How to create it? We'll need a simple way to create these db files automatically with every release. All information except descriptions can be filled in automatically. So what do you guys think about this? Greetings, Pierre -- Pierre Schmitz, https://users.archlinux.de/~pierre

2 1

[arch-releng] [testiso 2011.01.24] test experience
by Thomas Dziedzic 27 Jan '11

27 Jan '11

Hey there, tested another install iso: source: netinstall in virtualbox using eth0 with dhcp mirror: http://mirrors.kernel.org/ editor: vi time: chicago, with utc hwclock partitions: /boot 200mb; swap 200mb; / rest filesystems: /boot ext2; swap swap; / ext4 package selection: default base group with sudo installation: failed to download grub, didn't move on, retried selecting install, only had to grab the package that failed to download. Everything continued as expected. NICE! configs: left them vanilla except changing the hostname package selection still suffers from https://bugs.archlinux.org/task/22428 FS#22428 - [test iso 2011.01.09] aif package selection should be sorted more naturally. I didn't try btrfs so I can't comment on FS#22427 - [test iso 2011.01.09] aif fails to ask for mountpoint when creating btrfs partition. Overall this installation went without a hitch and I would say that this iso is ready for an audience who doesn't use btrfs. The only usability annoyance I encountered was again, FS#22428 which made it rather difficult to find something like sudo. Thanks for the great work to Dieter and everyone involved! Hope this feedback helps you guys.

2 1

Re: [arch-releng] nullglob
by Dieter Plaetinck 26 Jan '11

26 Jan '11

On Mon, 24 Jan 2011 17:22:18 -0500 Matthew Gyurgyik <pyther(a)pyther.net> wrote: > Falconindy was right. nullglob affects everything just not the > function. I simply fixed it by adding shopt -u nullglob at the end of > the function. > > I rebased and forced a push therefore there is no "new" commit: > https://github.com/pyther/aif/commit/43d02dc58a63599ff67d3a7e9f89c8b81831b5… > > I don't think it is worth it to resend the patch yet again for one > line of code, though let me know if that is what you want me to do. can you be sure that nullglob must be unset after the function? Could it be that nullglob was already set in the beginning? > > Here is the link for "Moved kernel parameter generation into > get_kernel_parameters" > https://github.com/pyther/aif/commit/2265ae387d898e68cc0bfa964d59fcf3737ba2… > let me know you want me to send a patch to the list for that. I think you're missing the point of the mailing list. on the mailing list, there are multiple people interested in the releng code. Multiple people who can review your stuff and comment it. It's pointless to put all the work on my shoulders, and I only when I say it looks good, to send it to the ML. It's ok to get in touch with me for very specific questions, but for stuff like this: Send it to ML right away, it's ok if there are still problems in it, we can discuss it on the ML. Dieter

1 1

[arch-releng] Use xz for squashfs with kernel 2.6.38
by Sven-Hendrik Haase 26 Jan '11

26 Jan '11

I wondered whether we are going to use xz for squashfs compression in the upcoming kernel 2.6.38? -- Sven-Hendrik

4 4

[arch-releng] [PATCH 2/2] Use globbing and parameter expansion to get the raid slave devices
by pyther＠pyther.net 23 Jan '11

23 Jan '11

Uses bash globbing and parameter expansion to find all of the slaves for a raid device. This is a much better method then using ls. Also, by looking at DEVNAME in the uevent file we provide support for block device that are not in the root of /dev. call die_error if a block device is not passed into device_is_raid() --- src/core/libs/lib-blockdevices-filesystems.sh | 12 ++++++++---- 1 files changed, 8 insertions(+), 4 deletions(-) diff --git a/src/core/libs/lib-blockdevices-filesystems.sh b/src/core/libs/lib-blockdevices-filesystems.sh index 8451f15..f388d5b 100644 --- a/src/core/libs/lib-blockdevices-filesystems.sh +++ b/src/core/libs/lib-blockdevices-filesystems.sh @@ -921,7 +921,8 @@ get_blockdevice_size () # $1 blockdevice (ex: /dev/md0 or /dev/sda1) # All MD RAID block devices have a major id of 9 device_is_raid() { - [[ $1 && -f /proc/mdstat ]] || return 1 + [[ -b "$1" ]] || die_error "device_is_raid needs a blockdevice as \$1 ($1 given)" + [[ -f /proc/mdstast ]] || return 1 local devmajor=$(stat -c %t "$1") (( devmajor == 9 )) } @@ -942,10 +943,13 @@ mdraid_slave0 () # ex: /dev/md0 has slaves: "/dev/sda1 /dev/sdb2 /dev/sdc2" mdraid_all_slaves () { + shopt -s nullglob local slave= - local slaves= - for slave in $(ls /sys/class/block/$(basename $1)/slaves/); do - slaves=$slaves"/dev/"$slave" " + local slaves= + for slave in /sys/class/block/${1##*/}/slaves/*; do + source "$slave/uevent" + slaves="$slaves/dev/$DEVNAME " + unset DEVNAME done echo $slaves } -- 1.7.3.5

