- Arch-releng - lists.archlinux.org

[arch-releng] [test iso 2011.01.09] testing results
by Thomas Dziedzic 14 Jan '11

14 Jan '11

I have decided to test the iso since I heard testing is at an all time low. I have filed 3 bugs from my experience so far: https://bugs.archlinux.org/task/22427 FS#22427 - [test iso 2011.01.09] aif fails to ask for mountpoint when creating btrfs partition. This really was the blocker for me because it didn't mount a / partition, so I was unable to install correctly. https://bugs.archlinux.org/task/22428 FS#22428 - [test iso 2011.01.09] aif package selection should be sorted more naturally. This issue is more of a human interface bug. I'm just requesting the reorganization of a list, otherwise, there are no problems with it. https://bugs.archlinux.org/task/22429 FS#22429 - [test iso 2011.01.09] aif during install truncates target list. Though this isn't really necessary, I would like to see the full list of targets rather then the first 3 packages (out of 100) displayed on a single line. My overall experience with the iso was pleasant, though it definitely needs more work on the btrfs side. I would like to thank Dieter and any other devs that are actively working on getting the next iso out the door.

1 0

[arch-releng] mention `pacman -Syu` in official install guide
by Andre "Osku" Schmidt 11 Jan '11

11 Jan '11

hi, just noticed that theres no mention at all about `pacman -Syu` in https://wiki.archlinux.org/index.php/Official_Arch_Linux_Install_Guide and now that we can do `pacman -Syu foobar` i would think it should be the first(ish) command to tell the users? anyone good with words? ;) cheers .andre ps. or is there a reason not to have it mentioned in this guide?

2 1

[arch-releng] from now on: automatic testbuilds + heads-up on changes
by Dieter Plaetinck 09 Jan '11

09 Jan '11

Hi, I want to inform you of two things: 1) new testbuilds available! A whole bunch of changes, both in the actual build environment, as well as AIF itself, where the changes are mostly related to how we automatically export configuration to the target system (locales, networking, initcpio, etc). for users, the main thing to remember is that we now always export the network config to the target system, whereas we used to ask for confirmation before. actually, so much has changed that despite my own tests it would suprise me if i already found/fixed all the regressions. for hackers: the aif code is now more readable :) Please have a look at the changelog: http://releng.archlinux.org/isos/Changelog Known issues: - when selecting source 'net', "install packages" won't realize the step has completed successfully. so you can do only do core installs, but the fix is already in git and will be on the next images. - some kind of kernel/vbox bug wrt virtual memory: http://mailman.archlinux.org/pipermail/arch-general/2011-January/017713.html - btrfs-progs, nilfs-utils and dosfstools packages cannot be installed to the target system (although you should be able to use those FS'es fine) 2) automatic builds: from now on, there will be automatic builds available @ http://releng.archlinux.org/isos/ i'm still playing around with the parameters, probably the builds will be daily or weekly. I would appreciate it if people could test these images and post their findings. Checklist of things which need to be verified. (when you test, you can obviously verify as many things as you want in one run) * dual arch image boot (pick your arch of preference at boot) * i686 image boot * x86_64 image boot * pxe boot * automatic install generic example * automatic install fancy example * net install manual spec (check that it works + rc.conf, resolv.conf, mirrorlist) * net install dhcp (check that it works + rc.conf) * core install * autoprepare (check the installed system, incl fstab) * manual partitioning/FS (with optionally: encryption, nilfs2, lvm,btrfs). (check also fstab, menu.lst, initcpio.conf and crypttab if appropriate) * doing a rollback and re-doing autoprepare or manual config TODO: - some kind of web app /wiki page where users can mark things in the checklist as "working with testbuild $version" ? that would give a good indication of when we can promote testbuilds to official releases - some way to give these more visibility (post to al.org frontpage? can we mirror first?..?) Dieter

2 3

[arch-releng] [PATCH] Define grubmenu as a global variable instead of multiple local variables
by pyther＠pyther.net 08 Jan '11

