July 2009 - Anaconda-devel - Fedora mailing-lists

[PATCH 0/5] fix zFCP in anaconda and also make it work with changed sysfs interface of device driver (#510524)
by Steffen Maier 18 Aug '09

18 Aug '09

Steffen Maier (5): correctly activate zFCP LUN on s390 correctly delete a SCSI device provided by a zFCP LUN on s390 correctly deactivate zFCP LUN on s390 error messages of zFCP on s390: log or pass to the UI prevent getting started up or shutdown again while already in such state kickstart.py | 5 +- storage/zfcp.py | 177 ++++++++++++++++++++++++++++++++++++++++------- textw/partition_text.py | 5 +- 3 files changed, 161 insertions(+), 26 deletions(-)

2 19

[PATCH 1/2] Don't populate flags.cmdline with "True" values when no = is used.
by Peter Jones 12 Aug '09

12 Aug '09

Since all tests are either flags.cmdline.has_key() or "if flags.cmdline.get()", we want this to be None when there's no equals sign, rather than True. --- flags.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/flags.py b/flags.py index 8dbcc9f..d49b252 100644 --- a/flags.py +++ b/flags.py @@ -59,7 +59,7 @@ class Flags: (key, val) = i.split("=", 1) except: key = i - val = True + val = None cmdlineDict[key] = val -- 1.6.2.2

3 3

[PATCH] Update linuxrc.s390 and friends to reflect review comments.
by Steffen Maier 10 Aug '09