1 0

[arch-releng] [PATCH 1/2] Renamed mdraid_is_raid to device_is_raid
by pyther＠pyther.net 23 Jan '11

23 Jan '11

device_is_raid now uses dev major number to determine if a device is a raid device. A major ID of 8 is a /dev/sd? device whereas a major ID of 9 is a 'metadata' (raid) device. --- src/core/libs/lib-blockdevices-filesystems.sh | 20 ++++++-------------- src/core/libs/lib-ui-interactive.sh | 2 +- 2 files changed, 7 insertions(+), 15 deletions(-) diff --git a/src/core/libs/lib-blockdevices-filesystems.sh b/src/core/libs/lib-blockdevices-filesystems.sh index 8573b19..8451f15 100644 --- a/src/core/libs/lib-blockdevices-filesystems.sh +++ b/src/core/libs/lib-blockdevices-filesystems.sh @@ -919,20 +919,12 @@ get_blockdevice_size () # $1 blockdevice (ex: /dev/md0 or /dev/sda1) -# return true when blockdevice is an md raid, otherwise return a unset value -mdraid_is_raid() -{ - local israid - if [ -z $1 ]; then - # Don't call mdadm on empty blockdevice parameter! - israid="" - elif [ "$(mdadm --query $1 | cut -d':' -f2)" == " is not an md array" ]; then - israid="" - else - israid=true - fi - echo $israid -} +# All MD RAID block devices have a major id of 9 +device_is_raid() { + [[ $1 && -f /proc/mdstat ]] || return 1 + local devmajor=$(stat -c %t "$1") + (( devmajor == 9 )) +} # $1 md raid blockdevice (ex: /dev/md0) # return the array member device which is slave 0 in the given array diff --git a/src/core/libs/lib-ui-interactive.sh b/src/core/libs/lib-ui-interactive.sh index 881af90..829556a 100644 --- a/src/core/libs/lib-ui-interactive.sh +++ b/src/core/libs/lib-ui-interactive.sh @@ -878,7 +878,7 @@ interactive_grub() { bootdev=$(mount | grep $var_TARGET_DIR/boot | cut -d' ' -f 1) # check if bootdev or PART_ROOT is on a md raid array # This dialog is only shown when we detect / or /boot on a raid device. - if [ -n "$(mdraid_is_raid $bootdev)" -o -n "$(mdraid_is_raid $PART_ROOT)" ]; then + if device_is_raid $bootdev || device_is_raid $PART_ROOT; then ask_yesno "Do you have your system installed on software raid?\nAnswer 'YES' to install grub to another hard disk." no if [ $? -eq 0 ]; then onraid=true -- 1.7.3.5

1 0

[arch-releng] [PATCH] Renamed mdraid_is_raid to device_is_raid; new mdraid_all_slaves function
by pyther＠pyther.net 23 Jan '11