08 Jan '11

--- src/core/libs/lib-ui-interactive.sh | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/core/libs/lib-ui-interactive.sh b/src/core/libs/lib-ui-interactive.sh index 2297078..11cde02 100644 --- a/src/core/libs/lib-ui-interactive.sh +++ b/src/core/libs/lib-ui-interactive.sh @@ -1,6 +1,9 @@ #!/bin/bash # A library which allows you to do backend stuff by using user interfaces +# Global Variables +grubmenu="$var_TARGET_DIR/boot/grub/menu.lst" + # check if a worker has completed successfully. if not -> tell user he must do it + return 1 # if ok -> don't warn anything and return 0 check_depend () @@ -855,7 +858,6 @@ interactive_install_bootloader () { interactive_grub() { get_grub_map - local grubmenu="$var_TARGET_DIR/boot/grub/menu.lst" [ ! -f $grubmenu ] && show_warning "No grub?" "Error: Couldn't find $grubmenu. Is GRUB installed?" && return 1 debug FS "starting interactive_grub" @@ -995,7 +997,6 @@ interactive_grub() { } generate_grub_menulst() { - local grubmenu="$var_TARGET_DIR/boot/grub/menu.lst" get_device_with_mount '/' || return 1 local _rootpart=$ANSWER_DEVICE @@ -1128,7 +1129,6 @@ EOF } interactive_grub_menulst () { - local grubmenu="$var_TARGET_DIR/boot/grub/menu.lst" generate_grub_menulst helptext= grep -q '^/dev/mapper' $TMP_FSTAB && helptext=" /dev/mapper/ users: Pay attention to the kernel line!" -- 1.7.3.4

3 5

[arch-releng] [PATCH] Call target_configure_network directly instead of worker_auto_network.
by pyther＠pyther.net 02 Jan '11