10 Aug '09

This brings linuxrc.s390, lsznet.raw, and controlunits.sh to the latest version as posted here https://www.redhat.com/archives/anaconda-devel-list/2009-February/msg00392.… --- loader/controlunits.sh | 54 +++ loader/linuxrc.s390 | 1165 ++++++++++++++++++++++-------------------------- loader/lsznet.raw | 89 +--- 3 files changed, 593 insertions(+), 715 deletions(-) create mode 100644 loader/controlunits.sh mode change 100644 => 100755 loader/linuxrc.s390 mode change 100644 => 100755 loader/lsznet.raw diff --git a/loader/controlunits.sh b/loader/controlunits.sh new file mode 100644 index 0000000..d45386d --- /dev/null +++ b/loader/controlunits.sh @@ -0,0 +1,54 @@ +# controlunits.sh: define some common control unit mappings +# Copyright (C) IBM Corp. 2009 +# Author: Steffen Maier <maier(a)de.ibm.com> +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; version 2 of the License only. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + +# The arrays (among other things) should be adapted, if any of those device +# drivers start supporting different CU types/models. + +readonly -a CU=( + 1731/01 + 1731/05 + 3088/08 + 3088/1f + 3088/1e + 3088/01 + 3088/60 +) + +readonly -a CU_DEVDRV=( + qeth + qeth + ctcm + ctcm + ctcm + lcs + lcs +) + +# Searches for a match of argument 1 on the array $CU and sets $cu_idx +# to the matched array index on success. +# Returns 0 on success, 1 on failure. +function search_cu() { + local scu=$1 + local i + for ((i=0; i < ${#CU[@]}; i++)); do + if [ "$scu" == "${CU[i]}" ]; then + cu_idx=$i + return 0 + fi + done + return 1 +} diff --git a/loader/linuxrc.s390 b/loader/linuxrc.s390 old mode 100644 new mode 100755 index d9ccca6..d0dfedb --- a/loader/linuxrc.s390 +++ b/loader/linuxrc.s390 @@ -27,16 +27,22 @@ # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # -# prerequisites of this script to run inside the RHEL5.3 installer initrd: -# - copy udevsettle and udevd to initrd -# - copy lsznet.raw to initrd as /sbin/lsznet -# - create /etc/udev/udev.conf with at least one comment line as content +# prerequisites of this script to run inside the installer initrd: +# - udevadm and udevd need to be there +# - have /etc/udev/udev.conf with at least one comment line as content +# - if necessary, have udev rules +# - copy lsznet.raw and controlunits.sh to initrd +# - pack kernel modules and module-init-tools (no longer use busybox for that) +# - "multi on" in /etc/host.conf [RH bugs 486457,486461,483244] + +# TODOs: +# - make sure driver modules get loaded automatically +# - modprobe.conf entry for loading dasd_mod without setting any DASD online +# - udev rule for lcs/ctcm vs. cu3088 +# - make sure netiucv gets loaded even without udev events since on no bus VERSION=1.2 -export TEXTDOMAIN=s390installer -export TEXTDOMAINDIR=/usr/lib/locale - # helper function to execute command in arguments and print command on stdout function debug() { # uncomment the following echo "$*" to enable debug output @@ -108,7 +114,7 @@ function doreboot() exit 0 fi - echo $"about to exec shutdown -r" + echo "about to exec shutdown -r" exec /sbin/shutdown -r exit 0 } @@ -131,19 +137,18 @@ function sysecho () { function startinetd() { echo - echo $"Starting telnetd and sshd to allow login over the network." + echo "Starting sshd to allow login over the network." if [ -z "$testing" ]; then - echo $"Welcome to the anaconda install environment $VERSION for $S390ARCH" > /etc/issue.net - echo $"Welcome to the anaconda install environment $VERSION for $S390ARCH" > /etc/motd + echo "Welcome to the anaconda install environment $VERSION for $S390ARCH" > /etc/issue.net + echo "Welcome to the anaconda install environment $VERSION for $S390ARCH" > /etc/motd echo >> /etc/motd fi # testing - /sbin/xinetd -stayalive -reuse -pidfile /tmp/xinetd.pid /sbin/sshd if [ -z "$RUNKS" ]; then echo - echo $"Connect now to $IPADDR and login as user root to start the installation." - echo $"E.g. using: ssh -X root@$IPADDR" + echo "Connect now to $IPADDR and login as user root to start the installation." + echo "E.g. using: ssh -X root@$IPADDR" read while : ; do /bin/sh --login @@ -157,33 +162,42 @@ function readcmsfile() # $1=dasdport $2=filename { local dev if [ $# -ne 2 ]; then return; fi - mknod /dev/dasda b 94 0 + # precondition: udevd created dasda block device node if ! sysecho /proc/cio_ignore "free $1"; then - echo $"DASD $1 could not be cleared from device blacklist" + echo "DASD $1 could not be cleared from device blacklist" return 1 fi # /proc/cio_ignore won't block on freeing devices until resensing # has been completed, so wait until the udev event queue depletes - # (without udevsettle we could wait 2 seconds unconditionally) + # (without udevadm settle we could wait 2 seconds unconditionally) #debug ls -laF /dev/.udev - udevsettle + udevadm settle + # even though the device might now be online, some of its + # sysfs attributes might not yet be available sleep 1 - insmod dasd_mod$LO dasd=$1 - insmod dasd_eckd_mod$LO - if ! cmsfscat -d /dev/dasda -a $2 > /tmp/$2; then - echo $"Could not read conf file $2 on CMS DASD $1." - fi - # maybe replace with a more robust: dev=$(printf "0.0.%04x" $((0x${1}))) + # precondition: dasd_eckd_mod driver incl. dependencies loaded, + # dasd_mod must be loaded without setting any DASD online if [ ${#1} == 3 ]; then dev="0.0.0${1}" elif [ ${#1} == 4 ]; then dev="0.0.${1}" fi - sysecho /sys/bus/ccw/drivers/dasd-eckd/$dev/online 0 - rmmod dasd_eckd_mod - rmmod dasd_mod + if ! sysecho /sys/bus/ccw/devices/$dev/online 1; then + echo "DASD $dev could not be set online" + return 1 + fi + if ! cmsfscat -d /dev/dasda -a $2 > /tmp/$2; then + echo "Could not read conf file $2 on CMS DASD $1." + fi + if ! sysecho /sys/bus/ccw/devices/$dev/online 0; then + echo "DASD $dev could not be set offline again" + return 1 + fi + # consequences of no more module unload: loader/anaconda can no longer + # use DASD option to online DASDs and set other DASD parameters! } +# adaption of the same function in init.c (udevd gets started later) function createDevices() { awk '{ printf("mknod /dev/%s %s %s %s\n", $1, $2, $3, $4); @@ -202,6 +216,7 @@ random c 1 8 644 root:root urandom c 1 9 644 root:root rtc c 10 135 644 root:root EOF + # tty handling is different from init.c since s390 does not have all for i in 2 3 4 5 6 7 8 9 ; do ln -s console /dev/tty$i done @@ -209,6 +224,70 @@ EOF ln -s /proc/self/fd /dev/fd } +# approximately the main() function of init.c +function init_main() { + S390ARCH=$(uname -m) + if [ "$S390ARCH" = "s390" ]; then + export S390ARCH="S/390" + else + export S390ARCH="zSeries" + fi + + echo + echo "Starting the $S390ARCH initrd to configure networking. Version is $VERSION" + + # set up env vars as we do in init.c + if [ $(uname -m) = "s390x" ]; then + LD_LIBRARY_PATH=/lib64:/usr/lib64:/usr/X11R6/lib64:/usr/kerberos/lib64:/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib + else + LD_LIBRARY_PATH=/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib + fi + export LD_LIBRARY_PATH + + PATH="$PATH:/usr/bin:/bin:/sbin:/usr/sbin:/mnt/sysimage/bin:/mnt/sysimage/usr/bin:/mnt/sysimage/usr/sbin:/mnt/sysimage/sbin:/mnt/sysimage/usr/X11R6/bin" + export PATH + HOME=/ + export HOME + PYTHONPATH=/tmp/updates + export PYTHONPATH + + if [ -z "$testing" ]; then + + mount -t proc none /proc + + mount -t tmpfs none /dev + createDevices + # udevd req'd by udevadm settle (/dev/.udev/queue) + # in readcmsfile, dialog_network_table, semantic_check_subchannels. + # (important: start udevd at the right time, e.g. after setup of /dev) + echo "starting udev..." + udevd --daemon + + mount -t devpts /dev/pts /dev/pts + mount -t sysfs none /sys + # remount root fs rw + mount /dev/root / -o remount,rw + + # limit output on 3270 console + # (console_loglevel of 4 is just right to not get driver info, + # e.g. from qeth, since that would mix up with the user dialog) + echo "4 4 1 7" > /proc/sys/kernel/printk + + # make /tmp/ramfs + mount -t ramfs none /tmp + + ifconfig lo 127.0.0.1 netmask 255.0.0.0 + route add -host 127.0.0.1 dev lo + + echo -e "127.0.0.1\tlocalhost.localdomain localhost localhost4 localhost4.localdomain4" > /etc/hosts + echo -e "::1\t\tlocalhost.localdomain localhost localhost6 localhost6.localdomain6" >> /etc/hosts + + # init.c would also fork /sbin/dbus-uuidgen --ensure + # and /sbin/dbus-daemon --system + + fi # testing +} + # from here on accesses to sysfs try to follow # linux/Documentation/sysfs-rules.txt @@ -251,14 +330,14 @@ function clear_screen() { function dialog_network_table() { while : ; do - echo $"Scanning for available network devices..." + echo "Scanning for available network devices..." # This may take a long time so we show "progress": #( while true; do echo -n "."; sleep 1; done ) & #local childpid=$! read_lsznet_output #kill $childpid #echo - echo $"Autodetection found ${count} devices." + echo "Autodetection found ${count} devices." # count==0: there might still be a blacklist the user wants to clear. # do not flood user with long list if there are many devices if [ "$count" -le 15 ]; then @@ -267,7 +346,7 @@ function dialog_network_table() { else # [ $count -gt 15 ] echo while : ; do - echo $"s) show all, m) manual config:" + echo "s) show all, m) manual config:" local answer read answer case $answer in @@ -290,27 +369,27 @@ function dialog_network_table() { local cio_wc=$(wc -c /proc/cio_ignore) read cio_wc_bytes cio_wc_filename cio_wc_foo <<< "$cio_wc" if [ "$cio_wc_bytes" != "0" ]; then - echo $"Note: There is a device blacklist active! (Clearing might take long)" + echo "Note: There is a device blacklist active! (Clearing might take long)" #cat /proc/cio_ignore | tr '\n' ',' #echo else if [ "$count" -eq 0 ]; then # count==0 AND no device blacklist => manual mode - echo $"Entering manual configuration mode." + echo "Entering manual configuration mode." break fi fi fi # selection dialog while : ; do - [ "$count" -gt 0 ] && echo -n $"<num>) use config, " - [ "$cio_wc_bytes" != "0" ] && echo -n $"c) clear blacklist, " - echo $"m) manual config, r) rescan, s) shell:" + [ "$count" -gt 0 ] && echo -n "<num>) use config, " + [ "$cio_wc_bytes" != "0" ] && echo -n "c) clear blacklist, " + echo "m) manual config, r) rescan, s) shell:" local choice read choice [ -z "$choice" ] && continue if [ "$choice" = "s" ]; then - echo $"Enter 'exit' at the shell prompt to get back to the installation dialog." + echo "Enter 'exit' at the shell prompt to get back to the installation dialog." /bin/bash continue 2 fi @@ -336,7 +415,7 @@ function dialog_network_table() { esac done if [ "$choice" = "c" ]; then - echo $"Clearing device blacklist..." + echo "Clearing device blacklist..." if sysecho /proc/cio_ignore "free all"; then cio_wc_bytes=0 # /proc/cio_ignore won't block on freeing devices @@ -346,17 +425,15 @@ function dialog_network_table() { #( while true; do echo -n "."; sleep 3; done ) & #local childpid=$! #debug ls -laF /dev/.udev - udevsettle + udevadm settle # (virtual) CTC/A takes some more time to appear in sysfs # FIXME: how long to wait? 3 seconds seems to be enough. sleep 3 #kill $childpid #echo - # udevsettle only works if udevd is running (/dev/.udev/queue)! - # (without udevsettle we could wait 5-10 seconds uncond.) continue else - echo $"Device blacklist could not be cleared" + echo "Device blacklist could not be cleared" fi fi [ "$choice" = "m" ] && break @@ -366,7 +443,11 @@ function dialog_network_table() { local item cutype chp chpidtype devdrv devname chlist cardtype read item cutype chp chpidtype devdrv devname chlist cardtype <<< ${nettable[$choice]} # $NETTYPE happens to be exactly the network driver name - NETTYPE=$devdrv + if [ "$devdrv" = "ctcm" ]; then + NETTYPE="ctc" + else + NETTYPE=$devdrv + fi SUBCHANNELS=$chlist break done @@ -379,49 +460,7 @@ declare -r BUSIDFORMAT=[[:xdigit:]][[:xdigit:]][[:xdigit:]][[:xdigit:]] declare -r IDFORMAT=$PREFIXFORMAT.$SSIDFORMAT.$BUSIDFORMAT declare -r SUBCHANNEL_TYPE_IO=0 -# The following combinations of control unit type and model were taken from the -# Linux network device drivers for s390 on 2008-06-09 from Linux 2.6.25.4. -# The list (among other things) should be adapted, if any of those device -# drivers start supporting different CU types/models. -# (Alternatively, the list could be generated by reading the modaliases -# directly out of the device driver modules: -# modinfo qeth/cu3088 | fgrep 'alias:' -# However, this would still require a list of device driver modules.) - -declare -r -a CU=( - 1731/01 - 1731/05 - 3088/08 - 3088/1f - 3088/1e - 3088/01 - 3088/60 -) - -declare -r -a CU_DEVDRV=( - qeth - qeth - ctc - ctc - ctc - lcs - lcs -) - -# Searches for a match of argument 1 on the array $CU and sets $cu_idx -# to the matched array index on success. -# Returns 0 on success, 1 on failure. -function search_cu() { - local scu=$1 - local i - for ((i=0; i < ${#CU[@]}; i++)); do - if [ "$scu" == "${CU[i]}" ]; then - cu_idx=$i - return 0 - fi - done - return 1 -} +. /sbin/controlunits.sh function cardtype2cleartext() { local cardtype=$1 @@ -436,9 +475,9 @@ function cardtype2cleartext() { OSD_Express) echo "OSA card in OSD mode, unknown link type" ;; HSTR) echo "OSA card in OSD mode, High Speed Token Ring" ;; OSN) echo "OSA for NCP, ESCON/CDLC bridge" ;; - HiperSockets) echo "HiperSocket with CHPID type IQD" ;; + HiperSockets) echo "HiperSockets with CHPID type IQD" ;; "GuestLAN QDIO") echo "GuestLAN based on OSA (QDIO)" ;; - "GuestLAN Hiper") echo "GuestLAN based on HiperSocket" ;; + "GuestLAN Hiper") echo "GuestLAN based on HiperSockets" ;; unknown) echo "other" ;; *) echo "unknown" echo "l.$LINENO: found unknown card_type, code needs to be fixed" 1>&2 @@ -456,29 +495,6 @@ function isVM() { fi } -# echos one comparable integer built from 3 arguments in kernel version style -function kernel_version() { - [ $# -eq 3 ] || return 1 - local krn_ver=$1 krn_patch=$2 krn_sub=$3 - local version=$(( (krn_ver << 16) + (krn_patch << 8) + krn_sub )) - # <<EOF convince syntax highlighter that above shifts are no here documents - echo $version - return 0 -} - -# echos one comparable integer built from actual running kernel version -function linux_version() { - local krn_ver krn_patch krn_sub - local krn_foo krn_extra - IFS=. - read krn_ver krn_patch krn_foo < <(uname -r) - unset IFS - IFS=- - read krn_sub krn_extra <<< "$krn_foo" - unset IFS - kernel_version $krn_ver $krn_patch $krn_sub -} - # watch out: potential error message as side effect function isLayer2Default() { # Read default from sysfs because according to device @@ -487,7 +503,7 @@ function isLayer2Default() { # This only works here in installer where nobody has overwritten # the default setting with another custom value already! if [ ! -f /sys/devices/${NETTYPE}/$SCH_R_DEVBUSID/layer2 ]; then - echo $"Could not read layer mode from sysfs" + echo "Could not read layer mode from sysfs" return 1 fi local layer2 @@ -514,114 +530,37 @@ function isLayer2() { esac } -# returns true iff qeth device $SCH_R_DEVBUSID and kernel version +# returns true iff qeth device $SCH_R_DEVBUSID # is capable of supporting IPv6 # watch out: potential error message as side effect function ipv6_capable() { [ "$NETTYPE" = "qeth" ] || return 1 case $cardtype in - OSD_10GIG|OSD_1000|OSD_100|OSD_Express|"GuestLAN QDIO") return 0 ;; - HiperSockets) - if isLayer2; then - return 0 - elif [ $(linux_version) -ge $(kernel_version 2 6 26) ]; then - return 0 - else - return 1 - fi - ;; + OSD_10GIG|OSD_1000|OSD_100|OSD_Express|HiperSockets|"GuestLAN QDIO") + return 0 ;; OSD_GbE_LANE|OSD_FE_LANE|OSD_TR_LANE|OSD_ATM_LANE) return 1 ;; HSTR|OSN|unknown) return 1 ;; "GuestLAN Hiper") return 1 ;; - *) echo $"Unknown card_type to determine IPv6 support" + *) echo "Unknown card_type to determine IPv6 support" return 1 ;; esac } -function insert_module() { - local module=${1} - if grep "^${module%.ko} .*$" /proc/modules &> /dev/null ; then - # module has already been loaded => do not try to load again - return 0 - fi - if ! insmod $module; then - echo $"Could not load kernel module $module" - fi -} - -function load_kernel_modules() { - if [ ":$NETTYPE" = ":ctc" ]; then - insert_module ccwgroup$LO - insert_module cu3088$LO - insert_module fsm$LO - insert_module ctc$LO - elif [ ":$NETTYPE" = ":iucv" ]; then - insert_module fsm$LO - insert_module iucv$LO - insert_module netiucv$LO - elif [ "$NETTYPE" = "lcs" ]; then - insert_module ccwgroup$LO - insert_module cu3088$LO - insert_module lcs$LO - elif [ "$NETTYPE" = "qeth" ]; then - insert_module crypto_api$LO - insert_module xfrm_nalgo$LO - insert_module ipv6$LO - insert_module ccwgroup$LO - insert_module qdio$LO - insert_module qeth$LO - fi -} - -function remove_module() { - if ! rmmod $1; then - #echo $"Could not remove kernel module $1" - : - fi -} - -function remove_kernel_modules() { - if [ ":$NETTYPE" = ":ctc" ]; then - remove_module ctc - remove_module fsm - remove_module cu3088 - remove_module ccwgroup - elif [ ":$NETTYPE" = ":iucv" ]; then - remove_module netiucv - remove_module iucv - remove_module fsm - elif [ "$NETTYPE" = "lcs" ]; then - remove_module lcs - remove_module cu3088 - remove_module ccwgroup - elif [ "$NETTYPE" = "qeth" ]; then - remove_module qeth - remove_module qdio - remove_module ccwgroup - # The remaining 3 modules can hardly be unloaded - # since e.g. loopback holds an inet6 address and thus a reference. - # But that's no problem as long as the device drivers are removed. - remove_module ipv6 - remove_module xfrm_nalgo - remove_module crypto_api - fi -} - # sets device online _and_ retrieves DEVICE at the same time # iucv cannot be set online since it's not based on ccw(group) function set_device_online() { - echo $"Activating network device..." + echo "Activating network device..." local sysnettype case "${NETTYPE}" in qeth) sysnettype=${NETTYPE} ;; lcs|ctc) sysnettype=cu3088 ;; esac if ! [ -f /sys/devices/${sysnettype}/$SCH_R_DEVBUSID/online ]; then - echo $"Sysfs path to set device online does not exist." + echo "Sysfs path to set device online does not exist." return 1 fi if ! sysecho /sys/devices/${sysnettype}/$SCH_R_DEVBUSID/online "1"; then - echo $"Could not set device ($SUBCHANNELS) online" + echo "Could not set device ($SUBCHANNELS) online" return 1 fi local i=1 @@ -632,46 +571,46 @@ function set_device_online() { sleep 1 i=$((i+1)) if [ "$i" -gt 10 ]; then - echo $"Could not set device ($SUBCHANNELS) online within timeout" + echo "Could not set device ($SUBCHANNELS) online within timeout" return 1 fi done if [ "$NETTYPE" = "lcs" -o "$NETTYPE" = "ctc" ]; then # KH FIXME: Workaround for missing sysfs interface - # DEVICE=`cat /sys/devices/lcs/${SUBCHANNELS//,*/}/if_name` + # DEVICE=$(cat /sys/devices/lcs/${SUBCHANNELS//,*/}/if_name) # replaced with flexible solution: # https://bugzilla.redhat.com/show_bug.cgi?id=204803#c9 # "sys/bus/ccwgroup/devices/${SUBCHANNEL}/net\:* # for lcs after setting online" if [ ! -h /sys/devices/${sysnettype}/$SCH_R_DEVBUSID/net:* ]; then - echo $"Device $SUBCHANNELS does not have required sysfs attribute 'net:*'" + echo "Device $SUBCHANNELS does not have required sysfs attribute 'net:*'" return 1 fi DEVICE=$(echo /sys/devices/${sysnettype}/$SCH_R_DEVBUSID/net:*) DEVICE=${DEVICE//*:/} if [ "$DEVICE" = "" ]; then - echo $"Could not get device name for $SUBCHANNELS" + echo "Could not get device name for $SUBCHANNELS" return 1 fi else # qeth if [ ! -f /sys/devices/qeth/$SCH_R_DEVBUSID/if_name ]; then - echo $"Device $SUBCHANNELS does not have required sysfs attribute 'if_name'" + echo "Device $SUBCHANNELS does not have required sysfs attribute 'if_name'" return 1 fi # (device needs to be online to read if_name from sysfs attribute!) read DEVICE < /sys/devices/qeth/$SCH_R_DEVBUSID/if_name if [ "$DEVICE" = "" ]; then - echo $"Could not get device name for $SUBCHANNELS" + echo "Could not get device name for $SUBCHANNELS" return 1 fi if [ -f /sys/devices/qeth/$SCH_R_DEVBUSID/card_type ]; then read cardtype < /sys/devices/qeth/$SCH_R_DEVBUSID/card_type #debug echo "$cardtype" # device is now online and link type will be known - echo -n $"Detected: " + echo -n "Detected: " cardtype2cleartext "$cardtype" else - echo $"Could not read qeth network card type from sysfs." + echo "Could not read qeth network card type from sysfs." fi fi } @@ -679,12 +618,12 @@ function set_device_online() { # sets device up and blocks until device appears to be up function set_device_up() { if [ -z "$DEVICE" ]; then - echo $"Could not determine interface name to bring up device $SUBCHANNELS" + echo "Could not determine interface name to bring up device $SUBCHANNELS" return 1 fi # Device does not come up fast enough to use "ip" to configure, so block. # While OSA come up themselves after setting online, - # e.g. HiperSocket won't, so set them up explicitly for the following check + # e.g. HiperSockets won't => set them up explicitly for the following check debug ip link set up $DEVICE local i=1 while : ; do @@ -693,7 +632,7 @@ function set_device_up() { sleep 1 i=$((i+1)) if [ "$i" -gt 10 ]; then - echo $"Could not bring up device $DEVICE within timeout" + echo "Could not bring up device $DEVICE within timeout" return 1 fi done @@ -727,11 +666,13 @@ function syntax_check_domainname() { function modprobe_alias() { if [ ":$NETTYPE" = ":iucv" ]; then echo "alias $DEVICE netiucv" >> /tmp/modprobe.conf + elif [ ":$NETTYPE" = ":ctc" ]; then + echo "alias $DEVICE ctcm" >> /tmp/modprobe.conf else echo "alias $DEVICE $NETTYPE" >> /tmp/modprobe.conf fi if [ $? -ne 0 ]; then - echo $"Could not append alias for network device $DEVICE to modprobe.conf" + echo "Could not append alias for network device $DEVICE to modprobe.conf" return 1 fi return 0 @@ -752,14 +693,14 @@ function configure_ipv6_address() { # happens automatically by the driver. do NOT mess with default setting. #NO#debug ip link set dev $DEVICE arp on if ! debug ip -6 address add $IPADDR/$NETMASK dev $DEVICE; then - echo $"Could net set IPv6 address $IPADDR/$NETMASK for device $DEVICE" + echo "Could net set IPv6 address $IPADDR/$NETMASK for device $DEVICE" return 1 fi # network route has been set by above "ip address add" already # take care of MTU, which is bundled with ifconfig in the other IPv4 cases if [ -n "$MMTU" ]; then if ! debug ip link set $DEVICE $MMTU; then - echo $"Could net set maximum transfer unit ($MMTU) for device $DEVICE" + echo "Could net set maximum transfer unit ($MMTU) for device $DEVICE" return 1 fi fi @@ -771,21 +712,21 @@ function configure_ipv4_address() { if ipcalc -bmnp $ipcalc_arg > /tmp/ipcalc.$$.out 2> /dev/null; then . /tmp/ipcalc.$$.out else - echo $"Could not calculate network address and broadcast address from" - echo $" IPv4 address $IPADDR and netmask $NETMASK" + echo "Could not calculate network address and broadcast address from" + echo " IPv4 address $IPADDR and netmask $NETMASK" return 1 fi rm /tmp/ipcalc.$$.out # device needs to be online if ! debug ifconfig $DEVICE $IPADDR $MMTU netmask $NETMASK broadcast $BROADCAST; then - echo $"Could not set IPv4 address $IPADDR for device $DEVICE" - echo $" with network mask $NETMASK and broadcast address $BROADCAST" - [ -n "$MMTU" ] && echo $" and maximum transfer unit: $MMTU" + echo "Could not set IPv4 address $IPADDR for device $DEVICE" + echo " with network mask $NETMASK and broadcast address $BROADCAST" + [ -n "$MMTU" ] && echo " and maximum transfer unit: $MMTU" return 1 fi # This network route is already there after ifconfig! #if ! debug route add -net $NETWORK netmask $NETMASK dev $DEVICE; then - # echo $"Could not add network route to $NETWORK/$NETMASK on device $DEVICE" + # echo "Could not add network route to $NETWORK/$NETMASK on device $DEVICE" # return 1 #fi return 0 @@ -829,9 +770,6 @@ function rollback_config() { [ -z "$mtu_was_set" ] && unset MTU [ -z "$mmtu_was_set" ] && unset MMTU [ -z "$vswitch_was_set" ] && unset VSWITCH - # Try to silently unload modules. We ignore failures since - # load_kernel_modules only tries to load missing modules. - tv remove_kernel_modules # prevent possible reuse of an old DEVICE on restarting dialog unset DEVICE } @@ -871,9 +809,9 @@ function reenter_menu() { #[ -z "$1" ] && return 0 while : ; do if [ -n "$helptext" ]; then - echo $"0) default is previous \"$oldvalue\", 1) new value, ?) help" + echo "0) default is previous \"$oldvalue\", 1) new value, ?) help" else - echo $"0) default is previous \"$oldvalue\", 1) new value" + echo "0) default is previous \"$oldvalue\", 1) new value" fi # uncoded alternative: 2) skip parameter local answer @@ -885,7 +823,7 @@ function reenter_menu() { # temporarily for this parameter. # To put it differently: redoing does NOT present old values. redoitem="yes" - echo -n $"new value: " + echo -n "new value: " return 0 ;; "?") input="?" @@ -903,9 +841,9 @@ function workflow_item_menu() { while : ; do unset redoitem if [ "$noredo" = "noredo" ]; then - echo $"1) continue, 2) restart dialog, 3) halt, 4) shell" + echo "1) continue, 2) restart dialog, 3) halt, 4) shell" else - echo $"0) redo this parameter, 1) continue, 2) restart dialog, 3) halt, 4) shell" + echo "0) redo this parameter, 1) continue, 2) restart dialog, 3) halt, 4) shell" fi local answer read answer @@ -922,7 +860,7 @@ function workflow_item_menu() { 3) tv doshutdown exit 0 ;; - 4) echo $"Enter 'exit' at the shell prompt to get back to the installation dialog." + 4) echo "Enter 'exit' at the shell prompt to get back to the installation dialog." /bin/bash if [ "$noredo" != "noredo" ] && [ -n "$question_prefix" ]; then $question_prefix @@ -960,7 +898,7 @@ function ask() { c) handle=$OPTARG ;; f) finish=$OPTARG ;; "?") ;; # ignore invalid option - :) echo $"l.$LINENO: Missing parameter to option -$OPTARG" 1>&2 ;; + :) echo "l.$LINENO: Missing parameter to option -$OPTARG" 1>&2 ;; esac done while : ; do @@ -1034,7 +972,7 @@ function syntax_check_nettype() { 0) # string matched the pattern if [ "$NETTYPE" = "iucv" ] && ! isVM; then - echo $"IUCV network type is only available under z/VM (NETTYPE)." + echo "IUCV network type is only available under z/VM (NETTYPE)." return 1 fi return 0 @@ -1049,33 +987,33 @@ function syntax_check_nettype() { echo "l.$LINENO: unexpected return code of regex match operator =~, code needs to be fixed" 1>&2 ;; esac - echo $"Incorrect format or value for network type (NETTYPE): $NETTYPE" + echo "Incorrect format or value for network type (NETTYPE): $NETTYPE" return 1 } function question_prefix_nettype() { - echo -n $"Network type" + echo -n "Network type" } function question_choices_nettype() { if isVM; then - echo $" (qeth, lcs, ctc, iucv, ? for help). Default is qeth:" + echo " (qeth, lcs, ctc, iucv, ? for help). Default is qeth:" else - echo $" (qeth, lcs, ctc, ? for help). Default is qeth:" + echo " (qeth, lcs, ctc, ? for help). Default is qeth:" fi } function helptext_nettype() { - echo $" Help text for network type:" - echo $" qeth: OSA-Express Fast Ethernet, Gigabit Ethernet (including 1000Base-T)," - echo $" High Speed Token Ring, and ATM (running Ethernet LAN emulation)" - echo $" features in QDIO mode." - echo $" [default]" - echo $" lcs: OSA-2 Ethernet/Token Ring, OSA-Express Fast Ethernet in non-QDIO mode," - echo $" OSA-Express High Speed Token Ring in non-QDIO mode and Gigabit Ethernet" - echo $" in non-QDIO mode." - echo $" ctc: Deprecated, useful for migration." - isVM && echo $" iucv: Deprecated, useful for migration." + echo " Help text for network type:" + echo " qeth: OSA-Express Fast Ethernet, Gigabit Ethernet (including 1000Base-T)," + echo " High Speed Token Ring, and ATM (running Ethernet LAN emulation)" + echo " features in QDIO mode." + echo " [default]" + echo " lcs: OSA-2 Ethernet/Token Ring, OSA-Express Fast Ethernet in non-QDIO mode," + echo " OSA-Express High Speed Token Ring in non-QDIO mode and Gigabit Ethernet" + echo " in non-QDIO mode." + echo " ctc: Deprecated, useful for migration." + isVM && echo " iucv: Deprecated, useful for migration." } function exception_nettype() { @@ -1106,15 +1044,15 @@ function do_nettype() { function do_chandev() { echo - echo $"The CHANDEV variable isn't used anymore, please update your " - echo $".parm or the .conf file to use NETTYPE, SUBCHANNELS, etc. instead." + echo "The CHANDEV variable isn't used anymore, please update your " + echo ".parm or the .conf file to use NETTYPE, SUBCHANNELS, etc. instead." echo } ### SUBCHANNELS function syntax_check_subchannels() { - SUBCHANNELS=`echo $SUBCHANNELS | tr ABCDEF abcdef` + SUBCHANNELS=$(echo $SUBCHANNELS | tr ABCDEF abcdef) # - make subchannel question dependent on NETTYPE (2 vs. 3 subchannels) if [ "$NETTYPE" = "qeth" ]; then # - match against regex, depending on qeth @@ -1130,7 +1068,7 @@ function syntax_check_subchannels() { ;; 1) # string did not match the pattern - echo $"Incorrect format for channels (SUBCHANNELS): $SUBCHANNELS" + echo "Incorrect format for channels (SUBCHANNELS): $SUBCHANNELS" ;; 2) echo "l.