23 Jan '11

Renamed mdraid_is_raid to device_is_raid falconindy rewrote the device_is_raid and mdraid_all_slaves functions device_is_raid now uses dev major number to determine if a device is a raid device. A major ID of 8 is a /dev/sd? device whereas a major ID of 9 is a 'metadata' (raid) device. mdraid_all_slaves now uses a parameter expansion, reads the uevents file to get the DEVNAME and appends /dev to DEVNAME --- src/core/libs/lib-blockdevices-filesystems.sh | 31 +++++++++---------------- src/core/libs/lib-ui-interactive.sh | 2 +- 2 files changed, 12 insertions(+), 21 deletions(-) diff --git a/src/core/libs/lib-blockdevices-filesystems.sh b/src/core/libs/lib-blockdevices-filesystems.sh index 8573b19..2bb0a74 100644 --- a/src/core/libs/lib-blockdevices-filesystems.sh +++ b/src/core/libs/lib-blockdevices-filesystems.sh @@ -920,19 +920,11 @@ get_blockdevice_size () # $1 blockdevice (ex: /dev/md0 or /dev/sda1) # return true when blockdevice is an md raid, otherwise return a unset value -mdraid_is_raid() -{ - local israid - if [ -z $1 ]; then - # Don't call mdadm on empty blockdevice parameter! - israid="" - elif [ "$(mdadm --query $1 | cut -d':' -f2)" == " is not an md array" ]; then - israid="" - else - israid=true - fi - echo $israid -} +device_is_raid() { + [[ $1 && -f /proc/mdstat ]] || return 1 + local devmajor=$(stat -c %t "$1") + (( devmajor == 9 )) +} # $1 md raid blockdevice (ex: /dev/md0) # return the array member device which is slave 0 in the given array @@ -948,12 +940,11 @@ mdraid_slave0 () # $1 md raid blockdevice (ex: /dev/md0) # return a list of array members from given md array # ex: /dev/md0 has slaves: "/dev/sda1 /dev/sdb2 /dev/sdc2" -mdraid_all_slaves () -{ - local slave= - local slaves= - for slave in $(ls /sys/class/block/$(basename $1)/slaves/); do - slaves=$slaves"/dev/"$slave" " +mdraid_all_slaves () { + shopt -s nullglob + for slave in /sys/class/block/${1##*/}/slaves/*; do + eval $(cat $slave/uevent) + echo /dev/$DEVNAME + unset DEVNAME done - echo $slaves } diff --git a/src/core/libs/lib-ui-interactive.sh b/src/core/libs/lib-ui-interactive.sh index 881af90..829556a 100644 --- a/src/core/libs/lib-ui-interactive.sh +++ b/src/core/libs/lib-ui-interactive.sh @@ -878,7 +878,7 @@ interactive_grub() { bootdev=$(mount | grep $var_TARGET_DIR/boot | cut -d' ' -f 1) # check if bootdev or PART_ROOT is on a md raid array # This dialog is only shown when we detect / or /boot on a raid device. - if [ -n "$(mdraid_is_raid $bootdev)" -o -n "$(mdraid_is_raid $PART_ROOT)" ]; then + if device_is_raid $bootdev || device_is_raid $PART_ROOT; then ask_yesno "Do you have your system installed on software raid?\nAnswer 'YES' to install grub to another hard disk." no if [ $? -eq 0 ]; then onraid=true -- 1.7.3.5

3 2

[arch-releng] [PATCH] Renamed mdraid_is_raid, new device_is_raid mdraid_all_slaves function
by pyther＠pyther.net 21 Jan '11

21 Jan '11

Renamed mdraid_is_raid to device_is_raid falconindy rewrote the device_is_raid and mdraid_all_slaves functions device_is_raid now uses dev major number to determine if a device is a raid device. A major ID of 8 is a /dev/sd? device whereas a major ID of 9 is a 'metadata' (raid) device. mdraid_all_slaves now uses a parameter expansion to get the basename of the device --- src/core/libs/lib-blockdevices-filesystems.sh | 30 +++++++----------------- src/core/libs/lib-ui-interactive.sh | 2 +- 2 files changed, 10 insertions(+), 22 deletions(-) diff --git a/src/core/libs/lib-blockdevices-filesystems.sh b/src/core/libs/lib-blockdevices-filesystems.sh index 8573b19..a0ca15f 100644 --- a/src/core/libs/lib-blockdevices-filesystems.sh +++ b/src/core/libs/lib-blockdevices-filesystems.sh @@ -920,19 +920,11 @@ get_blockdevice_size () # $1 blockdevice (ex: /dev/md0 or /dev/sda1) # return true when blockdevice is an md raid, otherwise return a unset value -mdraid_is_raid() -{ - local israid - if [ -z $1 ]; then - # Don't call mdadm on empty blockdevice parameter! - israid="" - elif [ "$(mdadm --query $1 | cut -d':' -f2)" == " is not an md array" ]; then - israid="" - else - israid=true - fi - echo $israid -} +device_is_raid() { + [[ $1 && -f /proc/mdadm ]] || return 1 + local devmajor=$(stat -c %t "$1") + (( devmajor == 9 )) +} # $1 md raid blockdevice (ex: /dev/md0) # return the array member device which is slave 0 in the given array @@ -948,12 +940,8 @@ mdraid_slave0 () # $1 md raid blockdevice (ex: /dev/md0) # return a list of array members from given md array # ex: /dev/md0 has slaves: "/dev/sda1 /dev/sdb2 /dev/sdc2" -mdraid_all_slaves () -{ - local slave= - local slaves= - for slave in $(ls /sys/class/block/$(basename $1)/slaves/); do - slaves=$slaves"/dev/"$slave" " - done - echo $slaves +mdraid_all_slaves () { + shopt -s nullglob + local slaves=(/sys/class/block/${1##*/}/slaves/*) + echo "${slaves[@]##*/}" } diff --git a/src/core/libs/lib-ui-interactive.sh b/src/core/libs/lib-ui-interactive.sh index 881af90..41c035b 100644 --- a/src/core/libs/lib-ui-interactive.sh +++ b/src/core/libs/lib-ui-interactive.sh @@ -878,7 +878,7 @@ interactive_grub() { bootdev=$(mount | grep $var_TARGET_DIR/boot | cut -d' ' -f 1) # check if bootdev or PART_ROOT is on a md raid array # This dialog is only shown when we detect / or /boot on a raid device. - if [ -n "$(mdraid_is_raid $bootdev)" -o -n "$(mdraid_is_raid $PART_ROOT)" ]; then + if [ -n "$(device_is_raid $bootdev)" -o -n "$(device_is_raid $PART_ROOT)" ]; then ask_yesno "Do you have your system installed on software raid?\nAnswer 'YES' to install grub to another hard disk." no if [ $? -eq 0 ]; then onraid=true -- 1.7.3.5