02 Jan '11

Will call target_configure_network instead of worker_auto_network in preconfigure_target. Calling worker_auto_network during worker_configure_system caused worker_configure_system to never set a return code. Calling target_configure_network follows the same format as other methods in preconfigure_target() worker_auto_network has been removed since no other function calls it. --- src/core/libs/lib-network.sh | 95 +++++++++++++++++++---------------- src/core/libs/lib-ui-interactive.sh | 2 +- src/core/procedures/base | 11 ---- src/core/procedures/interactive | 17 ------ 4 files changed, 52 insertions(+), 73 deletions(-) diff --git a/src/core/libs/lib-network.sh b/src/core/libs/lib-network.sh index bc17898..d374dbf 100644 --- a/src/core/libs/lib-network.sh +++ b/src/core/libs/lib-network.sh @@ -6,48 +6,55 @@ # we assume that you checked whether networking has been setup before calling us target_configure_network() { - source $RUNTIME_DIR/aif-network-settings 2>/dev/null || return 1 - - IFO=${INTERFACE_PREV:-eth0} # old iface: a previously entered one, or the arch default - IFN=${INTERFACE:-eth0} # new iface: a specified one, or the arch default - - # comment out any existing uncommented entries, whether specified by us, or arch defaults. - for var in eth0 $IFO INTERFACES gateway ROUTES - do - sed -i "s/^$var=/#$var=/" ${var_TARGET_DIR}/etc/rc.conf || return 1 - done - sed -i "s/^nameserver/#nameserver/" ${var_TARGET_DIR}/etc/resolv.conf || return 1 - if [ -f ${var_TARGET_DIR}/etc/profile.d/proxy.sh ] - then - sed -i "s/^export/#export/" ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 - fi - - if [ "$DHCP" = 0 ] ; then - local line="$IFN=\"$IFN ${IPADDR:-192.168.0.2} netmask ${SUBNET:-255.255.255.0} broadcast ${BROADCAST:-192.168.0.255}\"" - append_after_last "/$IFO\|eth0/" "$line" ${var_TARGET_DIR}/etc/rc.conf || return 1 - - if [ -n "$GW" ]; then - append_after_last "/gateway/" "gateway=\"default gw $GW\"" ${var_TARGET_DIR}/etc/rc.conf || return 1 - append_after_last "/ROUTES/" "ROUTES=(gateway)" ${var_TARGET_DIR}/etc/rc.conf || return 1 - fi - if [ -n "$DNS" ] - then - echo "nameserver $DNS" >> ${var_TARGET_DIR}/etc/resolv.conf || return 1 - fi - else - append_after_last "/$IFO\|eth0/" "$IFN=\"dhcp\"" ${var_TARGET_DIR}/etc/rc.conf || return 1 - fi - - append_after_last "/$IFO\|eth0/" "INTERFACES=($IFN)" ${var_TARGET_DIR}/etc/rc.conf || return 1 - - if [ -n "$PROXY_HTTP" ]; then - echo "export http_proxy=$PROXY_HTTP" >> ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 - chmod a+x ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 - fi - - if [ -n "$PROXY_FTP" ]; then - echo "export ftp_proxy=$PROXY_FTP" >> ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 - chmod a+x ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 - fi - return 0 + if [ -f $RUNTIME_DIR/aif-network-settings ]; then + + debug NETWORK "Configuring network settings on target system according to installer settings" + + source $RUNTIME_DIR/aif-network-settings 2>/dev/null || return 1 + + IFO=${INTERFACE_PREV:-eth0} # old iface: a previously entered one, or the arch default + IFN=${INTERFACE:-eth0} # new iface: a specified one, or the arch default + + # comment out any existing uncommented entries, whether specified by us, or arch defaults. + for var in eth0 $IFO INTERFACES gateway ROUTES + do + sed -i "s/^$var=/#$var=/" ${var_TARGET_DIR}/etc/rc.conf || return 1 + done + sed -i "s/^nameserver/#nameserver/" ${var_TARGET_DIR}/etc/resolv.conf || return 1 + if [ -f ${var_TARGET_DIR}/etc/profile.d/proxy.sh ] + then + sed -i "s/^export/#export/" ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 + fi + + if [ "$DHCP" = 0 ] ; then + local line="$IFN=\"$IFN ${IPADDR:-192.168.0.2} netmask ${SUBNET:-255.255.255.0} broadcast ${BROADCAST:-192.168.0.255}\"" + append_after_last "/$IFO\|eth0/" "$line" ${var_TARGET_DIR}/etc/rc.conf || return 1 + + if [ -n "$GW" ]; then + append_after_last "/gateway/" "gateway=\"default gw $GW\"" ${var_TARGET_DIR}/etc/rc.conf || return 1 + append_after_last "/ROUTES/" "ROUTES=(gateway)" ${var_TARGET_DIR}/etc/rc.conf || return 1 + fi + if [ -n "$DNS" ] + then + echo "nameserver $DNS" >> ${var_TARGET_DIR}/etc/resolv.conf || return 1 + fi + else + append_after_last "/$IFO\|eth0/" "$IFN=\"dhcp\"" ${var_TARGET_DIR}/etc/rc.conf || return 1 + fi + + append_after_last "/$IFO\|eth0/" "INTERFACES=($IFN)" ${var_TARGET_DIR}/etc/rc.conf || return 1 + + if [ -n "$PROXY_HTTP" ]; then + echo "export http_proxy=$PROXY_HTTP" >> ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 + chmod a+x ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 + fi + + if [ -n "$PROXY_FTP" ]; then + echo "export ftp_proxy=$PROXY_FTP" >> ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 + chmod a+x ${var_TARGET_DIR}/etc/profile.d/proxy.sh || return 1 + fi + else + debug NETWORK "Skipping Host Network Configuration - aif-network-settings not found" + fi + return 0 } diff --git a/src/core/libs/lib-ui-interactive.sh b/src/core/libs/lib-ui-interactive.sh index 19a0cbb..6175689 100644 --- a/src/core/libs/lib-ui-interactive.sh +++ b/src/core/libs/lib-ui-interactive.sh @@ -23,7 +23,7 @@ preconfigure_target () { target_configure_initial_locale || failed+=('initial locale') target_configure_initial_keymap_font || failed+=('keymap/font setting') target_configure_fstab || failed+=('fstab configuration') - execute worker auto_network || failed+=('network config export') + target_configure_network || failed+=('network config export') target_configure_mirrorlist || failed+=('mirrorlist configuration') target_configure_time || failed+=('time configuration') target_configure_initcpio || failed+=('initcpio configuration') diff --git a/src/core/procedures/base b/src/core/procedures/base index b6f0ea4..ab95cea 100644 --- a/src/core/procedures/base +++ b/src/core/procedures/base @@ -142,17 +142,6 @@ worker_install_packages () installpkg } - -worker_auto_network () -{ - if [ -f $RUNTIME_DIR/aif-network-settings ] - then - target_configure_network || return $? - fi - return 0 -} - - worker_configure_system () { preconfigure_target diff --git a/src/core/procedures/interactive b/src/core/procedures/interactive index 179866c..f499a4b 100644 --- a/src/core/procedures/interactive +++ b/src/core/procedures/interactive @@ -167,23 +167,6 @@ worker_install_bootloader () interactive_install_bootloader } - -worker_auto_network () -{ - # if the user has been through networking setup, it may be useful to export the users' choices to the target system - # networking setup could have happened in a separate process (eg partial-configure-network), so check if the settings file was created to be sure - if [ -f $RUNTIME_DIR/aif-network-settings ] - then - ask_yesno "Do you want to use the network settings from the installer in rc.conf and resolv.conf?\n\nIf you used Proxy settings, they will be written to /etc/profile.d/proxy.sh" || return 0 - if ! target_configure_network - then - show_warning "Automatic network settings propagation failed" "Failed to import current network settings into target system" - return 1 - fi - fi - return 0 -} - # override show_report because this procedure does not use phases, but it's own menu show_report () { -- 1.7.3.4