$LINENO: syntax error in regex of match operator =~, code needs to be fixed" 1>&2 @@ -1169,25 +1107,25 @@ function semantic_check_subchannels() { # - check for existence of devnos in sysfs if [ ! -d "$dev_p" -a "$cio_wc_bytes" != "0" ]; then # - try to free from /proc/cio_ignore if they don't exist - echo $"Device $devbusid not present, trying to clear from blacklist and resense..." + echo "Device $devbusid not present, trying to clear from blacklist and resense..." if sysecho /proc/cio_ignore "free $devbusid"; then # /proc/cio_ignore won't block on freeing devices # until resensing has been completed, so wait until - # the udev event queue depletes (without udevsettle we + # the udev event queue depletes (without udevadm settle we # could wait 2 seconds unconditionally) #debug ls -laF /dev/.udev - udevsettle + udevadm settle # even though the device might now be online, some of its # sysfs attributes (e.g. cutype) might not yet be available sleep 1 else - echo $"Device $devbusid could not be cleared from device blacklist" + echo "Device $devbusid could not be cleared from device blacklist" fi fi # reevaluate since globbing might not have worked before device existed dev_p=$(echo /sys/devices/css$prefix/$IDFORMAT/$devbusid) if [ ! -d "$dev_p" ]; then - echo $"Device $devbusid does not exist" + echo "Device $devbusid does not exist" all_subch_good=1 continue fi @@ -1200,34 +1138,40 @@ function semantic_check_subchannels() { local type read type < $subch_p/type if [ "$type" != "$SUBCHANNEL_TYPE_IO" ]; then - echo $"Channel $subch (device $devbusid) is not of type I/O" + echo "Channel $subch (device $devbusid) is not of type I/O" all_subch_good=1 continue fi fi # - check for correct CU type/model, depending on qeth/lcs/ctc if [ ! -f $dev_p/cutype ]; then - echo $"Device $devbusid does not have required sysfs attribute 'cutype'" + echo "Device $devbusid does not have required sysfs attribute 'cutype'" all_subch_good=1 continue fi local cutype read cutype < $dev_p/cutype if search_cu $cutype; then - if [ "${CU_DEVDRV[$cu_idx]}" != "$NETTYPE" ]; then - echo $"Device $devbusid has control unit type $cutype," - echo $" which does not match your selected network type $NETTYPE" + local driver + if [ "$NETTYPE" = "ctc" ]; then + driver="ctcm" + else + driver=$NETTYPE + fi + if [ "${CU_DEVDRV[$cu_idx]}" != "$driver" ]; then + echo "Device $devbusid has control unit type $cutype," + echo " which does not match your selected network type $NETTYPE" all_subch_good=1 continue fi else - echo $"Device $devbusid has control unit type $cutype which is unknown" + echo "Device $devbusid has control unit type $cutype which is unknown" all_subch_good=1 continue fi # read CHPIDs information about subchannels if [ ! -f $subch_p/chpids ]; then - echo $"Channel $subch (device $devbusid) does not have required sysfs attribute 'chpids'" + echo "Channel $subch (device $devbusid) does not have required sysfs attribute 'chpids'" all_subch_good=1 continue fi @@ -1236,10 +1180,10 @@ function semantic_check_subchannels() { local -a chpids read -a chpids <<< "$chpid_list" if [ ${#chpids[@]} -ne 8 ]; then - echo $"sysfs reported ${#chpids[@]} CHPIDs instead of expected 8, code needs fix" + echo "sysfs reported ${#chpids[@]} CHPIDs instead of expected 8, code needs fix" fi if [ ! -f $subch_p/pimpampom ]; then - echo $"Channel $subch (device $devbusid) does not have required sysfs attribute 'pimpampom'" + echo "Channel $subch (device $devbusid) does not have required sysfs attribute 'pimpampom'" all_subch_good=1 continue fi @@ -1256,7 +1200,7 @@ function semantic_check_subchannels() { done local pimchpids=${pimchpidZ//ZZ/} if [ "x$pimchpids" == "x" ]; then - echo $"Channel $subch (device $devbusid) does not have any installed channel path" + echo "Channel $subch (device $devbusid) does not have any installed channel path" all_subch_good=1 continue fi @@ -1274,27 +1218,28 @@ function semantic_check_subchannels() { # => be sure to quote all variable accesses in test statements. # - all subchannels must be of same CU type/model if [ "$cutype" != "$sch_r_cutype" ]; then - echo $"Device $devbusid does not have the same control unit type as device $SCH_R_DEVBUSID" + echo "Device $devbusid does not have the same control unit type as device $SCH_R_DEVBUSID" comparison=1 fi # - all subchannels must have same CHPIDs if [ "$pimchpidZ" != "$sch_r_pimchipidZ" ]; then - echo $"Device $devbusid does not have the same CHPIDs as device $SCH_R_DEVBUSID" + echo "Device $devbusid does not have the same CHPIDs as device $SCH_R_DEVBUSID" comparison=1 fi # - all subchannels should have same prefix & ssid ? if [ "$prefix" != "$sch_r_prefix" \ -o "$ssid" != "$sch_r_ssid" ]; then - echo $"Device $devbusid does not have the same prefix and subchannel set ID as device $SCH_R_DEVBUSID" + echo "Device $devbusid does not have the same prefix and subchannel set ID as device $SCH_R_DEVBUSID" comparison=1 fi if [ "$i" -eq 2 ]; then local sch_w_devbusid=$devbusid local sch_w_devno=$devno + # TODO: not true for CTCM => relax # - write_devbusid == read_devbusid+1 if [ $((0x$devno)) -ne $((0x$sch_r_devno + 1)) ]; then - echo $"Device bus ID of write channel (dev $devbusid) must be one larger than" - echo $" that of read channel (dev $SCH_R_DEVBUSID)" + echo "Device bus ID of write channel (dev $devbusid) must be one larger than" + echo " that of read channel (dev $SCH_R_DEVBUSID)" comparison=1 fi elif [ "$i" -eq 3 ]; then @@ -1302,8 +1247,8 @@ function semantic_check_subchannels() { # (also seems to be handled by ccwgroup kernel subsystem) if [ "$devbusid" = "$sch_w_devbusid" \ -o "$devbusid" = "$SCH_R_DEVBUSID" ]; then - echo $"Device bus ID of data channel (dev $devbusid) must be different to that of" - echo $" read channel ($SCH_R_DEVBUSID) and write channel ($sch_w_devbusid)" + echo "Device bus ID of data channel (dev $devbusid) must be different to that of" + echo " read channel ($SCH_R_DEVBUSID) and write channel ($sch_w_devbusid)" comparison=1 fi fi @@ -1314,32 +1259,32 @@ function semantic_check_subchannels() { fi # filter potentially good subchannels ... if [ -h $dev_p/group_device ]; then - echo $"Device $devbusid is already in a ccwgroup and thus unavailable" + echo "Device $devbusid is already in a ccwgroup and thus unavailable" all_subch_good=1 continue fi if [ ! -f $dev_p/online ]; then - echo $"Device $devbusid does not have required sysfs attribute 'online'" + echo "Device $devbusid does not have required sysfs attribute 'online'" all_subch_good=1 continue fi local online read online < $dev_p/online if [ "$online" = "1" ]; then - echo $"Device $devbusid is already in use and thus unavailable" + echo "Device $devbusid is already in use and thus unavailable" all_subch_good=1 continue fi # - check availability if [ ! -f $dev_p/availability ]; then - echo $"Device $devbusid does not have required sysfs attribute 'availability'" + echo "Device $devbusid does not have required sysfs attribute 'availability'" all_subch_good=1 continue fi local availability read availability < $dev_p/availability if [ "$availability" != "good" ]; then - echo $"Device $devbusid is not available but '$availiability'" + echo "Device $devbusid is not available but '$availiability'" all_subch_good=1 continue fi @@ -1353,62 +1298,70 @@ function semantic_check_subchannels() { function handle_subchannels() { # - try to establish ccwgroup right here and fail out on error - if sysecho /sys/bus/ccwgroup/drivers/${NETTYPE}/group "$SUBCHANNELS"; then + local driver + if [ "$NETTYPE" = "ctc" ]; then + driver="ctcm" + else + driver=$NETTYPE + fi + if sysecho /sys/bus/ccwgroup/drivers/${driver}/group "$SUBCHANNELS"; then case "$NETTYPE" in qeth) # Just preliminary card_type info until device goes online! - # In fact it seems enough to separate OSA from HiperSocket. + # In fact it seems enough to separate OSA from HiperSockets. if [ -f /sys/devices/qeth/$SCH_R_DEVBUSID/card_type ]; then read cardtype < /sys/devices/qeth/$SCH_R_DEVBUSID/card_type else - echo $"Could not read qeth network card type from sysfs." + echo "Could not read qeth network card type from sysfs." fi ;; ctc|lcs) if [ -f /sys/devices/cu3088/$SCH_R_DEVBUSID/type ]; then - local type=$(cat /sys/devices/cu3088/$SCH_R_DEVBUSID/type) - [ "$type" = "CTC/A" ] && type="channel-to-channel adapter" + local type + read type < /sys/devices/cu3088/$SCH_R_DEVBUSID/type + [ "$type" = "CTC/A" ] && \ + type="channel-to-channel adapter (CTC/A)" echo "Detected: $type" else - echo $"Could not read ctc network card type from sysfs." + echo "Could not read ctc network card type from sysfs." fi ;; esac return 0 else - echo $"Channels $SUBCHANNELS could not be grouped" + echo "Channels $SUBCHANNELS could not be grouped" fi return 1 } function question_prefix_subchannels() { if [ "$NETTYPE" = "qeth" ]; then - echo -n $"Read,write,data channel" + echo -n "Read,write,data channel" else - echo -n $"Read,write channel" + echo -n "Read,write channel" fi } function question_choices_subchannels() { if [ "$NETTYPE" = "qeth" ]; then - echo $" (e.g. 0.0.0300,0.0.0301,0.0.0302 or ? for help)." + echo " (e.g. 0.0.0300,0.0.0301,0.0.0302 or ? for help)." else - echo $" (e.g. 0.0.0600,0.0.0601 or ? for help)" + echo " (e.g. 0.0.0600,0.0.0601 or ? for help)" fi } function helptext_subchannels() { if [ "$NETTYPE" = "qeth" ]; then - echo $" Help text for qeth channels:" - echo $" Enter the device bus ID of your CCW devices." - echo $" QETH needs three channels for read, write, and data," - echo $" e.g. 0.0.0300,0.0.0301,0.0.0302" + echo " Help text for qeth channels:" + echo " Enter the device bus ID of your CCW devices." + echo " QETH needs three channels for read, write, and data," + echo " e.g. 0.0.0300,0.0.0301,0.0.0302" else - echo $" Help text for lcs/ctc channels:" - echo $" Enter the device bus ID of your CCW devices." - echo $" CTC/ESCON and LCS need two channels for read and write," - echo $" e.g. 0.0.0600,0.0.0601 will configure the CTC or ESCON interface" - echo $" with the channels 0x600 and 0x601" + echo " Help text for lcs/ctc channels:" + echo " Enter the device bus ID of your CCW devices." + echo " CTC/ESCON and LCS need two channels for read and write," + echo " e.g. 0.0.0600,0.0.0601 will configure the CTC or ESCON interface" + echo " with the channels 0x600 and 0x601" fi } @@ -1436,12 +1389,12 @@ function do_subchannels() { function syntax_check_portname() { # - 1-8 characters, we convert it to upper case - PORTNAME=`echo $PORTNAME | tr '[:lower:]' '[:upper:]'` + PORTNAME=$(echo $PORTNAME | tr '[:lower:]' '[:upper:]') local portname_len=${#PORTNAME} if [ "$portname_len" -ge 1 -a "$portname_len" -le 8 ]; then return 0 fi - echo $"Incorrect string length [1..8] for portname (PORTNAME): $PORTNAME" + echo "Incorrect string length [1..8] for portname (PORTNAME): $PORTNAME" return 1 } @@ -1451,7 +1404,7 @@ function handle_portname() { if sysecho /sys/devices/${NETTYPE}/$SCH_R_DEVBUSID/portname "$PORTNAME"; then return 0 else - echo $"Portname '$PORTNAME' could not be configured for $SUBCHANNELS" + echo "Portname '$PORTNAME' could not be configured for $SUBCHANNELS" fi return 1 } @@ -1461,35 +1414,35 @@ function hint_portname() { local pname_hint read pname_hint < /sys/devices/${NETTYPE}/$SCH_R_DEVBUSID/portname if [ "$pname_hint" = "no portname required" ]; then - echo $" * Your configuration does not require a portname. *" + echo " * Your configuration does not require a portname. *" fi fi } function question_prefix_portname(){ - echo -n $"Portname" + echo -n "Portname" } function question_choices_portname(){ - echo $" (1..8 characters, or ? for help). Default is no portname:" + echo " (1..8 characters, or ? for help). Default is no portname:" } function helptext_portname(){ - echo $" Help text for portname:" + echo " Help text for portname:" # updated text describing when portname is obsolete; # taken from: # SA22-7935-09, Open Systems Adapter-Express Customer's # Guide and Reference, 10th ed. May 2008, IBM, p.17f. # SC33-8411-00, Device Drivers, Features, and Commands, # 1st ed. May 2008, IBM, p.116. - echo $" Portname of the OSA-Express feature in QDIO mode and z/VM Guest LAN." - echo $" This parameter is optional with:" - echo $" - z/VM 4.4.0 or z/VM 4.3.0 with APARs VM63308 and PQ73878" - echo $" - z800, z900 with >= Driver 3G - EC stream J11204, MCL032 (OSA level 3.33)" - echo $" - z890, z990, z9, z10 mainframes" + echo " Portname of the OSA-Express feature in QDIO mode and z/VM Guest LAN." + echo " This parameter is optional with:" + echo " - z/VM 4.4.0 or z/VM 4.3.0 with APARs VM63308 and PQ73878" + echo " - z800, z900 with >= Driver 3G - EC stream J11204, MCL032 (OSA level 3.33)" + echo " - z890, z990, z9, z10 mainframes" hint_portname - echo $" If portname is used, all operating systems sharing port must use same name." - echo $" Input empty string if you don't want to enter a portname. [default]" + echo " If portname is used, all operating systems sharing port must use same name." + echo " Input empty string if you don't want to enter a portname. [default]" } function exception_portname(){ @@ -1511,7 +1464,7 @@ function syntax_check_qeth_portno() { return 0 ;; esac - echo $"Incorrect format or value for relative port number (PORTNO): $PORTNO" + echo "Incorrect format or value for relative port number (PORTNO): $PORTNO" return 1 } @@ -1519,30 +1472,30 @@ function handle_qeth_portno() { if sysecho /sys/devices/qeth/$SCH_R_DEVBUSID/portno "$PORTNO"; then return 0 fi - echo $"Could not configure relative port number $PORTNO for $SUBCHANNELS" + echo "Could not configure relative port number $PORTNO for $SUBCHANNELS" return 1 } function question_prefix_portno() { - echo -n $"Relative port number for OSA" + echo -n "Relative port number for OSA" } function question_choices_portno() { - echo $" (0, 1, or ? for help). Default is 0:" + echo " (0, 1, or ? for help). Default is 0:" } function helptext_portno() { - echo $" Help text for relative port number for OSA with 2 ports per CHPID:" - echo $" This applies to:" - echo $" - OSA-Express3 Gigabit Ethernet on z10 systems" - echo $" - OSA-Express ATM on zSeries 800 and 900 systems" - echo $" 0 for relative port number 0 [default]" - echo $" 1 for relative port number 1" - echo $" Input empty string to not modify the default configuration." + echo " Help text for relative port number for OSA with 2 ports per CHPID:" + echo " This applies to:" + echo " - OSA-Express3 Gigabit Ethernet on z10 systems" + echo " - OSA-Express ATM on zSeries 800 and 900 systems" + echo " 0 for relative port number 0 [default]" + echo " 1 for relative port number 1" + echo " Input empty string to not modify the default configuration." } function exception_portno() { - # Writing portno of e.g. hipersocket device fails. + # Writing portno of e.g. hipersockets device fails. # Therefore, do not configure on empty default value. [ -z "$PORTNO" ] && break } @@ -1563,7 +1516,7 @@ function syntax_check_layer2() { return 0 ;; esac - echo $"Incorrect format or value for layer2 mode (LAYER2): $LAYER2" + echo "Incorrect format or value for layer2 mode (LAYER2): $LAYER2" return 1 } @@ -1574,37 +1527,41 @@ function handle_layer2() { if sysecho /sys/devices/${NETTYPE}/$SCH_R_DEVBUSID/layer2 "$LAYER2"; then return 0 else - echo $"Layer2 mode '$LAYER2' could not be configured for $SUBCHANNELS" + echo "Layer2 mode '$LAYER2' could not be configured for $SUBCHANNELS" fi return 1 } function question_prefix_layer2() { - echo -n $"Layer mode" + echo -n "Layer mode" } function question_choices_layer2() { - echo -n $" (0 for layer3, 1 for layer2, or ? for help)." + echo -n " (0 for layer3, 1 for layer2, or ? for help)." if [ "$isLayer2Default" = "yes" ]; then - echo $" Default is 1:" + echo " Default is 1:" else - echo $" Default is 0:" + echo " Default is 0:" fi } function helptext_layer2() { - echo $" Help text for OSA mode of operation: layer 2 vs. layer 3" + echo " Help text for OSA mode of operation: layer 2 vs. layer 3" if [ "$isLayer2Default" = "yes" ]; then - echo $" 0 for layer 3 mode (may not work with dhcp, tcpdump, etc.)" - echo $" 1 for layer 2 mode [default]" + echo " 0 for layer 3 mode (may not work with dhcp, tcpdump, etc.)" + echo " 1 for layer 2 mode [default]" else - echo $" 0 for layer 3 mode [default] (may not work with dhcp, tcpdump, etc.)" - echo $" 1 for layer 2 mode" + echo " 0 for layer 3 mode [default] (may not work with dhcp, tcpdump, etc.)" + echo " 1 for layer 2 mode" fi } function exception_layer2() { - [ -z "$LAYER2" ] && break + if [ -z "$LAYER2" ]; then + isLayer2Default && LAYER2=1 || LAYER2=0 + # do not break, always apply, default may differ from online layer mode + #break + fi } function do_layer2() { @@ -1627,7 +1584,7 @@ function syntax_check_macaddr() { ;; 1) # string did not match the pattern - echo $"Incorrect format for mac address (MACADDR): $MACADDR" + echo "Incorrect format for mac address (MACADDR): $MACADDR" ;; 2) echo "l.$LINENO: syntax error in regex of match operator =~, code needs to be fixed" 1>&2 @@ -1645,47 +1602,46 @@ function handle_macaddr() { if debug ifconfig $DEVICE hw ether $MACADDR; then return 0 fi - echo $"MAC address $MACADDR could not be configured for" - echo $" $SUBCHANNELS (network device $DEVICE)" + echo "MAC address $MACADDR could not be configured for" + echo " $SUBCHANNELS (network device $DEVICE)" return 1 } function question_prefix_macaddr() { - echo -n $"Unique MAC address" + echo -n "Unique MAC address" } function question_choices_macaddr() { - echo -n $" (e.g. 02:00:00:00:00:00, ? for help)." - # for virtual OSAs macaddr is optional and defaults to not specifying one - if [ -z "${cardtype//OSD_*/}" ]; then - echo - else - echo $" Default is automatic:" - fi + macaddr_default=$(ifconfig $DEVICE | grep 'HWaddr' | sed 's/.*HWaddr $[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]$.*/\1/') + echo " (e.g. 02:00:00:00:00:00, ? for help). Default is $macaddr_default:" } function helptext_macaddr() { - echo $" Help text for MAC address:" - echo $" This is required for real OSA in layer 2 mode." - echo $" If connecting to a layer 2 VSWITCH, a MAC address is automatically assigned." - echo -n $" You may accept the automatic MAC address with an empty input." + echo " Help text for MAC address:" if [ -z "${cardtype//OSD_*/}" ]; then - echo + echo " For real OSA in layer 2 mode, a random MAC address is automatically assigned." else - echo $" [default]" + echo " If connecting to a layer 2 VSWITCH, a MAC address is automatically assigned." fi - echo $" You may override the automatic MAC address with non-empty input." - echo $" An example MAC address would be: 02:00:00:00:00:00" + echo " You may accept the automatic MAC address with an empty input. [default]" + echo " If the automatic address is not unique, please provide a MAC address." + [ -z "${cardtype//OSD_*/}" ] && \ + echo " For real OSA, the provided address must be different from that of the OSA." + echo " You may override the automatic MAC address with non-empty input." + echo " An example MAC address would be: 02:00:00:00:00:00" } function exception_macaddr() { - # for real OSAs macaddr is required - if [ -n "${cardtype//OSD_*/}" ]; then - # for virtual OSAs macaddr is optional - if [ -z "$MACADDR" ]; then + if [ -z "$MACADDR" ]; then + if [ -z "${cardtype//OSD_*/}" ]; then + # keep random default MAC address of real OSA, + # so the OSA comes up with the same MAC each time in the future + MACADDR=$macaddr_default + else + # virtual OSA in layer2 is GuestLAN or VSWITCH VSWITCH=1 - break fi + break fi } @@ -1708,10 +1664,10 @@ function syntax_check_ctcprot() { return 0 ;; x2) - echo $"CTC tty's are not usable for this installation (CTCPROT)" + echo "CTC tty's are not usable for this installation (CTCPROT)" ;; *) - echo $"Incorrect format or value for CTC protocol (CTCPROT): $CTCPROT" + echo "Incorrect format or value for CTC protocol (CTCPROT): $CTCPROT" ;; esac return 1 @@ -1722,24 +1678,24 @@ function handle_ctcprot() { if sysecho /sys/devices/cu3088/${SCH_R_DEVBUSID}/protocol "$CTCPROT"; then return 0 fi - echo $"Could not configure CTC protocol $CTCPROT for $SUBCHANNELS" + echo "Could not configure CTC protocol $CTCPROT for $SUBCHANNELS" return 1 } function question_prefix_ctcprot() { - echo -n $"CTC protocol" + echo -n "CTC protocol" } function question_choices_ctcprot() { - echo $" (0, 1, 3, or ? for help). Default is 0:" + echo " (0, 1, 3, or ? for help). Default is 0:" } function helptext_ctcprot() { - echo $" Help text for CTC protocol:" - echo $" Protocol which should be used for the CTC interface" - echo $" 0 for compatibility with p.e. VM TCP service machine [default]" - echo $" 1 for enhanced package checking for Linux peers" - echo $" 3 for compatibility with OS/390 or z/OS peers" + echo " Help text for CTC protocol:" + echo " Protocol which should be used for the CTC interface" + echo " 0 for compatibility with p.e. VM TCP service machine [default]" + echo " 1 for enhanced package checking for Linux peers" + echo " 3 for compatibility with OS/390 or z/OS peers" } function do_ctcprot() { @@ -1767,7 +1723,7 @@ function syntax_check_lcs_portno() { echo "l.$LINENO: unexpected return code of regex match operator =~, code needs to be fixed" 1>&2 ;; esac - echo $"Incorrect format for LCS port number (PORTNAME): $PORTNAME" + echo "Incorrect format for LCS port number (PORTNAME): $PORTNAME" return 1 } @@ -1776,21 +1732,21 @@ function handle_lcs_portno() { if sysecho /sys/devices/cu3088/$SCH_R_DEVBUSID/portno "$PORTNAME"; then return 0 fi - echo $"Could not configure relative port number $PORTNAME for $SUBCHANNELS" + echo "Could not configure relative port number $PORTNAME for $SUBCHANNELS" return 1 } function question_prefix_lcs_portno() { - echo -n $"Relative port number of your LCS device" + echo -n "Relative port number of your LCS device" } function question_choices_lcs_portno() { - echo $" (number or ? for help). Default is 0:" + echo " (number or ? for help). Default is 0:" } function helptext_lcs_portno() { - echo $" Help text for relative port number of LCS device:" - echo $" Required for OSA-Express ATM cards only." + echo " Help text for relative port number of LCS device:" + echo " Required for OSA-Express ATM cards only." } function exception_lcs_portno() { @@ -1810,10 +1766,10 @@ function do_lcs_portno() { function syntax_check_peerid() { if [ "${#PEERID}" -lt 0 -o "${#PEERID}" -gt 8 ]; then - echo $"Incorrect string length [0..8] for IUCV connection to peer (PEERID): $PEERID" + echo "Incorrect string length [0..8] for IUCV connection to peer (PEERID): $PEERID" return 1 fi - PEERID=`echo $PEERID | tr '[:lower:]' '[:upper:]'` + PEERID=$(echo $PEERID | tr '[:lower:]' '[:upper:]') [[ "$PEERID" =~ ^[[:alnum:]$]{0,8}$ ]] case $? in 0) @@ -1830,13 +1786,13 @@ function syntax_check_peerid() { echo "l.$LINENO: unexpected return code of regex match operator =~, code needs to be fixed" 1>&2 ;; esac - echo $"Incorrect format for IUCV connection to peer (PEERID): $PEERID" + echo "Incorrect format for IUCV connection to peer (PEERID): $PEERID" return 1 } function handle_peerid() { if ! sysecho /sys/bus/iucv/drivers/netiucv/connection "$PEERID"; then - echo $"Could not create new IUCV connection to peer $PEERID" + echo "Could not create new IUCV connection to peer $PEERID" return 1 fi # find netiucv with proper guest user ID in @@ -1848,41 +1804,41 @@ function handle_peerid() { [ "$user" = "$PEERID" ] && break done < <(ls -1 /sys/devices/iucv/netiucv[0-9]*/user 2> /dev/null) if [ "$user" != "$PEERID" ]; then - echo $"Could not find configured connection to peer $PEERID" + echo "Could not find configured connection to peer $PEERID" return 1 fi local netiucv=${userpath%%/user} netiucv=${netiucv##*/} if [ "$netiucv" = "" ]; then - echo $"Could not get netiucv instance for peer $PEERID" + echo "Could not get netiucv instance for peer $PEERID" return 1 fi if [ ! -h /sys/devices/iucv/${netiucv}/net:* ]; then - echo $"Device $netiucv does not have required sysfs attribute 'net:*'" + echo "Device $netiucv does not have required sysfs attribute 'net:*'" return 1 fi DEVICE=$(echo /sys/devices/iucv/${netiucv}/net:*) DEVICE=${DEVICE//*:/} if [ "$DEVICE" = "" ]; then - echo $"Could not get device name for $netiucv" + echo "Could not get device name for $netiucv" return 1 fi return 0 } function question_prefix_peerid() { - echo -n $"Peer ID of the VM guest to connect to" + echo -n "Peer ID of the VM guest to connect to" } function question_choices_peerid() { - echo $" (0..8 characters or ? for help):" + echo " (0..8 characters or ? for help):" } function helptext_peerid() { - echo $" Help text for the peer ID of the VM guest you want to connect to:" - echo $" User ID of a VM guest you want to connect to." - echo $" 0..8 alphabetic/numeric characters or dollar-signs '$'." - echo $" Your input will be converted to uppercase." + echo " Help text for the peer ID of the VM guest you want to connect to:" + echo " User ID of a VM guest you want to connect to." + echo " 0..8 alphabetic/numeric characters or dollar-signs '$'." + echo " Your input will be converted to uppercase." } function do_peerid() { @@ -1894,28 +1850,28 @@ function do_peerid() { ### HOSTNAME function syntax_check_hostname() { - syntax_check_domainname "$HOSTNAME" $"Incorrect format for hostname (HOSTNAME): $HOSTNAME" + syntax_check_domainname "$HOSTNAME" "Incorrect format for hostname (HOSTNAME): $HOSTNAME" } function handle_hostname() { if ! hostname $HOSTNAME; then - echo $"Could not configure hostname $HOSTNAME" + echo "Could not configure hostname $HOSTNAME" return 1 fi return 0 } function question_prefix_hostname() { - echo -n $"Hostname of your new Linux guest" + echo -n "Hostname of your new Linux guest" } function question_choices_hostname() { - echo $" (FQDN e.g. s390.redhat.com or ? for help):" + echo " (FQDN e.g. s390.redhat.com or ? for help):" } function helptext_hostname() { - echo $" Help text for hostname:" - echo $" Enter the full qualified domain name of your host." + echo " Help text for hostname:" + echo " Enter the full qualified domain name of your host." } function do_hostname() { @@ -1936,33 +1892,31 @@ function syntax_check_ipaddr() { ipv6="yes" return 0 fi - echo $"Incorrect format for IP address (IPADDR): $IPADDR" + echo "Incorrect format for IP address (IPADDR): $IPADDR" return 1 } function question_prefix_ipaddr() { - echo -n $"IPv4 address" - [ "$ipv6_capable" = "yes" ] && echo -n $" / IPv6 addr." + echo -n "IPv4 address" + [ "$ipv6_capable" = "yes" ] && echo -n " / IPv6 addr." } function question_choices_ipaddr() { - echo -n $" (e.g. 10.0.0.2" - [ "$ipv6_capable" = "yes" ] && echo -n $" / 2001:0DB8::" - echo $" or ? for help)" + echo -n " (e.g. 10.0.0.2" + [ "$ipv6_capable" = "yes" ] && echo -n " / 2001:0DB8::" + echo " or ? for help)" } function helptext_ipaddr() { - echo $" Help text for IP address:" - echo $" Enter a valid IPv4 address of your new Linux guest (e.g. 10.0.0.2)" + echo " Help text for IP address:" + echo " Enter a valid IPv4 address of your new Linux guest (e.g. 10.0.0.2)" if [ "$ipv6_capable" = "yes" ]; then - echo $" or alternatively a valid IPv6 address without CIDR prefix (e.g. 2001:0DB8::)" - echo $" IPv6 is supported on:" - echo $" - Ethernet interfaces of the OSA-Express adapter running in QDIO mode." - echo $" - HiperSockets layer 2 interfaces" - [ $(linux_version) -ge $(kernel_version 2 6 26) ] && \ - echo $" - HiperSockets layer 3 interfaces" - echo $" - z/VM guest LAN interfaces running in QDIO mode." - echo $" IPv6 is not supported on HiperSocket guest LAN, OSA-Express Token Ring, ATM." + echo " or alternatively a valid IPv6 address without CIDR prefix (e.g. 2001:0DB8::)" + echo " IPv6 is supported on:" + echo " - Ethernet interfaces of the OSA-Express adapter running in QDIO mode." + echo " - HiperSockets interfaces" + echo " - z/VM guest LAN interfaces running in QDIO mode." + echo " IPv6 is not supported on HiperSockets guest LAN, OSA-Express Token Ring, ATM." fi } @@ -1983,18 +1937,18 @@ function syntax_check_netmask_v4() { ipcalc_arg="$IPADDR/$NETMASK" return 0 fi - echo $"Incorrect value for network prefix [1..32] (NETMASK): $NETMASK" + echo "Incorrect value for network prefix [1..32] (NETMASK): $NETMASK" return 1 elif checkipv4 $NETMASK; then ipcalc_arg="$IPADDR $NETMASK" return 0 fi - echo $"Incorrect format or value for network mask (NETMASK): $NETMASK" + echo "Incorrect format or value for network mask (NETMASK): $NETMASK" return 1 } function question_prefix_netmask() { - echo -n $"IPv4 netmask or CIDR prefix" + echo -n "IPv4 netmask or CIDR prefix" } function hint_netmask_v4() { @@ -2007,13 +1961,13 @@ function hint_netmask_v4() { # <<EOF convince syntax highlighter that above shifts are no here documents if [ $(( ip & 0x80000000 )) -eq $(( 0x00000000 )) ]; then # class a - echo $"255.0.0.0" + echo "255.0.0.0" elif [ $(( ip & 0xC0000000 )) -eq $(( 0x80000000 )) ]; then # class b - echo $"255.255.0.0" + echo "255.255.0.0" elif [ $(( ip & 0xE0000000 )) -eq $(( 0xC0000000 )) ]; then # class c - echo $"255.255.255.0" + echo "255.255.255.0" else # some other class that should not be used as host address return 1 @@ -2022,24 +1976,24 @@ function hint_netmask_v4() { } function question_choices_netmask() { - echo -n $" (e.g. 255.255.255.0 or 1..32 or ? for help)" + echo -n " (e.g. 255.255.255.0 or 1..32 or ? for help)" local default=$(hint_netmask_v4) if [ -n "$default" ]; then - echo $". Default is $default:" + echo ". Default is $default:" else - echo $":" - echo $"The IP address you entered previously should probably not be used for a host." + echo ":" + echo "The IP address you entered previously should probably not be used for a host." fi } function helptext_netmask() { - echo $" Help text for IPv4 netmask or CIDR prefix:" - echo $" Enter a valid IPv4 netmask or CIDR prefix (e.g. 255.255.255.0 or 1..32)" + echo " Help text for IPv4 netmask or CIDR prefix:" + echo " Enter a valid IPv4 netmask or CIDR prefix (e.g. 255.255.255.0 or 1..32)" local default=$(hint_netmask_v4) if [ -n "$default" ]; then - echo $" Default is $default" + echo " Default is $default" else - echo $"The IP address you entered previously should probably not be used for a host." + echo "The IP address you entered previously should probably not be used for a host." fi } @@ -2065,16 +2019,16 @@ function syntax_check_prefix_v6() { return 0 fi fi - echo $"Incorrect value for network prefix [1..128] (NETMASK): $NETMASK" + echo "Incorrect value for network prefix [1..128] (NETMASK): $NETMASK" return 1 } function question_prefix_netmask_v6() { - echo -n $"CIDR prefix for the IPv6 address" + echo -n "CIDR prefix for the IPv6 address" } function question_choices_netmask_v6() { - echo $" (1..128):" + echo " (1..128):" } function do_netmask_v6() { @@ -2098,11 +2052,11 @@ function configure_ipv4_gateway() { ip -4 route del default dev $DEVICE >& /dev/null [ -z "$GATEWAY" ] && return 0 if ! tv route add default gw $GATEWAY dev $DEVICE; then - echo $"Could net set default route on device $DEVICE via gateway $GATEWAY" + echo "Could net set default route on device $DEVICE via gateway $GATEWAY" return 1 fi # BH FIXME: Workaround for manual MACADDR, need ping to update arp table - echo $"Trying to reach gateway $GATEWAY..." + echo "Trying to reach gateway $GATEWAY..." if [ "$NETTYPE" = "ctc" ]; then # (virtual) CTC(/A) seems to need some time to get functional local i=1 @@ -2110,13 +2064,13 @@ function configure_ipv4_gateway() { ping -c 1 $GATEWAY >& /dev/null && break i=$((i+1)) if [ "$i" -gt 3 ]; then - echo $"Could not reach gateway $GATEWAY within timeout" + echo "Could not reach gateway $GATEWAY within timeout" return 1 fi done else if ! ping -c 1 $GATEWAY >& /dev/null; then - echo $"Could not reach your default gateway $GATEWAY" + echo "Could not reach your default gateway $GATEWAY" return 1 fi fi @@ -2142,29 +2096,29 @@ function hint_ipv4_gateway() { local hi_b=$(( (hi & 0x00FF0000) >> 16 )) local hi_c=$(( (hi & 0x0000FF00) >> 8 )) local hi_d=$(( (hi & 0x000000FE) )) - echo $" Depending on your network design patterns, the default gateway" - echo $" might be $lo_a.