2 1

Re: [arch-releng] Playing with unionfs-fuse (was: Status of AUFS2.1)
by Gerardo Exequiel Pozzi 18 Jan '11

18 Jan '11

On 01/17/2011 07:18 PM, Thomas Bächler wrote: > Am 17.01.2011 20:40, schrieb Tobias Powalowski: >> Hi guys, >> i'm getting tired of building aufs2 into the standard kernel, >> there are no stable snapshots, everytime you need grab patches and at the >> moment the utils don't even build without any real reason. >> Can we ditch aufs2 support? >> Does archiso really need this filesystem? >> >> Thanks for clarification, i would be happy if we can remove this. > We need a unioning file system. But I don't we think we actually need > the utils. As far as I know, aufs is the most feature-complete and > stable unioning file system. We could maintain an extra kernel only for > the live CD if that is easier. > I am currently playing with unionfs-fuse and archiso, so no need for extra modules. Two issues: 1) There is no "append" feature like with aufs, so all branchs should be mounted in one step. 2) Since is an userspace solution and there is no support for "omit_pid" [#1] in initscripts at the time, shutdown and reboot fails. Notes: 1) I never used it so I do not know how stable is. 2) No patches at this moment I just experimenting from ramfs# command line. [#1] https://bugs.archlinux.org/task/10536 PS: I am back from vacation :) -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1

1 1

[arch-releng] Syslinux Installer / Update Script - Testers Needed
by Matthew Gyurgyik 16 Jan '11

16 Jan '11

Hello Community, Over the last few weeks I have been working on Syslinux support for the installer. With the help Thomas and Dieter I am nearing the completion of this project. As part of this project, I have written a script that will help install and update Syslinux (similar to that of grub-install). Some key features of the script: syslinux-install_update.sh * Install Syslinux to the FS + Partition Boot Loader (extlinux --install /boot/syslinux) * Install Syslinux MBR * Detect and optionally set the boot flag on the boot partition * Update Syslinux – copy files and execute (extilnux --update /boot/syslinux) * Support for GPT disks * Support for RAID configurations The goal is to include this script in the official Syslinux package. Therefore we need your help to test it. syslinux-install_update.sh -i -a -m ..... install Syslinux, set the boot flag (if needed), and install the MBR We need tests for the following setups: / + /boot on the *same* partition / + /boot on the *same* partition - RAID /boot + root on *separate* partition /boot + root on *separate* partition - RAID All of the above using but using the GPT partition layout NOTE: This is an alpha/beta stage script. The script modifies the first 440 bytes of the disk (using dd) and the partition table (using either sfdisk or sgdisk). Although the script should be safe to run, I am not responsible for any data loss that may occur. Let us know the following: * Did the script work for you? * What was your partition setup? (see above) * What version did you use? * If the script did not work, please provide as much information as possible Get the script here: https://gist.github.com/772138 Syslinux Sample Config File: http://projects.archlinux.org/svntogit/packages.git/plain/syslinux/trunk/sy… The Syslinux package in testing includes the above configuration file. Cheers, pyther

3 2