2 1

[arch-releng] [PATCH 0/1] *** Call target_configure_network instead of worker_auto_netwok ***
by pyther＠pyther.net 02 Jan '11

02 Jan '11

*** Call target_configure_network instead of worker_auto_network. We may want to notify or allow the user to edit /etc/profile.d/proxy.sh since target_configure_network will edit the file if proxy setting were used. Calling worker_auto_network during worker_configure_system caused worker_configure_system to never have a return code set. Also worker_auto_network has been removed since no function will call it.*** Matthew Gyurgyik (1): Call target_configure_network lib-ui-interactive.sh instead of worker_auto_network. Removed worker_auto_network

2 2

[arch-releng] [PATCH 0/2] *** Rename mdraid_is-raid and mdraid_all-slaves ***
by pyther＠pyther.net 02 Jan '11

02 Jan '11

*** Renamed mdraid_is-raid and mdraid_all-slaves to mdraid_is_raid and mdraid_all_slaves *** Matthew Gyurgyik (2): change mdraid_is-raid to mdraid_is_raid change mdraid_all-slaves to mdraid_all_slaves src/core/libs/lib-blockdevices-filesystems.sh | 4 ++-- src/core/libs/lib-ui-interactive.sh | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) -- 1.7.3.4

2 3

[arch-releng] away mode on
by Gerardo Exequiel Pozzi 29 Dec '10

29 Dec '10

Hi I am going to take some days of vacation and detox of electronic devices from 01/01 to 15/01. PS: I am a bit out from here in these days because lots of problems with electrical supply. :( Good year \forall -- Gerardo Exequiel Pozzi \cos^2\alpha + \sin^2\alpha = 1

2 1

[arch-releng] releng build environment moved
by Dieter Plaetinck 23 Dec '10

23 Dec '10

Hi, we just moved the releng image building stuff from sigurd.al.org to alberich.al.org To clarify: Moved: - image generation environment - http serving of testbuilds, releng packages, releng repos Untouched: - hosting of official images (http://www.archlinux.org/download/) - Torrent tracker (tracker.al.org) - releng mailing list The new machine is the VPS donated by airVM, see https://wiki.archlinux.org/index.php/Category:DeveloperWiki:Server_Configur… Two reasons for this move: - releng load patterns affected the AUR/community repos, running on sigurd, also disk space was getting tight - I wanted a separate env to perform all the critical releng stuff without worrying too much I might break some production service Folks following releng stuff, note: - build.archlinux.org has been deprecated, the new url is releng.archlinux.org - the same stuff (testbuilds, packages, etc) are available on the new location Releng TODO: - update scripts so we can fully automatically build testing images - update scripts to update tracker on sigurd - schedule automatic builds Dieter

1 0

[arch-releng] btrfs support in aif
by Dieter Plaetinck 21 Dec '10

21 Dec '10

Anthony and other interested folks, I've been looking a bit further, and it seems like btrfs support shouldn't be too hard to implement. It actually seems simpler then LVM (because lvm has 3 levels: PV,VG and LV; btrfs has just the btrfs itself (~default subvolume) and other subvolumes) subvolumes don't get a new devicefile but i'll probably use something like: /dev/sda:$spec to denote what's what. if $spec is a number, it will become mount option subvolid=$spec; otherwise subvol=$spec although from what i can tell, the id's aren't used often. and it seems more robust to me use names anyway. * which are the requirements your btrfs_advanced mkinitcpio hook implies? what things does aif need to do other then just doing mkfs.btrfs to get the full potential out of btrfs/your hook? please explain why a default btrfs configuration does not suffice. does it have something to do with https://btrfs.wiki.kernel.org/index.php/UseCases#Can_a_snapshot_be_replaced… ? * I've read a bit more about btrfs and I think an implementation like this will suffice for most users: - allow creation of a btrfs on top of 1-n blockdevices (user can pick raid levels for data and metadata) - allow creation of 0-m subvolumes - each subvolume as well as the default can get an arbitrary mountpoint, as well as specific mount options like compress, ssd, etc. if i understood correctly, that is. However, to be fully compatible with your hook, I will probably "strongly recommend" to create a subvolume __active and mount that as / Right? anything I missed? Dieter

2 7