$lo_b.$lo_c.$lo_d or $hi_a.$hi_b.$hi_c.$hi_d" + echo " Depending on your network design patterns, the default gateway" + echo " might be $lo_a.$lo_b.$lo_c.$lo_d or $hi_a.$hi_b.$hi_c.$hi_d" } function question_prefix_gateway() { - echo -n $"IPv4 address of your default gateway" + echo -n "IPv4 address of your default gateway" } function question_choices_gateway() { - echo $" or ? for help:" + echo " or ? for help:" } function helptext_gateway() { - echo $" Help text for IPv4 default gateway:" - echo $" For HiperSockets with internal traffic only you may want to leave this empty" - echo $" and choose continue afterwards to go on without gateway." + echo " Help text for IPv4 default gateway:" + echo " For HiperSockets with internal traffic only you may want to leave this empty" + echo " and choose continue afterwards to go on without gateway." hint_ipv4_gateway } function finish_gateway() { if ! checkipv4 $GATEWAY; then # above checkipv4 is silent, so make up for syntax error - echo $"Incorrect format for IPv4 address of gateway (GATEWAY): $GATEWAY" + echo "Incorrect format for IPv4 address of gateway (GATEWAY): $GATEWAY" workflow_item_menu fi if configure_ipv4_gateway; then @@ -2203,37 +2157,37 @@ function configure_ipv6_gateway() { # ip -6 route show | grep ^default # However, we currently use manual IPv6 configuration only.) if ! debug ip -6 route add ::/0 dev $DEVICE via $GATEWAY; then - echo $"Could net set default route on device $DEVICE" - echo $" via gateway $GATEWAY" + echo "Could net set default route on device $DEVICE" + echo " via gateway $GATEWAY" return 1 fi # BH FIXME: Workaround for manual MACADDR, need ping to update arp table - echo $"Trying to reach gateway $GATEWAY..." + echo "Trying to reach gateway $GATEWAY..." if ! ping6 -c 1 $GATEWAY >& /dev/null; then - echo $"Could not reach your default gateway $GATEWAY" + echo "Could not reach your default gateway $GATEWAY" return 1 fi return 0 } function question_prefix_gateway_v6() { - echo -n $"IPv6 address of your default gateway" + echo -n "IPv6 address of your default gateway" } function question_choices_gateway_v6() { - echo $":" + echo ":" } function helptext_gateway_v6() { - echo $" Help text for IPv6 default gateway:" - echo $" For HiperSockets with internal traffic only you may want to leave this empty" - echo $" and choose continue afterwards to go on without gateway." + echo " Help text for IPv6 default gateway:" + echo " For HiperSockets with internal traffic only you may want to leave this empty" + echo " and choose continue afterwards to go on without gateway." } function finish_gateway_v6() { if ! checkipv6 $GATEWAY; then # above checkipv6 is silent, so make up for syntax error - echo $"Incorrect format for IPv6 address of gateway (GATEWAY): $GATEWAY" + echo "Incorrect format for IPv6 address of gateway (GATEWAY): $GATEWAY" workflow_item_menu fi if configure_ipv6_gateway; then @@ -2259,37 +2213,37 @@ function configure_ipv4_ptp() { configure_ipv4_gateway return $? fi - echo $"Could not set IPv4 address $IPADDR for device $DEVICE" - echo $" to peer $GATEWAY" - [ -n "$MMTU" ] && echo $" and maximum transfer unit: $MMTU" + echo "Could not set IPv4 address $IPADDR for device $DEVICE" + echo " to peer $GATEWAY" + [ -n "$MMTU" ] && echo " and maximum transfer unit: $MMTU" return 1 } function question_prefix_ptp_gateway() { - echo -n $"IPv4 address of your point-to-point partner" + echo -n "IPv4 address of your point-to-point partner" } function question_choices_ptp_gateway() { - echo $" or ? for help:" + echo " or ? for help:" # no hinting possible here } function helptext_ptp_gateway() { - echo $" Help text for point-to-point partner:" - echo $" IPv4 address of your CTC / ESCON / IUCV point-to-point partner." + echo " Help text for point-to-point partner:" + echo " IPv4 address of your CTC / ESCON / IUCV point-to-point partner." } function finish_ptp_gateway() { if checkipv4 $GATEWAY; then if [ "$GATEWAY" = "$IPADDR" ]; then - echo $"IPv4 address of partner should probably be different from the guest's address" + echo "IPv4 address of partner should probably be different from the guest's address" workflow_item_menu && break else break fi else # above checkipv4 is silent, so make up for syntax error - echo $"Incorrect format for IPv4 address of partner (GATEWAY): $GATEWAY" + echo "Incorrect format for IPv4 address of partner (GATEWAY): $GATEWAY" workflow_item_menu && break fi # too early to actually configure gateway @@ -2305,7 +2259,7 @@ function do_ptp_gateway() { function syntax_check_dns() { if [ -z "$DNS" ]; then - echo $"You might encounter problems without a nameserver, especially with FTP installs" + echo "You might encounter problems without a nameserver, especially with FTP installs" return 1 fi local dnsitem @@ -2313,14 +2267,14 @@ function syntax_check_dns() { if [ "$ipv6" ]; then while read dnsitem; do if ! checkipv6 $dnsitem; then - echo $"Not a valid IPv6 address for DNS server: $dnsitem" + echo "Not a valid IPv6 address for DNS server: $dnsitem" allgood="no" fi done < <(echo $DNS | sed 's/,/\n/g') else while read dnsitem; do if ! checkipv4 $dnsitem; then - echo $"Not a valid IPv4 address for DNS server: $dnsitem" + echo "Not a valid IPv4 address for DNS server: $dnsitem" allgood="no" fi done < <(echo $DNS | sed 's/:/\n/g') @@ -2337,11 +2291,11 @@ function handle_dns() { [ -z "$DNS" ] && return 0 local dnsitem local allgood="yes" - echo $"Trying to reach DNS servers..." + echo "Trying to reach DNS servers..." if [ "$ipv6" ]; then while read dnsitem; do if ! ping6 -c 1 $dnsitem >& /dev/null; then - echo $"Could not ping DNS server (might still serve DNS requests): $dnsitem" + echo "Could not ping DNS server (might still serve DNS requests): $dnsitem" allgood="no" # this should not be a hard failure since some network # environments may prevent pings to DNS servers @@ -2355,15 +2309,15 @@ function handle_dns() { # However, nslookup fails with "Resolver Error 0 (no error)" # at this stage of the setup progress => not useful if ! ping -c 1 $dnsitem >& /dev/null; then - echo $"Could not ping DNS server: $dnsitem" + echo "Could not ping DNS server: $dnsitem" # if nslookup $dnsitem $dnsitem >& /dev/null; then -# echo $" but could resolve DNS server with itself: $dnsitem" +# echo " but could resolve DNS server with itself: $dnsitem" # else -# echo $"Could not resolve DNS server with itself: $dnsitem" +# echo "Could not resolve DNS server with itself: $dnsitem" # allgood="no" # fi # elif ! nslookup $dnsitem $dnsitem >& /dev/null; then -# echo $"Could not resolve DNS server with itself: $dnsitem" +# echo "Could not resolve DNS server with itself: $dnsitem" allgood="no" fi done < <(echo $DNS | sed 's/:/\n/g') @@ -2377,34 +2331,34 @@ function handle_dns() { function question_prefix_dns() { if [ "$ipv6" ]; then - echo -n $"IPv6 addresses of DNS servers" + echo -n "IPv6 addresses of DNS servers" else - echo -n $"IPv4 addresses of DNS servers" + echo -n "IPv4 addresses of DNS servers" fi } function question_choices_dns() { if [ "$ipv6" ]; then - echo $" (separated by commas ',' or ? for help):" + echo " (separated by commas ',' or ? for help):" else - echo $" (separated by colons ':' or ? for help):" + echo " (separated by colons ':' or ? for help):" fi } function helptext_dns() { - echo $" Help text for DNS servers:" + echo " Help text for DNS servers:" if [ "$ipv6" ]; then - echo $" Enter IPv6 addresses of DNS servers separated by commas ','" + echo " Enter IPv6 addresses of DNS servers separated by commas ','" else - echo $" Enter IPv4 addresses of DNS servers separated by colons ':'" + echo " Enter IPv4 addresses of DNS servers separated by colons ':'" fi - echo $" Default are no DNS servers at all." - echo $" However, you might encounter problems without a nameserver," - echo $" especially with FTP installs." + echo " Default are no DNS servers at all." + echo " However, you might encounter problems without a nameserver," + echo " especially with FTP installs." if [ "$ipv6" ]; then - echo $" An example with 2 servers would be: 2001:0DB8::42,2001:0DB8::BE:AF" + echo " An example with 2 servers would be: 2001:0DB8::42,2001:0DB8::BE:AF" else - echo $" An example with 2 servers would be: 10.0.0.250:10.1.1.1" + echo " An example with 2 servers would be: 10.0.0.250:10.1.1.1" fi } @@ -2421,7 +2375,7 @@ function syntax_check_searchdns() { local dnsitem local allgood="yes" while read dnsitem; do - syntax_check_domainname "$dnsitem" $"Not a valid DNS search domain: $dnsitem" || allgood="no" + syntax_check_domainname "$dnsitem" "Not a valid DNS search domain: $dnsitem" || allgood="no" done < <(echo $SEARCHDNS | sed 's/:/\n/g') if [ "$allgood" = "yes" ]; then return 0 @@ -2431,18 +2385,18 @@ function syntax_check_searchdns() { } function question_prefix_searchdns() { - echo -n $"DNS search domains" + echo -n "DNS search domains" } function question_choices_searchdns() { - echo $" (separated by colons ':' or ? for help):" + echo " (separated by colons ':' or ? for help):" } function helptext_searchdns() { - echo $" Help text for DNS search domains:" - echo $" Enter search domains according to hostname syntax separated by colons." - echo $" Default are no DNS search domains at all." - echo $" An example would be: subdomain.domain.com:domain.com" + echo " Help text for DNS search domains:" + echo " Enter search domains according to hostname syntax separated by colons." + echo " Default are no DNS search domains at all." + echo " An example would be: subdomain.domain.com:domain.com" } function do_searchdns() { @@ -2473,7 +2427,7 @@ function parse_dasd() { 0) ;; # string matched the pattern 1) # string did not match the pattern if [ -z "$handle" ]; then - echo $"Incorrect format for lower bound of DASD range $range: $lo" + echo "Incorrect format for lower bound of DASD range $range: $lo" allgood="no" fi ;; @@ -2487,7 +2441,7 @@ function parse_dasd() { 0) ;; # string matched the pattern 1) # string did not match the pattern if [ -z "$handle" ]; then - echo $"Incorrect format for upper bound of DASD range $range: $hi" + echo "Incorrect format for upper bound of DASD range $range: $hi" allgood="no" fi ;; @@ -2497,13 +2451,13 @@ function parse_dasd() { fi if [ "$handle" = "yes" ]; then if ! sysecho /proc/cio_ignore "free $range"; then - echo $"Could not free DASD device $range from device blacklist" + echo "Could not free DASD device $range from device blacklist" allgood="no" fi fi if [ "${features//*)/}" != "" ]; then if [ -z "$handle" ]; then - echo $"Missing closing parenthesis at features of DASD range $range: ($features" + echo "Missing closing parenthesis at features of DASD range $range: ($features" allgood="no" fi fi @@ -2513,7 +2467,7 @@ function parse_dasd() { case $feature in ro|diag|erplog) continue ;; *) if [ -z "$handle" ]; then - echo $"Unknown DASD feature for device range $range: $feature" + echo "Unknown DASD feature for device range $range: $feature" allgood="no" fi ;; @@ -2523,7 +2477,9 @@ function parse_dasd() { esac done < <(echo $DASD | sed 's/,/\n/g') if [ "$handle" = "yes" ]; then - udevsettle + udevadm settle + # even though the device might now be online, some of its + # sysfs attributes might not yet be available sleep 1 fi if [ "$allgood" = "yes" ]; then @@ -2544,18 +2500,18 @@ function handle_dasd() { } function question_prefix_dasd() { - echo -n $"DASD range" + echo -n "DASD range" } function question_choices_dasd() { - echo $" (e.g. 200-203,205 or ? for help). Default is autoprobing:" + echo " (e.g. 200-203,205 or ? for help). Default is autoprobing:" } function helptext_dasd() { - echo $" Help text for DASD range:" - echo $" Comma separated list of ranges of device bus IDs." - echo $" Default is autoprobing (not recommended)." - echo $" Examples would be: 200-203 or 200,201,202,203 or 0.0.0200-0.0.0203,0.0.0205" + echo " Help text for DASD range:" + echo " Comma separated list of ranges of device bus IDs." + echo " Default is autoprobing (not recommended)." + echo " Examples would be: 200-203 or 200,201,202,203 or 0.0.0200-0.0.0203,0.0.0205" } function exception_dasd() { @@ -2587,14 +2543,14 @@ function syntax_check_fcp() { devno=${fcp[0]} wwpn=${fcp[2]} lun=${fcp[4]} - echo $"Deprecated number of FCP arguments (5 instead of 3): " - echo $" $i=\"${!i}\"" - echo $" should instead be: " - echo $" $i=\"$devno $wwpn $lun\"" + echo "Deprecated number of FCP arguments (5 instead of 3): " + echo " $i=\"${!i}\"" + echo " should instead be: " + echo " $i=\"$devno $wwpn $lun\"" ;; *) - echo $"Unsupported number of FCP arguments (${#fcp[@]} instead of 3) in:" - echo $" $i=\"${!i}\"" + echo "Unsupported number of FCP arguments (${#fcp[@]} instead of 3) in:" + echo " $i=\"${!i}\"" allgood="no" continue ;; @@ -2603,8 +2559,8 @@ function syntax_check_fcp() { case $? in 0) ;; # string matched the pattern 1) # string did not match the pattern - echo $"Incorrect format for FCP device $devno in:" - echo $" $i=\"${!i}\"" + echo "Incorrect format for FCP device $devno in:" + echo " $i=\"${!i}\"" allgood="no" ;; 2) @@ -2619,8 +2575,8 @@ function syntax_check_fcp() { case $? in 0) ;; # string matched the pattern 1) # string did not match the pattern - echo $"Incorrect format for FCP WWPN $wwpn in:" - echo $" $i=\"${!i}\"" + echo "Incorrect format for FCP WWPN $wwpn in:" + echo " $i=\"${!i}\"" allgood="no" ;; 2) echo "l.$LINENO: syntax error in regex of match operator =~, code needs to be fixed" 1>&2 ;; @@ -2632,8 +2588,8 @@ function syntax_check_fcp() { case $? in 0) ;; # string matched the pattern 1) # string did not match the pattern - echo $"Incorrect format for FCP LUN $lun in:" - echo $" $i=\"${!i}\"" + echo "Incorrect format for FCP LUN $lun in:" + echo " $i=\"${!i}\"" allgood="no" ;; 2) @@ -2699,7 +2655,7 @@ function final_check() { # optionally consider "continue" as default # but then again the user may inadvertently continue echo - echo $"c) continue, p) parm file/configuration, n) network state, r) restart, s) shell" + echo "c) continue, p) parm file/configuration, n) network state, r) restart, s) shell" local answer read answer case $answer in @@ -2717,7 +2673,7 @@ function final_check() { fi ;; r) break ;; - s) echo $"Enter 'exit' at the shell prompt to get back to the installation dialog." + s) echo "Enter 'exit' at the shell prompt to get back to the installation dialog." /bin/bash ;; esac @@ -2727,101 +2683,43 @@ function final_check() { ### MAIN ### -S390ARCH=`uname -m` -if [ "$S390ARCH" = "s390" ]; then - export S390ARCH="S/390" -else - export S390ARCH="zSeries" -fi +init_main -echo -echo $"Starting the $S390ARCH initrd to configure networking. Version is $VERSION" - -# set up env vars as we do in init.c -if [ `uname -m` = "s390x" ]; then - LD_LIBRARY_PATH=/lib64:/usr/lib64:/usr/X11R6/lib64:/usr/kerberos/lib64:/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib -else - LD_LIBRARY_PATH=/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib +# Parse configuration +if [ -n "$CMSDASD" -a -n "$CMSCONFFILE" ]; then + readcmsfile $CMSDASD $CMSCONFFILE + source /tmp/$CMSCONFFILE #2>/dev/null fi -export LD_LIBRARY_PATH - -PATH="$PATH:/usr/bin:/bin:/sbin:/usr/sbin:/mnt/sysimage/bin:/mnt/sysimage/usr/bin:/mnt/sysimage/usr/sbin:/mnt/sysimage/sbin:/mnt/sysimage/usr/X11R6/bin" -export PATH -HOME=/ -export HOME -PYTHONPATH=/tmp/updates -export PYTHONPATH - -if [ -z "$testing" ]; then - - mount -t proc none /proc - mount -t tmpfs none /dev - createDevices - mount -t devpts /dev/pts /dev/pts - mount -t sysfs none /sys - # remount root fs rw - mount /dev/root / -o remount,rw - - # limit output on x3270 console - # (console_loglevel of 4 is just right to not get driver info, e.g. from qeth, - # since that would mix up with the user dialog) - echo "4 4 1 7" > /proc/sys/kernel/printk - - # make /tmp/ramfs - mount -t ramfs none /tmp - - ifconfig lo 127.0.0.1 netmask 255.0.0.0 - route add -host 127.0.0.1 dev lo - - echo -e "127.0.0.1\tlocalhost.localdomain localhost" > /etc/hosts - echo -e "::1\t\tlocalhost6.localdomain6 localhost6" >> /etc/hosts - - LO="" - - [ -L /sbin/insmod ] && LO=".ko" - - # udevd req'd by udevsettle in readcmsfile, dialog_network_table, and - # semantic_check_subchannels. - # (important to start udevd at the right time, e.g. after setup of /dev) - udevd --daemon - - # Parse configuration - if [ -n "$CMSDASD" -a -n "$CMSCONFFILE" ]; then - readcmsfile $CMSDASD $CMSCONFFILE - source /tmp/$CMSCONFFILE #2>/dev/null - fi - -fi # testing - -### MAIN (continuation) ### if [ -r /sys/firmware/ipl/ipl_type ]; then - if [ "`cat /sys/firmware/ipl/ipl_type`" = "fcp" ]; then - while [ 1 ]; do - echo $"Your IPL device is set to FCP." - echo $"Would you like to perform a CD-ROM/DVD-ROM installation? (y/n)" + #local ipl_type + read ipl_type < /sys/firmware/ipl/ipl_type + if [ "$ipl_type" = "fcp" ]; then + while : ; do + echo "Your IPL device is set to FCP." + echo "Would you like to perform a CD-ROM/DVD-ROM installation? (y/n)" + #local do_cd_install read do_cd_install case $do_cd_install in y|Y|[Yy][Ee][Ss]) - insert_module qdio$LO - insert_module scsi_mod$LO - insert_module scsi_transport_fc$LO - insert_module zfcp$LO - CD_DEVICE="`cat /sys/firmware/ipl/device`" - WWPN="`cat /sys/firmware/ipl/wwpn`" - LUN="`cat /sys/firmware/ipl/lun`" + # precondition: zfcp driver incl. dependencies loaded + #local CD_DEVICE WWPN LUN + read CD_DEVICE < /sys/firmware/ipl/device + read WWPN < /sys/firmware/ipl/wwpn + read LUN < /sys/firmware/ipl/lun if sysecho /proc/cio_ignore "free $CD_DEVICE"; then - udevsettle + udevadm settle + # even though device might now be online, some of its + # sysfs attributes might not yet be available sleep 1 else - echo $"Device $CD_DEVICE could not be cleared from device blacklist" + echo "Device $CD_DEVICE could not be cleared from device blacklist" fi - echo 1 > /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/online - [ $? == 0 ] || echo $"Could not set FCP device $CD_DEVICE online" - echo $WWPN > /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/port_add - [ $? == 0 ] || echo $"Could not add WWPN $WWPN on FCP device $CD_DEVICE" - echo $LUN > /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/$WWPN/unit_add - [ $? == 0 ] || echo $"Could not add LUN $LUN at WWPN $WWPN on FCP device $CD_DEVICE" + sysecho /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/online 1 \ + || echo "Could not set FCP device $CD_DEVICE online" + # port (WWPN) appears automatically + sysecho /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/$WWPN/unit_add $LUN \ + || echo "Could not add LUN $LUN at WWPN $WWPN on FCP device $CD_DEVICE" break ;; n|N|[Nn][Oo]) @@ -2829,7 +2727,7 @@ if [ -r /sys/firmware/ipl/ipl_type ]; then ;; *) echo - echo $"*** INVALID ANSWER: $do_cd_install" + echo "*** INVALID ANSWER: $do_cd_install" echo unset do_cd_install ;; @@ -2846,7 +2744,6 @@ fi # [ -z "${cardtype//OSD_*/}" ] can be used to check for real OSA -# Parse configuration # Check for missing parameters, prompt for them if necessary while : ; do @@ -2859,10 +2756,10 @@ while : ; do -o -z "$RUNKS" -a $ -z "$NETTYPE" -o -z "$SUBCHANNELS" $ ] && \ dialog_network_table if isVM; then - echo $"* NOTE: To enter default or empty values press enter twice. *" + echo "* NOTE: To enter default or empty values press enter twice. *" fi do_nettype - tv load_kernel_modules + # precondition: driver (qeth/lcs/ctcm/netiucv) loaded incl. dependencies # only works, if ipv6 kernel module has been loaded before: [ "$NETTYPE" = "qeth" ] && tv disable_ipv6_autoconf if [ "$NETTYPE" != "iucv" ]; then @@ -2886,7 +2783,7 @@ while : ; do fi do_layer2 # set device online to know the device name - # and to know if it's OSD/HiperSocket/GuestLAN BUT do not + # and to know if it's OSD/HiperSockets/GuestLAN BUT do not # try to ifconfig the device up since that requires # setting the mac address before (if applicable). set_device_online || workflow_item_menu noredo @@ -2963,16 +2860,13 @@ while : ; do do_dasd - echo $"Initial configuration completed." + echo "Initial configuration completed." final_check && break rollback_config reenter="yes" done # outer dialog loop -#echo $"Starting portmap." -#portmap - if [ -z "$testing" ]; then # convert to space-separated lists @@ -2989,18 +2883,13 @@ if [ -z "$testing" ]; then for i in $RESOLVDNS; do echo "nameserver $i"; done >> /etc/resolv.conf fi - # make sure we have an /etc/hosts file (required for telnetd) + # make sure we have an /etc/hosts file (originally required for telnetd) if [ ! -z "$HOSTNAME" -a ! -z "$IPADDR" ]; then - echo -e "$IPADDR\t$HOSTNAME `echo $HOSTNAME | cut -d '.' -f 1`" >> /etc/hosts + echo -e "$IPADDR\t$HOSTNAME $(echo $HOSTNAME | cut -d '.' -f 1)" >> /etc/hosts fi fi # testing -if [ -z "$testing" ]; then - # undo startup of udevd, only needed it for possible calls of udevsettle - killall udevd -fi # testing - if [ -n "$DASD" ]; then echo "DASD=$DASD" > /tmp/dasd_ports fi @@ -3015,16 +2904,6 @@ done # cio_ignore handling for FCP should happen when the content of /tmp/fcpconfig # will actually be processed which is in anaconda's zfcp.py ZFCP::readConfig() -if [ -z "$testing" ]; then - - grep -q ext3 /proc/filesystems - if [ "$?" != "0" ]; then - insmod jbd$LO - insmod ext3$LO - fi - -fi # testing - # TODO/FIXME: also need to pass IPv6 decision to loader/anaconda # [ "$ipv6" ] && echo "IPV6=yes" @@ -3032,8 +2911,6 @@ fi # testing cat > /tmp/install.cfg << EOF LANG="$LANG" S390ARCH="$S390ARCH" -TEXTDOMAIN="$TEXTDOMAIN" -TEXTDOMAINDIR="$TEXTDOMAINDIR" PORTNAME="$PORTNAME" HOSTNAME="$HOSTNAME" DEVICE="$DEVICE" @@ -3051,12 +2928,12 @@ ONBOOT="yes" CTCPROT="$CTCPROT" EOF if [ "$ipv6" ]; then - echo DNS=\"`echo $DNS | cut -d ',' -f 1`\" >> /tmp/install.cfg + echo DNS=\"$(echo $DNS | cut -d ',' -f 1)\" >> /tmp/install.cfg else - echo DNS=\"`echo $DNS | cut -d ':' -f 1`\" >> /tmp/install.cfg + echo DNS=\"$(echo $DNS | cut -d ':' -f 1)\" >> /tmp/install.cfg fi cat >> /tmp/install.cfg << EOF -export LANG PORTNAME S390ARCH TEXTDOMAIN TEXTDOMAINDIR +export LANG PORTNAME S390ARCH export HOSTNAME DEVICE NETTYPE IPADDR GATEWAY MTU export NETWORK NETMASK BROADCAST DNS SEARCHDNS export PEERID ONBOOT SUBCHANNELS CTCPROT @@ -3066,7 +2943,14 @@ EOF if [ -z "$testing" ]; then cat /tmp/install.cfg >> /etc/profile fi # testing -cat > /tmp/netinfo << EOF + +NETSCRIPTS="/etc/sysconfig/network-scripts" +IFCFGFILE="$NETSCRIPTS/ifcfg-$DEVICE" +if [ ! -d "$NETSCRIPTS" ]; then + mkdir -p $NETSCRIPTS +fi + +cat > $IFCFGFILE << EOF DEVICE=$DEVICE ONBOOT=yes BOOTPROTO=static @@ -3077,34 +2961,35 @@ MTU=$MTU SUBCHANNELS=$SUBCHANNELS EOF if [ "$ipv6" ]; then - [ "$DNS" != "" ] && echo "DNS=`echo $DNS | cut -d ',' -f 1`" >> /tmp/netinfo -else - [ "$DNS" != "" ] && echo "DNS=`echo $DNS | cut -d ':' -f 1`" >> /tmp/netinfo -fi -[ "$NETTYPE" != "" ] && echo "NETTYPE=$NETTYPE" >> /tmp/netinfo -[ "$PEERID" != "" ] && echo "PEERID=$PEERID" >> /tmp/netinfo -[ "$PORTNAME" != "" ] && echo "PORTNAME=$PORTNAME" >> /tmp/netinfo -[ "$CTCPROT" != "" ] && echo "CTCPROT=$CTCPROT" >> /tmp/netinfo -[ "$LAYER2" != "" ] && echo "LAYER2=$LAYER2" >> /tmp/netinfo -[ "$MACADDR" != "" ] && echo "MACADDR=$MACADDR" >> /tmp/netinfo -[ "$PORTNO" != "" ] && echo "PORTNO=$PORTNO" >> /tmp/netinfo -# also needs support in stage2 to make PORTNO persistent, -# i.e. OPTION="portno=1" in /etc/sysconfig/network-scripts/ifcfg-<netdevname> -# this is done by the loader starting with RHEL 5.3 -if [ "$ipv6" ]; then - cat >> /tmp/netinfo << EOF + cat >> $IFCFGFILE << EOF IPV6INIT=yes IPV6_AUTOCONF=no IPV6ADDR=$IPADDR/$NETMASK +IPV6_DEFAULTGW=$GATEWAY EOF # FIXME: /etc/sysconfig/network:IPV6_DEFAULTGW=$GATEWAY # /etc/sysconfig/network:NETWORKING_IPV6=yes else - cat >> /tmp/netinfo << EOF + cat >> $IFCFGFILE << EOF IPADDR=$IPADDR NETMASK=$NETMASK EOF fi +if [ "$ipv6" ]; then + [ "$DNS" != "" ] && echo "DNS=$(echo $DNS | cut -d ',' -f 1)" >> $IFCFGFILE +else + [ "$DNS" != "" ] && echo "DNS=$(echo $DNS | cut -d ':' -f 1)" >> $IFCFGFILE +fi +[ "$NETTYPE" != "" ] && echo "NETTYPE=$NETTYPE" >> $IFCFGFILE +[ "$PEERID" != "" ] && echo "PEERID=$PEERID" >> $IFCFGFILE +[ "$PORTNAME" != "" ] && echo "PORTNAME=$PORTNAME" >> $IFCFGFILE +[ "$CTCPROT" != "" ] && echo "CTCPROT=$CTCPROT" >> $IFCFGFILE +[ "$LAYER2" != "" ] && echo "LAYER2=$LAYER2" >> $IFCFGFILE +[ "$MACADDR" != "" ] && echo "MACADDR=$MACADDR" >> $IFCFGFILE +[ "$PORTNO" != "" ] && echo "PORTNO=$PORTNO" >> $IFCFGFILE +# also needs support in stage2 (loader.c,(net.c)) to make PORTNO persistent, +# i.e. OPTION="portno=1" in /etc/sysconfig/network-scripts/ifcfg-<netdevname> +# (see loader of RHEL 5.3) if [ -z "$testing" ]; then diff --git a/loader/lsznet.raw b/loader/lsznet.raw old mode 100644 new mode 100755 index 5b3df91..ac9bd07 --- a/loader/lsznet.raw +++ b/loader/lsznet.raw @@ -23,73 +23,29 @@ readonly DEBUG=0 # nothing to be changed below here -export TEXTDOMAIN=lsznet.raw - readonly CMD=${0##*/} function error() { - echo $"$CMD: ERROR: $*" 1>&2 + echo "$CMD: ERROR: $*" 1>&2 exit 1; } # currently requires bash version 3.0 or later -# (this seems reasonable since bash-3.0 has been shipped with -# RHEL 4.0, 4.1, 4.2, 4.3, 4.4, 4.5, 4.6, 4.7; -# bash-3.1 with RHEL 5.0, 5.1; and bash-3.2 with RHEL 5.2) -if [ ${BASH_VERSINFO[0]} -lt 3 -o \ - ${BASH_VERSINFO[1]} -lt 0 ]; then - error $"only works with BASH version 3.0 or later (current version used is ${BASH_VERSION})" -fi -# the script was designed for Linux kernel 2.6 and might work with newer ones -kernel_version=$(uname -r) -IFS=. -read krn_ver krn_patch krn_foo <<< "$kernel_version" -unset IFS -if [ $krn_ver -lt 2 -o $krn_patch -lt 6 ]; then - error $"only works for kernel versions 2.6 or probably later" -fi +. /sbin/controlunits.sh -# The following combinations of control unit type and model were taken from the -# Linux network device drivers for s390 on 2008-06-09 from Linux 2.6.25.4. -# The list (among other things) should be adapted, if any of those device +# The arrays (among other things) should be adapted, if any of those device # drivers start supporting different CU types/models. -# (Alternatively, the list could be generated by reading the modaliases -# directly out of the device driver modules: -# modinfo qeth/cu3088 | fgrep 'alias:' -# However, this would still require a list of device driver modules.) -readonly -a CU=( - 1731/01 - 1731/05 - 3088/08 - 3088/1f - 3088/1e - 3088/01 - 3088/60 - 3088/61 -) # $CU_CARDTYPE array is the only one which may contain entries with spaces readonly -a CU_CARDTYPE=( "OSA (QDIO)" - "HiperSocket" + "HiperSockets" "CTC adapter" "escon channel" "ficon channel" "LCS p390" "LCS OSA" - "LCS CLAW" -) - -readonly -a CU_DEVDRV=( - qeth - qeth - ctc - ctc - ctc - lcs - lcs - lcs ) readonly -a CU_DEVNAME=( @@ -100,7 +56,6 @@ readonly -a CU_DEVNAME=( ctc eth eth - eth ) readonly -a CU_GROUPCHANNELS=( @@ -111,8 +66,8 @@ readonly -a CU_GROUPCHANNELS=( 2 2 2 - 2 ) + readonly -a CHPIDTYPES=( [10]=OSE [11]=OSD @@ -131,21 +86,6 @@ function debug() { [ $DEBUG -ge $level ] && echo "$*" 1>&2 } -# Searches for a match of argument 1 on the array $CU and sets $cu_idx -# to the matched array index on success. -# Returns 0 on success, 1 on failure. -function search_cu() { - local scu=$1 - local i - for ((i=0; i < ${#CU[@]}; i++)); do - if [ $scu == ${CU[i]} ]; then - cu_idx=$i - return 0 - fi - done - return 1 -} - # Returns symbolic name of CHPID type in $chpidtype_symbolic, # if an entry in the array $CHPIDTYPES has been found at index of argument 1. # Returns "?" otherwise. @@ -198,10 +138,10 @@ function build_list() { continue ;; 2) - error $"syntax error in regex of match operator =~, code needs to be fixed" + error "syntax error in regex of match operator =~, code needs to be fixed" ;; *) - error $"unexpected return code of regex match operator =~, code needs to be fixed" + error "unexpected return code of regex match operator =~, code needs to be fixed" ;; esac debug 5 " sysfs directory matched regex $dir" @@ -218,7 +158,7 @@ function build_list() { read chpid_list < $subch_p/chpids read -a chpids <<< "$chpid_list" if [ ${#chpids[@]} -ne 8 ]; then - error $"sysfs reported ${#chpids[@]} CHPIDs instead of expected 8" + error "sysfs reported ${#chpids[@]} CHPIDs instead of expected 8" fi read pim pam pom foo < $subch_p/pimpampom pimchpidZ="" @@ -250,13 +190,13 @@ function build_list() { read chptype < $chpid_p/type # filter and output... if search_cu $cutype; then - if [ "${CU_DEVDRV[$cu_idx]}" == "ctc" ]; then + if [ "${CU_DEVDRV[$cu_idx]}" == "ctcm" ]; then # assume CTC are mostly virtual and ignore chpid from sysfs chpidtype_symbolic="-" else search_chpt $chptype fi - echo $pimchpids $devbusid $STARTPATH/$dev_p $cutype $chpidtype_symbolic $chptype ${CU_DEVDRV[$cu_idx]} ${CU_DEVNAME[$cu_idx]} ${CU_GROUPCHANNELS[$cu_idx]} $cu_idx ${CU_CARDTYPE[$cu_idx]} + echo $pimchpids $devbusid $cutype $chpidtype_symbolic ${CU_DEVDRV[$cu_idx]} ${CU_DEVNAME[$cu_idx]} ${CU_GROUPCHANNELS[$cu_idx]} ${CU_CARDTYPE[$cu_idx]} else debug 5 " skip non-network device $devbusid CU $cutype" fi @@ -278,7 +218,7 @@ function search_groups() { local w_prefix w_ssid w_devno local d_prefix d_ssid d_devno local prefix ssid devno x - local chp devbusid dev_p cutype chpidtypename chptype devdrv devname groupchs cu_idx cardtype + local chp devbusid cutype chpidtypename devdrv devname groupchs cardtype # remembered last state variables for possible ccwgroup: local r_prefix="Z" local r_ssid="Z" @@ -288,9 +228,8 @@ function search_groups() { local count=0 local item=1 local skipped=0 - # currently unused are: dev_p,chptype, cu_idx. - while read chp devbusid dev_p cutype chpidtypename chptype devdrv devname groupchs cu_idx cardtype; do - debug 1 " # $chp $devbusid $dev_p $cutype $chpidtypename $chptype $devdrv $devname $groupchs $cu_idx $cardtype" + while read chp devbusid cutype chpidtypename devdrv devname groupchs cardtype; do + debug 1 " # $chp $devbusid $cutype $chpidtypename $devdrv $devname $groupchs $cardtype" IFS=. read prefix ssid devno x <<< "$devbusid" unset IFS @@ -348,7 +287,7 @@ function search_groups() { chlist=$r_prefix.$r_ssid.$r_devno,$w_prefix.$w_ssid.$w_devno,$d_prefix.$d_ssid.$d_devno ;; *) - error $"unknown number of channels for group, code needs to be fixed" + error "unknown number of channels for group, code needs to be fixed" ;; esac echo $item $cutype $chp $chpidtypename $devdrv $devname $chlist "$cardtype" -- 1.6.3.1

1 1

[PATCH] Do not go interactive if timezone in ks is not valid (#473647).
by Radek Vykydal 07 Aug '09

07 Aug '09

Handle invalid timezones in ks the same way as in RHEL 5.4: in case that the timezone is invalid, the system time of install environment will be set correctly if it is in tzdata package, and even if it is not there the ks install will proceed. This will prevent possible breaking of installs with kickstarts working in older versions of RHEL. The question may be if we can/want to broaden valid timezones to all those present in tzdata package officialy (document it). Complete tzdata package is part of stage2 in Fedora and will be in RHEL 5.4. --- kickstart.py | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/kickstart.py b/kickstart.py index 474ecec..ae6ae3a 100644 --- a/kickstart.py +++ b/kickstart.py @@ -881,8 +881,7 @@ class Timezone(commands.timezone.FC6_Timezone): tab = zonetab.ZoneTab() if self.timezone not in (entry.tz.replace(' ','_') for entry in tab.getEntries()): - log.warning("Timezone %s set in kickstart is not valid, will ask" % (self.timezone,)) - return retval + log.warning("Timezone %s set in kickstart is not valid." % (self.timezone,)) self.handler.id.timezone.setTimezoneInfo(self.timezone, self.isUtc) self.handler.skipSteps.append("timezone") -- 1.5.4.3

2 1

[PATCH 1/8] Add functions to go from an iScsiDiskDevice to an libiscsi node
by Hans de Goede 04 Aug '09

04 Aug '09

Add functions to go from an iScsiDiskDevice to an libiscsi node and the other way around. This is a preparation patch for adding support for writing the necessary dracut cmdline options to grub.conf --- storage/iscsi.py | 30 +++++++++++++++++++++++------- 1 files changed, 23 insertions(+), 7 deletions(-) diff --git a/storage/iscsi.py b/storage/iscsi.py index cfe0c55..13ddc08 100644 --- a/storage/iscsi.py +++ b/storage/iscsi.py @@ -268,13 +268,10 @@ class iscsi(object): # devices used for root get started by the initrd if root.dependsOn(disk): continue - # find the iscsi node matching this disk - for node in self.nodes: - if node.name == disk.iscsi_name and \ - node.address == disk.iscsi_address and \ - node.port == disk.iscsi_port: - node.setParameter("node.startup", "automatic") - break + + node = self.getNode(disk.iscsi_name, disk.iscsi_address, disk.iscsi_port) + if node: + node.setParameter("node.startup", "automatic") if not os.path.isdir(instPath + "/etc/iscsi"): os.makedirs(instPath + "/etc/iscsi", 0755) @@ -289,4 +286,23 @@ class iscsi(object): shutil.copytree("/var/lib/iscsi", instPath + "/var/lib/iscsi", symlinks=True) + def getNode(self, name, address, port): + for node in self.nodes: + if node.name == name and node.address == address and \ + node.port == port: + return node + + return None + + def getNodeDisks(self, node, storage): + nodeDisks = [] + iscsiDisks = storage.devicetree.getDevicesByType("iscsi") + for disk in iscsiDisks: + if node.name == disk.iscsi_name and \ + node.address == disk.iscsi_address and \ + node.port == disk.iscsi_port: + nodeDisks.append(disk) + + return nodeDisks + # vim:tw=78:ts=4:et:sw=4 -- 1.6.2.2

2 8

[PATCH] Add yet more crud so we can run /sbin/start_udev instead of udev (#514501).
by Chris Lumens 31 Jul '09

31 Jul '09

start_udev appears to properly build the udev database, which otherwise is not fully populated for networking unless we run udev_settle multiple times. --- loader/init.c | 2 +- scripts/mk-images | 9 +++++++++ scripts/upd-instroot | 5 ++++- 3 files changed, 14 insertions(+), 2 deletions(-) diff --git a/loader/init.c b/loader/init.c index ff0f62e..594ebf1 100644 --- a/loader/init.c +++ b/loader/init.c @@ -512,7 +512,7 @@ int main(int argc, char **argv) { printf("done\n"); printf("starting udev..."); if (fork() == 0) { - execl("/sbin/udevd", "/sbin/udevd","--daemon",NULL); + execl("/sbin/start_udev", "/sbin/start_udev", NULL); exit(1); } } diff --git a/scripts/mk-images b/scripts/mk-images index 2a539a8..eed7daf 100755 --- a/scripts/mk-images +++ b/scripts/mk-images @@ -623,6 +623,7 @@ makeinitrd() { instbin $IMGPATH /usr/sbin/$cmd $MBD_DIR /sbin/$cmd done + instbin $IMGPATH /sbin/start_udev $MBD_DIR /sbin/start_udev instbin $IMGPATH /usr/sbin/udevd $MBD_DIR /sbin/udevd instbin $IMGPATH /usr/sbin/udevadm $MBD_DIR /sbin/udevadm instbin $IMGPATH /usr/bin/udevinfo $MBD_DIR /sbin/udevinfo @@ -637,6 +638,10 @@ makeinitrd() { cp -a functions $MBD_DIR/etc/rc.d/init.d ) + ( cd $IMGPATH/etc/sysconfig + cp -a udev $MBD_DIR/etc/sysconfig/udev + ) + ( cd $IMGPATH/etc/sysconfig/network-scripts cp -a network-functions $MBD_DIR/etc/sysconfig/network-scripts cp -a network-functions-ipv6 $MBD_DIR/etc/sysconfig/network-scripts @@ -728,6 +733,8 @@ makeinitrd() { # Misc instbin $IMGPATH /usr/bin/awk $MBD_DIR /sbin/awk + instbin $IMGPATH /usr/bin/cat $MBD_DIR /sbin/cat + instbin $IMGPATH /usr/bin/chown $MBD_DIR /sbin/chown instbin $IMGPATH /usr/bin/egrep $MBD_DIR /sbin/egrep instbin $IMGPATH /usr/bin/fgrep $MBD_DIR /sbin/fgrep instbin $IMGPATH /usr/bin/gawk $MBD_DIR /sbin/gawk @@ -735,6 +742,8 @@ makeinitrd() { instbin $IMGPATH /usr/bin/kill $MBD_DIR /sbin/kill instbin $IMGPATH /usr/bin/ln $MBD_DIR /sbin/ln instbin $IMGPATH /usr/bin/mkdir $MBD_DIR /sbin/mkdir + instbin $IMGPATH /usr/bin/mknod $MBD_DIR /sbin/mknod + instbin $IMGPATH /usr/sbin/pidof $MBD_DIR /sbin/pidof instbin $IMGPATH /usr/bin/readlink $MBD_DIR /sbin/readlink instbin $IMGPATH /usr/bin/rm $MBD_DIR /sbin/rm instbin $IMGPATH /usr/bin/rmdir $MBD_DIR /sbin/rmdir diff --git a/scripts/upd-instroot b/scripts/upd-instroot index 80b69f0..9db1fb5 100755 --- a/scripts/upd-instroot +++ b/scripts/upd-instroot @@ -364,6 +364,7 @@ etc/selinux/targeted etc/services etc/shells etc/sysconfig/network-scripts/network-functions* +etc/sysconfig/udev etc/udev etc/wpa_supplicant/wpa_supplicant.conf etc/yum.repos.d/* @@ -426,6 +427,7 @@ sbin/mount.ntfs* sbin/parted sbin/pcmcia-socket-startup sbin/pdisk +sbin/pidof sbin/probe sbin/reiserfsck sbin/reiserfstune @@ -434,6 +436,7 @@ sbin/route sbin/setfiles sbin/sfdisk sbin/silo +sbin/start_udev sbin/swapoff sbin/swapon sbin/tune2fs @@ -958,7 +961,7 @@ if [ -f $DEST/bin/bash ]; then fi if [ -f $DEST/bin/gawk ]; then - ln -sf awk $DEST/bin/gawk + ln -sf $DEST/bin/gawk awk fi [ -d $DEST/bin ] || die "ERROR: directory missing: $DEST/bin" -- 1.6.1.3

3 4

Build anaconda 12 for F11 for Kickstart?
by Nicholas Veeser 30 Jul '09

30 Jul '09

Trying to make Raid/kickstart work with F11. Need to be able to rebuilding / noformat a raid partition on install. Tried adding update.img from this bug: https://bugzilla.redhat.com/show_bug.cgi?id=503310 Fixed the problem, but produced a different bug (md device was detected as swap). Since this is not a real build of anaconda, I assume a bug is not appropriate. Fedora 11 kickstart/anaconda is broken for me. 1. Is there a known method for pushing updates to anaconda for install media? (am I missing something obvious) 2. Is there a method to build (install.img or update.img), current anaconda (ie 12), to work with the previous release of Fedora (ie 11)? Seems I can pull down git:anaconda@anaconda-12.3-1. I can "compile" with make. How can I make an appropriate update.img or install.img? I would be willing to write up a wiki page about this if I could get started. I assume that 'backporting' from Rawhide to the Last Release would useful if there is not already a mechanism for this. Thanx Nicholas

1 0

[PATCH] Let's not exit from buildinstall.functions, say, ever
by Jeremy Katz 30 Jul '09

30 Jul '09

Background: looking into why we don't have a gtk theme on x86_64 (#513879) and everything looks okay... /tmp/keepfile* is correct, etc. And it works on i386. In addition to no theme stuff, though, there's also not /sbin/grub. So maybe /sbin/grub has some non-static 32bit linkage now and that's tripping up our shlib stuff. I haven't tested (my box seems to have killed itself overnight) but this shouldn't break anything at least and might possibly fix things Exiting from a function being called could leave things in a weird state. The exit makes sense for mkinitrd, maybe less so when building installer images --- scripts/buildinstall.functions | 7 ++----- 1 files changed, 2 insertions(+), 5 deletions(-) diff --git a/scripts/buildinstall.functions b/scripts/buildinstall.functions index 3b10dff..2afb652 100755 --- a/scripts/buildinstall.functions +++ b/scripts/buildinstall.functions @@ -56,11 +56,8 @@ EOF while [ $n -lt ${#FILES[*]} ]; do FILE="${FILES[$n]}" if [ "$FILE" == "not found" ]; then - cat 1>&2 <<EOF -There are missing files on your system. The dynamic object $bin -requires ${NAMES[$n]} n order to properly function. mkinitrd cannot continue. -EOF - exit 1 + echo "WARNING: The dynamic object $bin requires ${NAMES[$n]} n order to properly function." + continue fi case "$FILE" in /lib*) -- 1.6.1

2 1

[PATCH] master. The new create.
by Joel Granados Moreno 30 Jul '09

30 Jul '09

Hello list Been working on a Custom Partitioning workflow. Based on https://fedoraproject.org/wiki/Design/AnacondaStorageUI/MiscMockupIdeas In this patchset there are two big changes: 1. In the LV edit the user no longer has to go through the intermediary step of the Volume Group. 2. There are only 4 buttons in the main customization window. The new "Create" button replaces the LVM, RAID and New buttons. It will automatically search for the valid options (So the user wont try to create an lvm when there are no PV present.) For 2. There are also some new information strings. The RAID info is taken from the old messages so no translation would be needed. There is an LVM message. Its long and I would like to see it a little shorter (anyone want to give it a try). Caveats: 1. LVM Logical Volume does not work yet. We need to relate it to a VG someway, not sure how to do that yet. The LV can still be created through the VG interface. 2. Definitely thinking of changing hasFreeHardDrive for something in storage.__init__. 3. This adds some click to some parts of the install but we add consistency and context info. We also check stuff before errors happen, so we don't need some error messages (Like trying to create partition when there is no free space) Review appreciated. regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

3 9

persistent device names, anaconda, udev, etc.
by Bill Nottingham 30 Jul '09

30 Jul '09

In general, we want the following in the OS: 1) Device names stay consistent across reboots The current persistent-net.rules udev files, and the current generator files, handle this. 2) Device names stay consistent across reinstalls Currently, anaconda doesn't use any existing information when starting udev (as far as I know), so interfaces can be in random order depending on timing. We don't currently have a mechanism for this. 3) Device names stay (reasonably) consistent across hardware swaps or image distribution to multiple machines (including virt) i.e., if the image is deployed to a new machine, the devices stay the same. We don't currently have a mechanism for this. So, how to handle this? For #2, we obviously can't have a set of rule files that does this; at least not well. (It would need to go back and rename other devices that we've already gotten events for, for example.) We'd need some sort of code that runs after udevsettle that applies an ordering. For #3, the solution proffered is to have the udev writer write two sets of rules: - one that maps based on MAC address - one that maps based on PCI/USB/etc. path The user could then set a configuration parameter that describes which ones would actually be used. Is any of this practical? Is any of this something that we can get done with existing resources? Bill

2 2

[PATCH] Do not offer going back when ugrade root for ks upgrade is not found (#499321)
by Radek Vykydal 30 Jul '09

30 Jul '09

Going back results in traceback which seems not right. Bug 499321 (preupgrade backtrace) has lots of tracebacks with various causes of upgrade root not found - I distincted two groups.: 1) dmraid not detected in f10 (=> installation on 2 disks) while detected in f11 (=> root not found). 2) root considered as non-upgradable (/etc/redhat-release changed by user). Such a broad bug is rather difficult to handle. OTOH we wouldn't get the tracebacks so easy if we just ended with error message (as with the patch). --- upgrade.py | 28 ++++++++++++++++++---------- 1 files changed, 18 insertions(+), 10 deletions(-) diff --git a/upgrade.py b/upgrade.py index c94cc3a..48edff0 100644 --- a/upgrade.py +++ b/upgrade.py @@ -188,18 +188,26 @@ def upgradeMountFilesystems(anaconda): except IndexError as e: # The upgrade root is search earlier but we give the message here. log.debug("No upgrade root was found.") - rc = anaconda.intf.messageWindow(_("Upgrade root not found"), + if anaconda.isKickstart and anaconda.id.ksdata.upgrade.upgrade: + anaconda.intf.messageWindow(_("Upgrade root not found"), _("The root for the previously installed system was not " - "found. You can exit installer or backtrack to choose " - "installation instead of upgrade."), - type="custom", - custom_buttons = [ _("_Back"), - _("_Exit installer") ], - custom_icon="question") - if rc == 0: - return DISPATCH_BACK - elif rc == 1: + "found."), type="custom", + custom_icon="info", + custom_buttons=[_("Exit installer")]) sys.exit(0) + else: + rc = anaconda.intf.messageWindow(_("Upgrade root not found"), + _("The root for the previously installed system was not " + "found. You can exit installer or backtrack to choose " + "installation instead of upgrade."), + type="custom", + custom_buttons = [ _("_Back"), + _("_Exit installer") ], + custom_icon="question") + if rc == 0: + return DISPATCH_BACK + elif rc == 1: + sys.exit(0) checkLinks = ( '/etc', '/var', '/var/lib', '/var/lib/rpm', '/boot', '/tmp', '/var/tmp', '/root', -- 1.6.0.6

2 1

Remove PACKAGES and PACKAGESRESCUE from upd-instroot
by David Cantrell 30 Jul '09

30 Jul '09

Based on notting's comment about removing the lib packages from PACKAGES, I thought we should be able to let upd-instroot do 'yum install anaconda' and get the tree we need. So I looked at what yum install anaconda would pull in at first, then moved remaining things that are listed in KEEPFILE* lists to Requires: tags in the spec file. This can also make life easier for lorax, as we could rely on 'yum install anaconda' to set up our tree more or less with the files we need. The first patch in this set is removing the package names that are pulled in when depsolving anaconda, as well as moving the font packages to their own list. The second patch is where I go and destroy the PACKAGES lists entirely. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

5 6

[PATCH] Make upgradeany boot option work again (#513227)
by Radek Vykydal 30 Jul '09

30 Jul '09

--- iw/examine_gui.py | 3 ++- rescue.py | 2 +- upgrade.py | 7 ++++--- 3 files changed, 7 insertions(+), 5 deletions(-) diff --git a/iw/examine_gui.py b/iw/examine_gui.py index 3a62e06..20af0c7 100644 --- a/iw/examine_gui.py +++ b/iw/examine_gui.py @@ -93,7 +93,8 @@ class UpgradeExamineWindow (InstallWindow): # we might get here after storage reset that obsoleted # root device objects we had found if not self.anaconda.id.rootParts: - self.anaconda.id.rootParts = upgrade.findExistingRoots(self.anaconda) + self.anaconda.id.rootParts = upgrade.findExistingRoots(self.anaconda, + upgradeany=flags.cmdline.has_key("upgradeany")) upgrade.setUpgradeRoot(self.anaconda) self.parts = self.anaconda.id.rootParts diff --git a/rescue.py b/rescue.py index 11da108..fe2d181 100644 --- a/rescue.py +++ b/rescue.py @@ -297,7 +297,7 @@ def runRescue(anaconda, instClass): import storage storage.storageInitialize(anaconda) - disks = upgrade.findExistingRoots(anaconda, upgradeany = 1) + disks = upgrade.findExistingRoots(anaconda, upgradeany=True) if not disks: root = None diff --git a/upgrade.py b/upgrade.py index c8b6757..43400b3 100644 --- a/upgrade.py +++ b/upgrade.py @@ -76,7 +76,8 @@ def findRootParts(anaconda): if anaconda.dir == DISPATCH_BACK: return if anaconda.id.rootParts is None: - anaconda.id.rootParts = findExistingRoots(anaconda) + anaconda.id.rootParts = findExistingRoots(anaconda, + upgradeany=flags.cmdline.has_key("upgradeany")) setUpgradeRoot(anaconda) @@ -88,10 +89,10 @@ def findRootParts(anaconda): anaconda.dispatch.skipStep("findinstall", skip = 1) anaconda.dispatch.skipStep("installtype", skip = 0) -def findExistingRoots(anaconda, upgradeany = 0): +def findExistingRoots(anaconda, upgradeany=False): if not flags.setupFilesystems: (prod, ver) = getReleaseString (anaconda.rootPath) - if flags.cmdline.has_key("upgradeany") or upgradeany == 1 or anaconda.id.instClass.productUpgradable(prod, ver): + if flags.cmdline.has_key("upgradeany") or upgradeany or anaconda.id.instClass.productUpgradable(prod, ver): return [(anaconda.rootPath, "")] return [] -- 1.6.0.6

2 2

[PATCH] Rework shutDown() to better accomidate "nokill" better.
by Peter Jones 29 Jul '09

29 Jul '09

Rework shutDown() so that we don't do unmounts until we're going to reboot or power off the machine. Also some minor cleanups like taking out all the negative-phrasing of "noKill" in the code itself. --- loader/init.c | 27 ++++++++-------- loader/init.h | 28 +++++++++++++++++ loader/shutdown.c | 87 ++++++++++++++++++++++++++++++++++------------------- 3 files changed, 97 insertions(+), 45 deletions(-) create mode 100644 loader/init.h diff --git a/loader/init.c b/loader/init.c index 1c275d8..ff0f62e 100644 --- a/loader/init.c +++ b/loader/init.c @@ -53,6 +53,7 @@ #include <termios.h> #include <libgen.h> +#include "init.h" #include "copy.h" #include "devt.h" #include "devices.h" @@ -110,10 +111,8 @@ char * env[] = { */ int testing=0; -void unmountFilesystems(void); -void disableSwap(void); -void shutDown(int noKill, int doReboot, int doPowerOff); -static int getNoKill(void); +void shutDown(int doKill, reboot_action rebootAction); +static int getKillPolicy(void); struct termios ts; static void printstr(char * string) { @@ -396,22 +395,22 @@ static void termReset(void) { /* reboot handler */ static void sigintHandler(int signum) { termReset(); - shutDown(getNoKill(), 1, 0); + shutDown(getKillPolicy(), REBOOT); } /* halt handler */ static void sigUsr1Handler(int signum) { termReset(); - shutDown(getNoKill(), 0, 0); + shutDown(getKillPolicy(), HALT); } /* poweroff handler */ static void sigUsr2Handler(int signum) { termReset(); - shutDown(getNoKill(), 0, 1); + shutDown(getKillPolicy(), POWEROFF); } -static int getNoKill(void) { +static int getKillPolicy(void) { int fd; int len; char buf[1024]; @@ -421,9 +420,9 @@ static int getNoKill(void) { len = read(fd, buf, sizeof(buf) - 1); close(fd); if ((len > 0) && strstr(buf, "nokill")) - return 1; + return 0; } - return 0; + return 1; } static int getInitPid(void) { @@ -453,7 +452,7 @@ int main(int argc, char **argv) { int doShutdown =0; int isSerial = 0; char * console = NULL; - int noKill = 0; + int doKill = 1; char * argvc[15]; char ** argvp = argvc; char twelve = 12; @@ -541,7 +540,7 @@ int main(int argc, char **argv) { } } - noKill = getNoKill(); + doKill = getKillPolicy(); #if !defined(__s390__) && !defined(__s390x__) static struct termios orig_cmode; @@ -749,7 +748,7 @@ int main(int argc, char **argv) { printf("running %s\n", argvc[0]); execve(argvc[0], argvc, env); - shutDown(1, 0, 0); + shutDown(1, HALT); } /* signal handlers for halt/poweroff */ @@ -797,7 +796,7 @@ int main(int argc, char **argv) { if (testing) exit(0); - shutDown(noKill, doReboot, 0); + shutDown(doKill, doReboot?REBOOT:HALT); return 0; } diff --git a/loader/init.h b/loader/init.h new file mode 100644 index 0000000..733bc8e --- /dev/null +++ b/loader/init.h @@ -0,0 +1,28 @@ +/* + * init.h + * + * Copyright (C) 2009 Red Hat, Inc. All rights reserved. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ +#ifndef INIT_H +#define INIT_H + +typedef enum { + REBOOT, + POWEROFF, + HALT +} reboot_action; + +#endif /* INIT_H */ diff --git a/loader/shutdown.c b/loader/shutdown.c index 044d021..54fd9a5 100644 --- a/loader/shutdown.c +++ b/loader/shutdown.c @@ -30,6 +30,8 @@ #include <sys/types.h> #include <unistd.h> +#include "init.h" + #ifdef AS_SHUTDOWN int testing = 0; #else @@ -39,25 +41,11 @@ extern int testing; void disableSwap(void); void unmountFilesystems(void); -static void rebootHandler(int signum) { - printf("rebooting system\n"); -#if USE_MINILIBC - reboot(0xfee1dead, 672274793, 0x1234567); -#else - reboot(RB_AUTOBOOT); -#endif -} - -void shutDown(int noKill, int doReboot, int doPowerOff) { - sync(); sync(); +static void performTerminations(int doKill) { + if (testing || !doKill) + return; - printf("disabling swap...\n"); - disableSwap(); - - printf("unmounting filesystems...\n"); - unmountFilesystems(); - - if (!testing && !noKill) { + sync(); printf("sending termination signals..."); kill(-1, 15); sleep(2); @@ -67,25 +55,60 @@ void shutDown(int noKill, int doReboot, int doPowerOff) { kill(-1, 9); sleep(2); printf("done\n"); - } +} - if (doReboot) { - printf("rebooting system\n"); - sleep(2); +static void performUnmounts(int doKill) { + if (testing || !doKill) + return; + + printf("disabling swap...\n"); + disableSwap(); + + printf("unmounting filesystems...\n"); + unmountFilesystems(); +} + +static void performReboot(reboot_action rebootAction) { + if (rebootAction == POWEROFF) { + printf("powering off system\n"); + sleep(2); + reboot(RB_POWER_OFF); + } else if (rebootAction == REBOOT) { + printf("rebooting system\n"); + sleep(2); #if USE_MINILIBC - reboot(0xfee1dead, 672274793, 0x1234567); + reboot(0xfee1dead, 672274793, 0x1234567); #else - reboot(RB_AUTOBOOT); + reboot(RB_AUTOBOOT); #endif - } else if (doPowerOff) { - printf("powering off system\n"); - reboot(RB_POWER_OFF); - } else { + } +} + +static int shouldReboot = 0; + +static void rebootHandler(int signum) { + shouldReboot = 1; +} + +void shutDown(int doKill, reboot_action rebootAction) { + if (rebootAction == POWEROFF || rebootAction == REBOOT) { + performUnmounts(doKill); + performTerminations(doKill); + if (!doKill) + performReboot(rebootAction); + } + printf("you may safely reboot your system\n"); - signal(SIGINT, rebootHandler); - while (1) sleep(60); - } + signal(SIGINT, rebootHandler); + while (1) { + sleep(1); + if (shouldReboot) { + performUnmounts(1); + performTerminations(1); + performReboot(REBOOT); + } + } exit(0); @@ -122,3 +145,5 @@ int main(int argc, char ** argv) { return 0; } #endif + +/* vim:set shiftwidth=4 softtabstop=4 ts=4: */ -- 1.6.4

2 1

[PATCH 1/5] Put mkdir into /sbin on the initrd, too.
by Chris Lumens 29 Jul '09

29 Jul '09

--- scripts/mk-images | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/scripts/mk-images b/scripts/mk-images index f746a29..8879d5d 100755 --- a/scripts/mk-images +++ b/scripts/mk-images @@ -728,6 +728,7 @@ makeinitrd() { instbin $IMGPATH /usr/bin/grep $MBD_DIR /sbin/grep instbin $IMGPATH /usr/bin/kill $MBD_DIR /sbin/kill instbin $IMGPATH /usr/bin/ln $MBD_DIR /sbin/ln + instbin $IMGPATH /usr/bin/mkdir $MBD_DIR /sbin/mkdir instbin $IMGPATH /usr/bin/readlink $MBD_DIR /sbin/readlink instbin $IMGPATH /usr/bin/rm $MBD_DIR /sbin/rm instbin $IMGPATH /usr/bin/rmdir $MBD_DIR /sbin/rmdir -- 1.6.1.3

2 8

[PATCH] Update device.map when upgrading (#513393)
by Radek Vykydal 29 Jul '09

29 Jul '09

The bug concerns update of grub when device names changed between install and upgrade. I am in doubts - are we supposed to handle such a case? How much of corner is the case? Isn't bootloader reinstall (safer) answer in such cases? --- booty/x86.py | 28 ++++++++++++++++++++++++++++ 1 files changed, 28 insertions(+), 0 deletions(-) diff --git a/booty/x86.py b/booty/x86.py index 3890e69..f90cb86 100644 --- a/booty/x86.py +++ b/booty/x86.py @@ -417,6 +417,31 @@ class x86BootloaderInfo(efiBootloaderInfo): return config + def updateDeviceMap(self, instRoot): + if os.access(instRoot + "/boot/grub/device.map", os.R_OK): + f = open(instRoot + "/boot/grub/device.map", "r") + updatedlines = [] + update = False + for line in f: + line = line.strip() + if line.startswith('(hd'): + (grubdisk, path) = line.split()[:2] + i = int(grubdisk.lstrip('(hd ').rstrip(') ')) + actual_path = self.storage.devicetree.getDeviceByName(self.drivelist[i]).path + if path != actual_path: + line = "%s %s" % (grubdisk, actual_path) + update = True + updatedlines.append(line) + f.close() + + if update: + os.rename(instRoot + "/boot/grub/device.map", + instRoot + "/boot/grub/device.map.rpmsave") + f = open(instRoot + "/boot/grub/device.map", "w+") + upd_comment = "# file updated by anaconda\n" + f.write(upd_comment + '\n'.join(updatedlines) + '\n') + f.close() + # this is a hackish function that depends on the way anaconda writes # out the grub.conf with a #boot= comment # XXX this falls into the category of self.doUpgradeOnly @@ -450,6 +475,9 @@ class x86BootloaderInfo(efiBootloaderInfo): # migrate info to /etc/sysconfig/grub self.writeSysconfig(instRoot, theDev) + # update device.map + self.updateDeviceMap(instRoot) + # more suckage. grub-install can't work without a valid /etc/mtab # so we have to do shenanigans to get updated grub installed... # steal some more code above -- 1.6.0.6

2 1

[PATCH] Fix CDLABEL substitution in syslinux.cfg for x86 boot.iso
by Jeremy Katz 29 Jul '09

29 Jul '09

--- scripts/mk-images.x86 | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/scripts/mk-images.x86 b/scripts/mk-images.x86 index 2d94e06..f62a80d 100644 --- a/scripts/mk-images.x86 +++ b/scripts/mk-images.x86 @@ -116,7 +116,7 @@ makeBootImages() { mv $MBD_BOOTTREE/syslinux.cfg $MBD_BOOTTREE/isolinux.cfg # set up the label for finding stage2 with a hybrid iso - sed -i 's/initrd=initrd.img/initrd=initrd.img stage2=hd:LABEL=$CDLABEL/' $MBD_BOOTTREE/isolinux.cfg + sed -i "s/initrd=initrd.img/initrd=initrd.img stage2=hd:LABEL=$CDLABEL/" $MBD_BOOTTREE/isolinux.cfg # copy in memtest if present if [ -f $IMGPATH/usr/lib/anaconda-runtime/boot/memtest* ]; then -- 1.6.1

2 1

[PATCH] use python-meh for exception reporting
by Chris Lumens 28 Jul '09

28 Jul '09

The following series of patches changes anaconda over to using the new python-meh for exception handling. I have tested this, though something is bound to go wrong once we get all the packaging done and into a release. Before this can be pushed, the new python-meh package needs to make it in, though that's supposed to be in the works. Doing this will allow me to later get rid of rhpl.exception. I think there's a bug and patch filed against all components that use it now. At least, I know I've cleaned up my code. - Chris

2 13

[PATCH 1/2] Use keyboard bits from system-config-keyboard now
by Jeremy Katz 28 Jul '09

28 Jul '09

system-config-keyboard has the keyboard bits now, so let's use those rather than the ones from rhpl --- anaconda | 2 +- anaconda.spec.in | 2 +- instdata.py | 2 +- scripts/getkeymaps | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/anaconda b/anaconda index aef0c95..03dbe5c 100755 --- a/anaconda +++ b/anaconda @@ -774,7 +774,7 @@ if __name__ == "__main__": import instdata - import rhpl.keyboard as keyboard + import system_config_keyboard.keyboard as keyboard log.info("anaconda called with cmdline = %s" %(sys.argv,)) log.info("Display mode = %s" %(opts.display_mode,)) diff --git a/anaconda.spec.in b/anaconda.spec.in index 81adcd6..964e495 100644 --- a/anaconda.spec.in +++ b/anaconda.spec.in @@ -115,7 +115,7 @@ Requires: mdadm Requires: lvm2 Requires: util-linux-ng >= 2.15.1 %ifnarch s390 s390x ppc64 -Requires: system-config-keyboard +Requires: system-config-keyboard >= 1.3.0 %endif Requires: hal, dbus-python Requires: cracklib-python diff --git a/instdata.py b/instdata.py index 488616b..6700528 100644 --- a/instdata.py +++ b/instdata.py @@ -40,7 +40,7 @@ import shlex from flags import * from constants import * from simpleconfig import SimpleConfigFile -import rhpl.keyboard as keyboard +import system_config_keyboard.keyboard as keyboard from pykickstart.version import versionToString, DEVEL diff --git a/scripts/getkeymaps b/scripts/getkeymaps index 2e003a7..8436900 100755 --- a/scripts/getkeymaps +++ b/scripts/getkeymaps @@ -51,7 +51,7 @@ else PATTERN=i386 fi -MAPS=$(python -c "import rhpl.keyboard_models ; rhpl.keyboard_models.get_supported_models()") +MAPS=$(python -c "import system_config_keyboard.keyboard_models ; system_config_keyboard.keyboard_models.get_supported_models()") for map in $MAPS ; do eval find /lib/kbd/keymaps/$PATTERN -name "$map.map*.gz" | while read n; do -- 1.6.1

3 8

[PATCH] Just pull in all python modules for stage2
by Jeremy Katz 28 Jul '09

28 Jul '09

in stage2. Rather than trying to pick out python module deps, let's just include everything. If it's a huge size hit, we'll blacklist big things we don't need --- scripts/upd-instroot | 35 +---------------------------------- 1 files changed, 1 insertions(+), 34 deletions(-) diff --git a/scripts/upd-instroot b/scripts/upd-instroot index 5d01a83..8b97d74 100755 --- a/scripts/upd-instroot +++ b/scripts/upd-instroot @@ -457,28 +457,6 @@ usr/$LIBDIR/hal usr/$LIBDIR/libuser/* usr/$LIBDIR/pango usr/$LIBDIR/python?.? -usr/$LIBDIR/python?.?/email -usr/$LIBDIR/python?.?/site-packages/*.pth -usr/$LIBDIR/python?.?/site-packages/*selinux* -usr/$LIBDIR/python?.?/site-packages/*sqlite* -usr/$LIBDIR/python?.?/site-packages/_cracklibmodule.so -usr/$LIBDIR/python?.?/site-packages/_dbus* -usr/$LIBDIR/python?.?/site-packages/bdevid.so -usr/$LIBDIR/python?.?/site-packages/block -usr/$LIBDIR/python?.?/site-packages/cElementTree.so -usr/$LIBDIR/python?.?/site-packages/cairo -usr/$LIBDIR/python?.?/site-packages/firstboot -usr/$LIBDIR/python?.?/site-packages/gtk* -usr/$LIBDIR/python?.?/site-packages/gtk*/gtk -usr/$LIBDIR/python?.?/site-packages/ixf86configmodule.so -usr/$LIBDIR/python?.?/site-packages/libiscsimodule.so -usr/$LIBDIR/python?.?/site-packages/libusermodule.so -usr/$LIBDIR/python?.?/site-packages/_pedmodule.so -usr/$LIBDIR/python?.?/site-packages/parted/ -usr/$LIBDIR/python?.?/site-packages/rhpl -usr/$LIBDIR/python?.?/site-packages/rpmmodule.so -usr/$LIBDIR/python?.?/site-packages/xf86config.py -usr/$LIBDIR/python?.?/xml/etree/ usr/$LIBDIR/rpm/rpmpopt usr/$LIBDIR/libiscsi.so* usr/$LIBDIR/libsqlite3.so* @@ -540,18 +518,7 @@ usr/lib/anaconda/storage/devicelibs usr/lib/anaconda/storage/formats usr/lib/kernel-wrapper usr/lib/locale -usr/lib/python?.?/site-packages/bugzilla* -usr/lib/python?.?/site-packages/cracklib.py* -usr/lib/python?.?/site-packages/decorator.py* -usr/lib/python?.?/site-packages/dbus -usr/lib/python?.?/site-packages/dbus_bindings.py* -usr/lib/python?.?/site-packages/elementtree -usr/lib/python?.?/site-packages/iniparse -usr/lib/python?.?/site-packages/pykickstart -usr/lib/python?.?/site-packages/repomd -usr/lib/python?.?/site-packages/rpmUtils -usr/lib/python?.?/site-packages/urlgrabber -usr/lib/python?.?/site-packages/yum +usr/lib/python?.? usr/lib/rpm/macros usr/lib/rpm/rpmpopt usr/lib/rpm/rpmrc -- 1.6.1

2 1

Patches for master branch
by David Cantrell 28 Jul '09

28 Jul '09

Misc s390 patches. Removal of ChangeLog per 512502. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

2 6

16 packages referred to in upd-instroot but not available in repositories
by Jeroen van Meeuwen 27 Jul '09

27 Jul '09

Hi, There's 16 packages in $PACKAGES that do not exist in the repositories (f11 or rawhide), so here's a set of patches that removes them from PACKAGES in upd-instroot. They're separate patches, so that you can choose and pick which packages to remove from the list and which packages not to. Removing these would be beneficial since YUM takes a long time searching for a package that does not exist, by going over all names (obviously), as well as Provides (and Obsoletes?). These patches apply to master. Kind regards, Jeroen van Meeuwen -kanarip

3 3

[PATCH] Update linuxrc.s390 and friends to reflect review comments.
by Steffen Maier 27 Jul '09

27 Jul '09

This brings linuxrc.s390 and friends to the latest version as posted here https://www.redhat.com/archives/anaconda-devel-list/2009-February/msg00392.… plus some updates after that. Changes: - update usage of udev to rawhide - no more telnetd for network login during install - adapt readcmsfile to automatic loading of device driver modules - refactor main part into function init_main - consider renaming of network device driver module from ctc to ctcm - refactor control unit lookup tables into separate file - no more rely on kernel versions to detect device driver features - upstream qeth driver supports IPv6 on HiperSockets in layer3 mode - no more manual (un)loading of device driver modules - correctly handle default MAC address for layer2 qeth devices - no more kill udevd at end of linuxrc.s390 - write network device configuration to correct new place (ifcfg) Splitting up into single patches is impossible without merge conflicts. --- loader/controlunits.sh | 56 +++++ loader/linuxrc.s390 | 575 +++++++++++++++++++----------------------------- 2 files changed, 288 insertions(+), 343 deletions(-) create mode 100644 loader/controlunits.sh diff --git a/loader/controlunits.sh b/loader/controlunits.sh new file mode 100644 index 0000000..e99db36 --- /dev/null +++ b/loader/controlunits.sh @@ -0,0 +1,56 @@ +# controlunits.sh: define some common control unit mappings +# Copyright (C) IBM Corp. 2009 +# Author: Steffen Maier <maier(a)de.ibm.com> +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; version 2 of the License only. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write to the Free Software +# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + +# The arrays (among other things) should be adapted, if any of those device +# drivers start supporting different CU types/models. + +readonly -a CU=( + 1731/01 + 1731/05 + 3088/08 + 3088/1f + 3088/1e + 3088/01 + 3088/60 + 3088/61 +) + +readonly -a CU_DEVDRV=( + qeth + qeth + ctcm + ctcm + ctcm + lcs + lcs + lcs +) + +# Searches for a match of argument 1 on the array $CU and sets $cu_idx +# to the matched array index on success. +# Returns 0 on success, 1 on failure. +function search_cu() { + local scu=$1 + local i + for ((i=0; i < ${#CU[@]}; i++)); do + if [ "$scu" == "${CU[i]}" ]; then + cu_idx=$i + return 0 + fi + done + return 1 +} diff --git a/loader/linuxrc.s390 b/loader/linuxrc.s390 index 5dec01d..260e393 100644 --- a/loader/linuxrc.s390 +++ b/loader/linuxrc.s390 @@ -27,10 +27,18 @@ # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA # -# prerequisites of this script to run inside the RHEL5.3 installer initrd: -# - copy udevsettle and udevd to initrd -# - copy lsznet.raw to initrd as /sbin/lsznet -# - create /etc/udev/udev.conf with at least one comment line as content +# prerequisites of this script to run inside the installer initrd: +# - udevadm and udevd need to be there +# - have /etc/udev/udev.conf with at least one comment line as content +# - if necessary, have udev rules +# - copy lsznet.raw and controlunits.sh to initrd +# - pack kernel modules and module-init-tools (no longer use busybox for that) +# - "multi on" in /etc/host.conf [RH bugs 486457,486461,483244] + +# TODOs: +# - make sure driver modules get loaded automatically +# - udev rule for lcs/ctcm vs. cu3088 +# - make sure netiucv gets loaded even without udev events since on no bus VERSION=1.2 @@ -131,14 +139,13 @@ function sysecho () { function startinetd() { echo - echo $"Starting telnetd and sshd to allow login over the network." + echo $"Starting sshd to allow login over the network." if [ -z "$testing" ]; then echo $"Welcome to the anaconda install environment $VERSION for $S390ARCH" > /etc/issue.net echo $"Welcome to the anaconda install environment $VERSION for $S390ARCH" > /etc/motd echo >> /etc/motd fi # testing - /sbin/xinetd -stayalive -reuse -pidfile /tmp/xinetd.pid /sbin/sshd if [ -z "$RUNKS" ]; then echo @@ -157,33 +164,42 @@ function readcmsfile() # $1=dasdport $2=filename { local dev if [ $# -ne 2 ]; then return; fi - mknod /dev/dasda b 94 0 + # precondition: udevd created dasda block device node if ! sysecho /proc/cio_ignore "free $1"; then echo $"DASD $1 could not be cleared from device blacklist" return 1 fi # /proc/cio_ignore won't block on freeing devices until resensing # has been completed, so wait until the udev event queue depletes - # (without udevsettle we could wait 2 seconds unconditionally) + # (without udevadm settle we could wait 2 seconds unconditionally) #debug ls -laF /dev/.udev - udevsettle + udevadm settle + # even though the device might now be online, some of its + # sysfs attributes might not yet be available sleep 1 - insmod dasd_mod$LO dasd=$1 - insmod dasd_eckd_mod$LO - if ! cmsfscat -d /dev/dasda -a $2 > /tmp/$2; then - echo $"Could not read conf file $2 on CMS DASD $1." - fi - # maybe replace with a more robust: dev=$(printf "0.0.%04x" $((0x${1}))) + # precondition: dasd_eckd_mod driver incl. dependencies loaded, + # dasd_mod must be loaded without setting any DASD online if [ ${#1} == 3 ]; then dev="0.0.0${1}" elif [ ${#1} == 4 ]; then dev="0.0.${1}" fi - sysecho /sys/bus/ccw/drivers/dasd-eckd/$dev/online 0 - rmmod dasd_eckd_mod - rmmod dasd_mod + if ! sysecho /sys/bus/ccw/devices/$dev/online 1; then + echo $"DASD $dev could not be set online" + return 1 + fi + if ! cmsfscat -d /dev/dasda -a $2 > /tmp/$2; then + echo $"Could not read conf file $2 on CMS DASD $1." + fi + if ! sysecho /sys/bus/ccw/devices/$dev/online 0; then + echo $"DASD $dev could not be set offline again" + return 1 + fi + # consequences of no more module unload: loader/anaconda can no longer + # use DASD option to online DASDs and set other DASD parameters! } +# adaption of the same function in init.c (udevd gets started later) function createDevices() { awk '{ printf("mknod /dev/%s %s %s %s\n", $1, $2, $3, $4); @@ -202,6 +218,7 @@ random c 1 8 644 root:root urandom c 1 9 644 root:root rtc c 10 135 644 root:root EOF + # tty handling is different from init.c since s390 does not have all for i in 2 3 4 5 6 7 8 9 ; do ln -s console /dev/tty$i done @@ -209,6 +226,70 @@ EOF ln -s /proc/self/fd /dev/fd } +# approximately the main() function of init.c +function init_main() { + S390ARCH=$(uname -m) + if [ "$S390ARCH" = "s390" ]; then + export S390ARCH="S/390" + else + export S390ARCH="zSeries" + fi + + echo + echo $"Starting the $S390ARCH initrd to configure networking. Version is $VERSION" + + # set up env vars as we do in init.c + if [ $(uname -m) = "s390x" ]; then + LD_LIBRARY_PATH=/lib64:/usr/lib64:/usr/X11R6/lib64:/usr/kerberos/lib64:/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib + else + LD_LIBRARY_PATH=/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib + fi + export LD_LIBRARY_PATH + + PATH="$PATH:/usr/bin:/bin:/sbin:/usr/sbin:/mnt/sysimage/bin:/mnt/sysimage/usr/bin:/mnt/sysimage/usr/sbin:/mnt/sysimage/sbin:/mnt/sysimage/usr/X11R6/bin" + export PATH + HOME=/ + export HOME + PYTHONPATH=/tmp/updates + export PYTHONPATH + + if [ -z "$testing" ]; then + + mount -t proc none /proc + + mount -t tmpfs none /dev + createDevices + # udevd req'd by udevadm settle (/dev/.udev/queue) + # in readcmsfile, dialog_network_table, semantic_check_subchannels. + # (important: start udevd at the right time, e.g. after setup of /dev) + echo $"starting udev..." + udevd --daemon + + mount -t devpts /dev/pts /dev/pts + mount -t sysfs none /sys + # remount root fs rw + mount /dev/root / -o remount,rw + + # limit output on 3270 console + # (console_loglevel of 4 is just right to not get driver info, + # e.g. from qeth, since that would mix up with the user dialog) + echo "4 4 1 7" > /proc/sys/kernel/printk + + # make /tmp/ramfs + mount -t ramfs none /tmp + + ifconfig lo 127.0.0.1 netmask 255.0.0.0 + route add -host 127.0.0.1 dev lo + + echo -e "127.0.0.1\tlocalhost.localdomain localhost localhost4 localhost4.localdomain4" > /etc/hosts + echo -e "::1\t\tlocalhost.localdomain localhost localhost6 localhost6.localdomain6" >> /etc/hosts + + # init.c would also fork /sbin/dbus-uuidgen --ensure + # and /sbin/dbus-daemon --system + + fi # testing +} + # from here on accesses to sysfs try to follow # linux/Documentation/sysfs-rules.txt @@ -346,14 +427,12 @@ function dialog_network_table() { #( while true; do echo -n "."; sleep 3; done ) & #local childpid=$! #debug ls -laF /dev/.udev - udevsettle + udevadm settle # (virtual) CTC/A takes some more time to appear in sysfs # FIXME: how long to wait? 3 seconds seems to be enough. sleep 3 #kill $childpid #echo - # udevsettle only works if udevd is running (/dev/.udev/queue)! - # (without udevsettle we could wait 5-10 seconds uncond.) continue else echo $"Device blacklist could not be cleared" @@ -366,7 +445,11 @@ function dialog_network_table() { local item cutype chp chpidtype devdrv devname chlist cardtype read item cutype chp chpidtype devdrv devname chlist cardtype <<< ${nettable[$choice]} # $NETTYPE happens to be exactly the network driver name - NETTYPE=$devdrv + if [ "$devdrv" = "ctcm" ]; then + NETTYPE="ctc" + else + NETTYPE=$devdrv + fi SUBCHANNELS=$chlist break done @@ -379,49 +462,7 @@ declare -r BUSIDFORMAT=[[:xdigit:]][[:xdigit:]][[:xdigit:]][[:xdigit:]] declare -r IDFORMAT=$PREFIXFORMAT.$SSIDFORMAT.$BUSIDFORMAT declare -r SUBCHANNEL_TYPE_IO=0 -# The following combinations of control unit type and model were taken from the -# Linux network device drivers for s390 on 2008-06-09 from Linux 2.6.25.4. -# The list (among other things) should be adapted, if any of those device -# drivers start supporting different CU types/models. -# (Alternatively, the list could be generated by reading the modaliases -# directly out of the device driver modules: -# modinfo qeth/cu3088 | fgrep 'alias:' -# However, this would still require a list of device driver modules.) - -declare -r -a CU=( - 1731/01 - 1731/05 - 3088/08 - 3088/1f - 3088/1e - 3088/01 - 3088/60 -) - -declare -r -a CU_DEVDRV=( - qeth - qeth - ctc - ctc - ctc - lcs - lcs -) - -# Searches for a match of argument 1 on the array $CU and sets $cu_idx -# to the matched array index on success. -# Returns 0 on success, 1 on failure. -function search_cu() { - local scu=$1 - local i - for ((i=0; i < ${#CU[@]}; i++)); do - if [ "$scu" == "${CU[i]}" ]; then - cu_idx=$i - return 0 - fi - done - return 1 -} +. /sbin/controlunits.sh function cardtype2cleartext() { local cardtype=$1 @@ -436,9 +477,9 @@ function cardtype2cleartext() { OSD_Express) echo "OSA card in OSD mode, unknown link type" ;; HSTR) echo "OSA card in OSD mode, High Speed Token Ring" ;; OSN) echo "OSA for NCP, ESCON/CDLC bridge" ;; - HiperSockets) echo "HiperSocket with CHPID type IQD" ;; + HiperSockets) echo "HiperSockets with CHPID type IQD" ;; "GuestLAN QDIO") echo "GuestLAN based on OSA (QDIO)" ;; - "GuestLAN Hiper") echo "GuestLAN based on HiperSocket" ;; + "GuestLAN Hiper") echo "GuestLAN based on HiperSockets" ;; unknown) echo "other" ;; *) echo "unknown" echo "l.$LINENO: found unknown card_type, code needs to be fixed" 1>&2 @@ -456,29 +497,6 @@ function isVM() { fi } -# echos one comparable integer built from 3 arguments in kernel version style -function kernel_version() { - [ $# -eq 3 ] || return 1 - local krn_ver=$1 krn_patch=$2 krn_sub=$3 - local version=$(( (krn_ver << 16) + (krn_patch << 8) + krn_sub )) - # <<EOF convince syntax highlighter that above shifts are no here documents - echo $version - return 0 -} - -# echos one comparable integer built from actual running kernel version -function linux_version() { - local krn_ver krn_patch krn_sub - local krn_foo krn_extra - IFS=. - read krn_ver krn_patch krn_foo < <(uname -r) - unset IFS - IFS=- - read krn_sub krn_extra <<< "$krn_foo" - unset IFS - kernel_version $krn_ver $krn_patch $krn_sub -} - # watch out: potential error message as side effect function isLayer2Default() { # Read default from sysfs because according to device @@ -514,22 +532,14 @@ function isLayer2() { esac } -# returns true iff qeth device $SCH_R_DEVBUSID and kernel version +# returns true iff qeth device $SCH_R_DEVBUSID # is capable of supporting IPv6 # watch out: potential error message as side effect function ipv6_capable() { [ "$NETTYPE" = "qeth" ] || return 1 case $cardtype in - OSD_10GIG|OSD_1000|OSD_100|OSD_Express|"GuestLAN QDIO") return 0 ;; - HiperSockets) - if isLayer2; then - return 0 - elif [ $(linux_version) -ge $(kernel_version 2 6 26) ]; then - return 0 - else - return 1 - fi - ;; + OSD_10GIG|OSD_1000|OSD_100|OSD_Express|HiperSockets|"GuestLAN QDIO") + return 0 ;; OSD_GbE_LANE|OSD_FE_LANE|OSD_TR_LANE|OSD_ATM_LANE) return 1 ;; HSTR|OSN|unknown) return 1 ;; "GuestLAN Hiper") return 1 ;; @@ -538,75 +548,6 @@ function ipv6_capable() { esac } -function insert_module() { - local module=${1} - if grep "^${module%.ko} .*$" /proc/modules &> /dev/null ; then - # module has already been loaded => do not try to load again - return 0 - fi - if ! insmod $module; then - echo $"Could not load kernel module $module" - fi -} - -function load_kernel_modules() { - if [ ":$NETTYPE" = ":ctc" ]; then - insert_module ccwgroup$LO - insert_module cu3088$LO - insert_module fsm$LO - insert_module ctc$LO - elif [ ":$NETTYPE" = ":iucv" ]; then - insert_module fsm$LO - insert_module iucv$LO - insert_module netiucv$LO - elif [ "$NETTYPE" = "lcs" ]; then - insert_module ccwgroup$LO - insert_module cu3088$LO - insert_module lcs$LO - elif [ "$NETTYPE" = "qeth" ]; then - insert_module crypto_api$LO - insert_module xfrm_nalgo$LO - insert_module ipv6$LO - insert_module ccwgroup$LO - insert_module qdio$LO - insert_module qeth$LO - fi -} - -function remove_module() { - if ! rmmod $1; then - #echo $"Could not remove kernel module $1" - : - fi -} - -function remove_kernel_modules() { - if [ ":$NETTYPE" = ":ctc" ]; then - remove_module ctc - remove_module fsm - remove_module cu3088 - remove_module ccwgroup - elif [ ":$NETTYPE" = ":iucv" ]; then - remove_module netiucv - remove_module iucv - remove_module fsm - elif [ "$NETTYPE" = "lcs" ]; then - remove_module lcs - remove_module cu3088 - remove_module ccwgroup - elif [ "$NETTYPE" = "qeth" ]; then - remove_module qeth - remove_module qdio - remove_module ccwgroup - # The remaining 3 modules can hardly be unloaded - # since e.g. loopback holds an inet6 address and thus a reference. - # But that's no problem as long as the device drivers are removed. - remove_module ipv6 - remove_module xfrm_nalgo - remove_module crypto_api - fi -} - # sets device online _and_ retrieves DEVICE at the same time # iucv cannot be set online since it's not based on ccw(group) function set_device_online() { @@ -638,7 +579,7 @@ function set_device_online() { done if [ "$NETTYPE" = "lcs" -o "$NETTYPE" = "ctc" ]; then # KH FIXME: Workaround for missing sysfs interface - # DEVICE=`cat /sys/devices/lcs/${SUBCHANNELS//,*/}/if_name` + # DEVICE=$(cat /sys/devices/lcs/${SUBCHANNELS//,*/}/if_name) # replaced with flexible solution: # https://bugzilla.redhat.com/show_bug.cgi?id=204803#c9 # "sys/bus/ccwgroup/devices/${SUBCHANNEL}/net\:* @@ -684,7 +625,7 @@ function set_device_up() { fi # Device does not come up fast enough to use "ip" to configure, so block. # While OSA come up themselves after setting online, - # e.g. HiperSocket won't, so set them up explicitly for the following check + # e.g. HiperSockets won't => set them up explicitly for the following check debug ip link set up $DEVICE local i=1 while : ; do @@ -727,6 +668,8 @@ function syntax_check_domainname() { function modprobe_alias() { if [ ":$NETTYPE" = ":iucv" ]; then echo "alias $DEVICE netiucv" >> /tmp/modprobe.conf + elif [ ":$NETTYPE" = ":ctc" ]; then + echo "alias $DEVICE ctcm" >> /tmp/modprobe.conf else echo "alias $DEVICE $NETTYPE" >> /tmp/modprobe.conf fi @@ -829,9 +772,6 @@ function rollback_config() { [ -z "$mtu_was_set" ] && unset MTU [ -z "$mmtu_was_set" ] && unset MMTU [ -z "$vswitch_was_set" ] && unset VSWITCH - # Try to silently unload modules. We ignore failures since - # load_kernel_modules only tries to load missing modules. - tv remove_kernel_modules # prevent possible reuse of an old DEVICE on restarting dialog unset DEVICE } @@ -960,7 +900,7 @@ function ask() { c) handle=$OPTARG ;; f) finish=$OPTARG ;; "?") ;; # ignore invalid option - :) echo $"l.$LINENO: Missing parameter to option -$OPTARG" 1>&2 ;; + :) echo "l.$LINENO: Missing parameter to option -$OPTARG" 1>&2 ;; esac done while : ; do @@ -1018,7 +958,7 @@ function ask() { # few parameters need special handling done by their own function: $finish else - echo "Unsupported calling of ask function, please fix calling code" + echo $"Unsupported calling of ask function, please fix calling code" fi done # PARMNAME # disable potential temporary redoing-mode during reenter-mode @@ -1068,7 +1008,7 @@ function question_choices_nettype() { function helptext_nettype() { echo $" Help text for network type:" echo $" qeth: OSA-Express Fast Ethernet, Gigabit Ethernet (including 1000Base-T)," - echo $" High Speed Token Ring, Hipersockets, and ATM (running Ethernet LAN emulation)" + echo $" High Speed Token Ring, and ATM (running Ethernet LAN emulation)" echo $" features in QDIO mode." echo $" [default]" echo $" lcs: OSA-2 Ethernet/Token Ring, OSA-Express Fast Ethernet in non-QDIO mode," @@ -1114,7 +1054,7 @@ function do_chandev() { ### SUBCHANNELS function syntax_check_subchannels() { - SUBCHANNELS=`echo $SUBCHANNELS | tr ABCDEF abcdef` + SUBCHANNELS=$(echo $SUBCHANNELS | tr ABCDEF abcdef) # - make subchannel question dependent on NETTYPE (2 vs. 3 subchannels) if [ "$NETTYPE" = "qeth" ]; then # - match against regex, depending on qeth @@ -1173,10 +1113,10 @@ function semantic_check_subchannels() { if sysecho /proc/cio_ignore "free $devbusid"; then # /proc/cio_ignore won't block on freeing devices # until resensing has been completed, so wait until - # the udev event queue depletes (without udevsettle we + # the udev event queue depletes (without udevadm settle we # could wait 2 seconds unconditionally) #debug ls -laF /dev/.udev - udevsettle + udevadm settle # even though the device might now be online, some of its # sysfs attributes (e.g. cutype) might not yet be available sleep 1 @@ -1214,7 +1154,13 @@ function semantic_check_subchannels() { local cutype read cutype < $dev_p/cutype if search_cu $cutype; then - if [ "${CU_DEVDRV[$cu_idx]}" != "$NETTYPE" ]; then + local driver + if [ "$NETTYPE" = "ctc" ]; then + driver="ctcm" + else + driver=$NETTYPE + fi + if [ "${CU_DEVDRV[$cu_idx]}" != "$driver" ]; then echo $"Device $devbusid has control unit type $cutype," echo $" which does not match your selected network type $NETTYPE" all_subch_good=1 @@ -1291,6 +1237,7 @@ function semantic_check_subchannels() { if [ "$i" -eq 2 ]; then local sch_w_devbusid=$devbusid local sch_w_devno=$devno + # TODO: not true for CTCM => relax # - write_devbusid == read_devbusid+1 if [ $((0x$devno)) -ne $((0x$sch_r_devno + 1)) ]; then echo $"Device bus ID of write channel (dev $devbusid) must be one larger than" @@ -1353,11 +1300,17 @@ function semantic_check_subchannels() { function handle_subchannels() { # - try to establish ccwgroup right here and fail out on error - if sysecho /sys/bus/ccwgroup/drivers/${NETTYPE}/group "$SUBCHANNELS"; then + local driver + if [ "$NETTYPE" = "ctc" ]; then + driver="ctcm" + else + driver=$NETTYPE + fi + if sysecho /sys/bus/ccwgroup/drivers/${driver}/group "$SUBCHANNELS"; then case "$NETTYPE" in qeth) # Just preliminary card_type info until device goes online! - # In fact it seems enough to separate OSA from HiperSocket. + # In fact it seems enough to separate OSA from HiperSockets. if [ -f /sys/devices/qeth/$SCH_R_DEVBUSID/card_type ]; then read cardtype < /sys/devices/qeth/$SCH_R_DEVBUSID/card_type else @@ -1366,9 +1319,11 @@ function handle_subchannels() { ;; ctc|lcs) if [ -f /sys/devices/cu3088/$SCH_R_DEVBUSID/type ]; then - local type=$(cat /sys/devices/cu3088/$SCH_R_DEVBUSID/type) - [ "$type" = "CTC/A" ] && type="channel-to-channel adapter" - echo "Detected: $type" + local type + read type < /sys/devices/cu3088/$SCH_R_DEVBUSID/type + [ "$type" = "CTC/A" ] && \ + type="channel-to-channel adapter (CTC/A)" + echo $"Detected: $type" else echo $"Could not read ctc network card type from sysfs." fi @@ -1436,7 +1391,7 @@ function do_subchannels() { function syntax_check_portname() { # - 1-8 characters, we convert it to upper case - PORTNAME=`echo $PORTNAME | tr '[:lower:]' '[:upper:]'` + PORTNAME=$(echo $PORTNAME | tr '[:lower:]' '[:upper:]') local portname_len=${#PORTNAME} if [ "$portname_len" -ge 1 -a "$portname_len" -le 8 ]; then return 0 @@ -1542,7 +1497,7 @@ function helptext_portno() { } function exception_portno() { - # Writing portno of e.g. hipersocket device fails. + # Writing portno of e.g. hipersockets device fails. # Therefore, do not configure on empty default value. [ -z "$PORTNO" ] && break } @@ -1604,7 +1559,11 @@ function helptext_layer2() { } function exception_layer2() { - [ -z "$LAYER2" ] && break + if [ -z "$LAYER2" ]; then + isLayer2Default && LAYER2=1 || LAYER2=0 + # do not break, always apply, default may differ from online layer mode + #break + fi } function do_layer2() { @@ -1655,37 +1614,36 @@ function question_prefix_macaddr() { } function question_choices_macaddr() { - echo -n $" (e.g. 02:00:00:00:00:00, ? for help)." - # for virtual OSAs macaddr is optional and defaults to not specifying one - if [ -z "${cardtype//OSD_*/}" ]; then - echo - else - echo $" Default is automatic:" - fi + macaddr_default=$(ifconfig $DEVICE | grep 'HWaddr' | sed 's/.*HWaddr $[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]:[[:xdigit:]][[:xdigit:]]$.*/\1/') + echo $" (e.g. 02:00:00:00:00:00, ? for help). Default is $macaddr_default:" } function helptext_macaddr() { echo $" Help text for MAC address:" - echo $" This is required for real OSA in layer 2 mode." - echo $" If connecting to a layer 2 VSWITCH, a MAC address is automatically assigned." - echo -n $" You may accept the automatic MAC address with an empty input." if [ -z "${cardtype//OSD_*/}" ]; then - echo + echo $" For real OSA in layer 2 mode, a random MAC address is automatically assigned." else - echo $" [default]" + echo $" If connecting to a layer 2 VSWITCH, a MAC address is automatically assigned." fi + echo $" You may accept the automatic MAC address with an empty input. [default]" + echo $" If the automatic address is not unique, please provide a MAC address." + [ -z "${cardtype//OSD_*/}" ] && \ + echo $" For real OSA, the provided address must be different from that of the OSA." echo $" You may override the automatic MAC address with non-empty input." echo $" An example MAC address would be: 02:00:00:00:00:00" } function exception_macaddr() { - # for real OSAs macaddr is required - if [ -n "${cardtype//OSD_*/}" ]; then - # for virtual OSAs macaddr is optional - if [ -z "$MACADDR" ]; then + if [ -z "$MACADDR" ]; then + if [ -z "${cardtype//OSD_*/}" ]; then + # keep random default MAC address of real OSA, + # so the OSA comes up with the same MAC each time in the future + MACADDR=$macaddr_default + else + # virtual OSA in layer2 is GuestLAN or VSWITCH VSWITCH=1 - break fi + break fi } @@ -1813,7 +1771,7 @@ function syntax_check_peerid() { echo $"Incorrect string length [0..8] for IUCV connection to peer (PEERID): $PEERID" return 1 fi - PEERID=`echo $PEERID | tr '[:lower:]' '[:upper:]'` + PEERID=$(echo $PEERID | tr '[:lower:]' '[:upper:]') [[ "$PEERID" =~ ^[[:alnum:]$]{0,8}$ ]] case $? in 0) @@ -1894,7 +1852,7 @@ function do_peerid() { ### HOSTNAME function syntax_check_hostname() { - syntax_check_domainname "$HOSTNAME" $"Incorrect format for hostname (HOSTNAME): $HOSTNAME" + syntax_check_domainname "$HOSTNAME" "Incorrect format for hostname (HOSTNAME): $HOSTNAME" } function handle_hostname() { @@ -1914,8 +1872,8 @@ function question_choices_hostname() { } function helptext_hostname() { - echo $" Help text for hostname:" - echo $" Enter the full qualified domain name of your host." + echo $" Help text for hostname:" + echo $" Enter the full qualified domain name of your host." } function do_hostname() { @@ -1958,11 +1916,9 @@ function helptext_ipaddr() { echo $" or alternatively a valid IPv6 address without CIDR prefix (e.g. 2001:0DB8::)" echo $" IPv6 is supported on:" echo $" - Ethernet interfaces of the OSA-Express adapter running in QDIO mode." - echo $" - HiperSockets layer 2 interfaces" - [ $(linux_version) -ge $(kernel_version 2 6 26) ] && \ - echo $" - HiperSockets layer 3 interfaces" + echo $" - HiperSockets interfaces" echo $" - z/VM guest LAN interfaces running in QDIO mode." - echo $" IPv6 is not supported on HiperSocket guest LAN, OSA-Express Token Ring, ATM." + echo $" IPv6 is not supported on HiperSockets guest LAN, OSA-Express Token Ring, ATM." fi } @@ -2007,13 +1963,13 @@ function hint_netmask_v4() { # <<EOF convince syntax highlighter that above shifts are no here documents if [ $(( ip & 0x80000000 )) -eq $(( 0x00000000 )) ]; then # class a - echo $"255.0.0.0" + echo "255.0.0.0" elif [ $(( ip & 0xC0000000 )) -eq $(( 0x80000000 )) ]; then # class b - echo $"255.255.0.0" + echo "255.255.0.0" elif [ $(( ip & 0xE0000000 )) -eq $(( 0xC0000000 )) ]; then # class c - echo $"255.255.255.0" + echo "255.255.255.0" else # some other class that should not be used as host address return 1 @@ -2523,7 +2479,9 @@ function parse_dasd() { esac done < <(echo $DASD | sed 's/,/\n/g') if [ "$handle" = "yes" ]; then - udevsettle + udevadm settle + # even though the device might now be online, some of its + # sysfs attributes might not yet be available sleep 1 fi if [ "$allgood" = "yes" ]; then @@ -2727,101 +2685,43 @@ function final_check() { ### MAIN ### -S390ARCH=`uname -m` -if [ "$S390ARCH" = "s390" ]; then - export S390ARCH="S/390" -else - export S390ARCH="zSeries" -fi +init_main -echo -echo $"Starting the $S390ARCH initrd to configure networking. Version is $VERSION" - -# set up env vars as we do in init.c -if [ `uname -m` = "s390x" ]; then - LD_LIBRARY_PATH=/lib64:/usr/lib64:/usr/X11R6/lib64:/usr/kerberos/lib64:/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib -else - LD_LIBRARY_PATH=/lib:/usr/lib:/usr/X11R6/lib:/usr/kerberos/lib +# Parse configuration +if [ -n "$CMSDASD" -a -n "$CMSCONFFILE" ]; then + readcmsfile $CMSDASD $CMSCONFFILE + source /tmp/$CMSCONFFILE #2>/dev/null fi -export LD_LIBRARY_PATH - -PATH="$PATH:/usr/bin:/bin:/sbin:/usr/sbin:/mnt/sysimage/bin:/mnt/sysimage/usr/bin:/mnt/sysimage/usr/sbin:/mnt/sysimage/sbin:/mnt/sysimage/usr/X11R6/bin" -export PATH -HOME=/ -export HOME -PYTHONPATH=/tmp/updates -export PYTHONPATH - -if [ -z "$testing" ]; then - - mount -t proc none /proc - mount -t tmpfs none /dev - createDevices - mount -t devpts /dev/pts /dev/pts - mount -t sysfs none /sys - # remount root fs rw - mount /dev/root / -o remount,rw - - # limit output on x3270 console - # (console_loglevel of 4 is just right to not get driver info, e.g. from qeth, - # since that would mix up with the user dialog) - echo "4 4 1 7" > /proc/sys/kernel/printk - - # make /tmp/ramfs - mount -t ramfs none /tmp - - ifconfig lo 127.0.0.1 netmask 255.0.0.0 - route add -host 127.0.0.1 dev lo - - echo -e "127.0.0.1\tlocalhost.localdomain localhost" > /etc/hosts - echo -e "::1\t\tlocalhost6.localdomain6 localhost6" >> /etc/hosts - - LO="" - - [ -L /sbin/insmod ] && LO=".ko" - - # udevd req'd by udevsettle in readcmsfile, dialog_network_table, and - # semantic_check_subchannels. - # (important to start udevd at the right time, e.g. after setup of /dev) - udevd --daemon - - # Parse configuration - if [ -n "$CMSDASD" -a -n "$CMSCONFFILE" ]; then - readcmsfile $CMSDASD $CMSCONFFILE - source /tmp/$CMSCONFFILE #2>/dev/null - fi - -fi # testing - -### MAIN (continuation) ### if [ -r /sys/firmware/ipl/ipl_type ]; then - if [ "`cat /sys/firmware/ipl/ipl_type`" = "fcp" ]; then - while [ 1 ]; do + #local ipl_type + read ipl_type < /sys/firmware/ipl/ipl_type + if [ "$ipl_type" = "fcp" ]; then + while : ; do echo $"Your IPL device is set to FCP." echo $"Would you like to perform a CD-ROM/DVD-ROM installation? (y/n)" + #local do_cd_install read do_cd_install case $do_cd_install in y|Y|[Yy][Ee][Ss]) - insert_module qdio$LO - insert_module scsi_mod$LO - insert_module scsi_transport_fc$LO - insert_module zfcp$LO - CD_DEVICE="`cat /sys/firmware/ipl/device`" - WWPN="`cat /sys/firmware/ipl/wwpn`" - LUN="`cat /sys/firmware/ipl/lun`" + # precondition: zfcp driver incl. dependencies loaded + #local CD_DEVICE WWPN LUN + read CD_DEVICE < /sys/firmware/ipl/device + read WWPN < /sys/firmware/ipl/wwpn + read LUN < /sys/firmware/ipl/lun if sysecho /proc/cio_ignore "free $CD_DEVICE"; then - udevsettle + udevadm settle + # even though device might now be online, some of its + # sysfs attributes might not yet be available sleep 1 else echo $"Device $CD_DEVICE could not be cleared from device blacklist" fi - echo 1 > /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/online - [ $? == 0 ] || echo $"Could not set FCP device $CD_DEVICE online" - echo $WWPN > /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/port_add - [ $? == 0 ] || echo $"Could not add WWPN $WWPN on FCP device $CD_DEVICE" - echo $LUN > /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/$WWPN/unit_add - [ $? == 0 ] || echo $"Could not add LUN $LUN at WWPN $WWPN on FCP device $CD_DEVICE" + sysecho /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/online 1 \ + || echo $"Could not set FCP device $CD_DEVICE online" + # port (WWPN) appears automatically + sysecho /sys/bus/ccw/drivers/zfcp/$CD_DEVICE/$WWPN/unit_add $LUN \ + || echo $"Could not add LUN $LUN at WWPN $WWPN on FCP device $CD_DEVICE" break ;; n|N|[Nn][Oo]) @@ -2846,7 +2746,6 @@ fi # [ -z "${cardtype//OSD_*/}" ] can be used to check for real OSA -# Parse configuration # Check for missing parameters, prompt for them if necessary while : ; do @@ -2862,7 +2761,7 @@ while : ; do echo $"* NOTE: To enter default or empty values press enter twice. *" fi do_nettype - tv load_kernel_modules + # precondition: driver (qeth/lcs/ctcm/netiucv) loaded incl. dependencies # only works, if ipv6 kernel module has been loaded before: [ "$NETTYPE" = "qeth" ] && tv disable_ipv6_autoconf if [ "$NETTYPE" != "iucv" ]; then @@ -2886,7 +2785,7 @@ while : ; do fi do_layer2 # set device online to know the device name - # and to know if it's OSD/HiperSocket/GuestLAN BUT do not + # and to know if it's OSD/HiperSockets/GuestLAN BUT do not # try to ifconfig the device up since that requires # setting the mac address before (if applicable). set_device_online || workflow_item_menu noredo @@ -2970,9 +2869,6 @@ while : ; do done # outer dialog loop -#echo $"Starting portmap." -#portmap - if [ -z "$testing" ]; then # convert to space-separated lists @@ -2989,18 +2885,13 @@ if [ -z "$testing" ]; then for i in $RESOLVDNS; do echo "nameserver $i"; done >> /etc/resolv.conf fi - # make sure we have an /etc/hosts file (required for telnetd) + # make sure we have an /etc/hosts file (originally required for telnetd) if [ ! -z "$HOSTNAME" -a ! -z "$IPADDR" ]; then - echo -e "$IPADDR\t$HOSTNAME `echo $HOSTNAME | cut -d '.' -f 1`" >> /etc/hosts + echo -e "$IPADDR\t$HOSTNAME $(echo $HOSTNAME | cut -d '.' -f 1)" >> /etc/hosts fi fi # testing -if [ -z "$testing" ]; then - # undo startup of udevd, only needed it for possible calls of udevsettle - killall udevd -fi # testing - if [ -n "$DASD" ]; then echo "DASD=$DASD" > /tmp/dasd_ports fi @@ -3015,16 +2906,6 @@ done # cio_ignore handling for FCP should happen when the content of /tmp/fcpconfig # will actually be processed which is in anaconda's zfcp.py ZFCP::readConfig() -if [ -z "$testing" ]; then - - grep -q ext3 /proc/filesystems - if [ "$?" != "0" ]; then - insmod jbd$LO - insmod ext3$LO - fi - -fi # testing - # TODO/FIXME: also need to pass IPv6 decision to loader/anaconda # [ "$ipv6" ] && echo "IPV6=yes" @@ -3051,9 +2932,9 @@ ONBOOT="yes" CTCPROT="$CTCPROT" EOF if [ "$ipv6" ]; then - echo DNS=\"`echo $DNS | cut -d ',' -f 1`\" >> /tmp/install.cfg + echo DNS=\"$(echo $DNS | cut -d ',' -f 1)\" >> /tmp/install.cfg else - echo DNS=\"`echo $DNS | cut -d ':' -f 1`\" >> /tmp/install.cfg + echo DNS=\"$(echo $DNS | cut -d ':' -f 1)\" >> /tmp/install.cfg fi cat >> /tmp/install.cfg << EOF export LANG PORTNAME S390ARCH TEXTDOMAIN TEXTDOMAINDIR @@ -3066,7 +2947,14 @@ EOF if [ -z "$testing" ]; then cat /tmp/install.cfg >> /etc/profile fi # testing -cat > /tmp/netinfo << EOF + +NETSCRIPTS="/etc/sysconfig/network-scripts" +IFCFGFILE="$NETSCRIPTS/ifcfg-$DEVICE" +if [ ! -d "$NETSCRIPTS" ]; then + mkdir -p $NETSCRIPTS +fi + +cat > $IFCFGFILE << EOF DEVICE=$DEVICE ONBOOT=yes BOOTPROTO=static @@ -3077,34 +2965,35 @@ MTU=$MTU SUBCHANNELS=$SUBCHANNELS EOF if [ "$ipv6" ]; then - [ "$DNS" != "" ] && echo "DNS=`echo $DNS | cut -d ',' -f 1`" >> /tmp/netinfo -else - [ "$DNS" != "" ] && echo "DNS=`echo $DNS | cut -d ':' -f 1`" >> /tmp/netinfo -fi -[ "$NETTYPE" != "" ] && echo "NETTYPE=$NETTYPE" >> /tmp/netinfo -[ "$PEERID" != "" ] && echo "PEERID=$PEERID" >> /tmp/netinfo -[ "$PORTNAME" != "" ] && echo "PORTNAME=$PORTNAME" >> /tmp/netinfo -[ "$CTCPROT" != "" ] && echo "CTCPROT=$CTCPROT" >> /tmp/netinfo -[ "$LAYER2" != "" ] && echo "LAYER2=$LAYER2" >> /tmp/netinfo -[ "$MACADDR" != "" ] && echo "MACADDR=$MACADDR" >> /tmp/netinfo -[ "$PORTNO" != "" ] && echo "PORTNO=$PORTNO" >> /tmp/netinfo -# also needs support in stage2 to make PORTNO persistent, -# i.e. OPTION="portno=1" in /etc/sysconfig/network-scripts/ifcfg-<netdevname> -# this is done by the loader starting with RHEL 5.3 -if [ "$ipv6" ]; then - cat >> /tmp/netinfo << EOF + cat >> $IFCFGFILE << EOF IPV6INIT=yes IPV6_AUTOCONF=no IPV6ADDR=$IPADDR/$NETMASK +IPV6_DEFAULTGW=$GATEWAY EOF # FIXME: /etc/sysconfig/network:IPV6_DEFAULTGW=$GATEWAY # /etc/sysconfig/network:NETWORKING_IPV6=yes else - cat >> /tmp/netinfo << EOF + cat >> $IFCFGFILE << EOF IPADDR=$IPADDR NETMASK=$NETMASK EOF fi +if [ "$ipv6" ]; then + [ "$DNS" != "" ] && echo "DNS=$(echo $DNS | cut -d ',' -f 1)" >> $IFCFGFILE +else + [ "$DNS" != "" ] && echo "DNS=$(echo $DNS | cut -d ':' -f 1)" >> $IFCFGFILE +fi +[ "$NETTYPE" != "" ] && echo "NETTYPE=$NETTYPE" >> $IFCFGFILE +[ "$PEERID" != "" ] && echo "PEERID=$PEERID" >> $IFCFGFILE +[ "$PORTNAME" != "" ] && echo "PORTNAME=$PORTNAME" >> $IFCFGFILE +[ "$CTCPROT" != "" ] && echo "CTCPROT=$CTCPROT" >> $IFCFGFILE +[ "$LAYER2" != "" ] && echo "LAYER2=$LAYER2" >> $IFCFGFILE +[ "$MACADDR" != "" ] && echo "MACADDR=$MACADDR" >> $IFCFGFILE +[ "$PORTNO" != "" ] && echo "PORTNO=$PORTNO" >> $IFCFGFILE +# also needs support in stage2 (loader.c,(net.c)) to make PORTNO persistent, +# i.e. OPTION="portno=1" in /etc/sysconfig/network-scripts/ifcfg-<netdevname> +# (see loader of RHEL 5.3) if [ -z "$testing" ]; then -- 1.6.3.1

2 1

Mutlipath support in anaconda
by Peter Jones 27 Jul '09

27 Jul '09

This patch adds multipath support. It still needs some work and some cleanup, and one chunk of it needs to be rewritten after some pyblock work gets done, but it's probably good enough for Feature Freeze. Let me know how much you hate it. Thanks. -- Peter

3 7

[PATCH] Do not raise UI dialog in stage2 if network is set in ks (#487503)
by Radek Vykydal 27 Jul '09

27 Jul '09

It concerns e.g. installation from media with ks on media. With the patch the dialog is raised only if bringing-up with ks-supplied data fails. --- gui.py | 3 +++ text.py | 3 +++ 2 files changed, 6 insertions(+), 0 deletions(-) diff --git a/gui.py b/gui.py index 3d75ae1..c4f697d 100755 --- a/gui.py +++ b/gui.py @@ -1161,6 +1161,9 @@ class InstallInterface: def enableNetwork(self, anaconda): if len(anaconda.id.network.netdevices) == 0: return False + if (anaconda.isKickstart and anaconda.id.ksdata.network + and anaconda.id.network.bringUp()): + return True from netconfig_dialog import NetworkConfigurator net = NetworkConfigurator(anaconda.id.network) ret = net.run() diff --git a/text.py b/text.py index fada4e2..388a97c 100644 --- a/text.py +++ b/text.py @@ -512,6 +512,9 @@ class InstallInterface: def enableNetwork(self, anaconda): if len(anaconda.id.network.netdevices) == 0: return False + if (anaconda.isKickstart and anaconda.id.ksdata.network + and anaconda.id.network.bringUp()): + return True from netconfig_text import NetworkConfiguratorText w = NetworkConfiguratorText(self.screen, anaconda) ret = w.run() -- 1.6.0.6

1 1

[PATCH mkinitrd rhel-5.4] Do not try to activate raid10 sub sets
by Hans de Goede 27 Jul '09

27 Jul '09

Hi, While reproducing some RHEL-5 dmraid 10 bugs I noticed that there is a really ugly error message when booting systems with / on dmraid 10, the problem is that we also try to activate the raid 0 subsets which make up the raid 1 set. The solution is to use a different command to get the set names, which also means we can get rid of the grep, patch attached. Peter, do you see any possible down sides to this patch ? Regards, Hans

1 0

[PATCH] master. The New button has to go.
by Joel Granados Moreno 27 Jul '09

27 Jul '09

Hello list The button that reads "New" on the customization screen does not describe what it does. Which is call the partition screen. I propose to call it "Partition" to correctly portray its functionality. This is also consistent with the LVM and RAID buttons. Review appreciated. regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

4 4

[PATCH] master.
by Joel Granados Moreno 24 Jul '09

24 Jul '09

Hello list When selecting the lv in the tree and hitting edit we go to the Volume group. The user has to then select the Logical Volume to continue editing. Wouldn't it be better to go to the LV editing window? This patch allows this but with some strangeness: 1. It goes to the LV screen directly but it "passes through" the Volume group screen. So the user will see the VolumeGroup screen behind the Lobical Volume screen pop up. 2. Once the user is done editing and hits OK, or Cancel, both windows close. IMO, its consistent behavior, but its still strange. In any case I think its a step in the right direction. comments greatly appreciated. regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

3 4

[PATCH] Disable devel repos on release (#503798)
by Jeremy Katz 23 Jul '09

23 Jul '09

We don't want to have rawhide listed as a repo once we disable betanag. This is a little bit hard-coded, but probably okay --- yuminstall.py | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index 035c4b1..3553927 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -533,6 +533,12 @@ class AnacondaYum(YumSorter): del(repo) raise RepoError, "Repo %s contains -source or -debuginfo, excluding" % name + # this is a little hard-coded, but it's effective + if not BETANAG and (repo.id.find("rawhide") or repo.id.find("development")): + name = repo.name + del(repo) + raise RepoError, "Excluding devel repo %s for non-devel anaconda" % name + if BETANAG and not repo.enabled: name = repo.name del(repo) -- 1.6.2.5

3 2

[PATCH] Work around problems with live installs and dpi other than 96 (#506512)
by Jeremy Katz 23 Jul '09

23 Jul '09

On the live install, we don't force the DPI to 96. Some of our text takes a bit of space, especially the autopart screen. So if we're on a live install (eg, without mini-wm running), let's just always get rid of the little header image. It doesn't really buy us much and then we get more real estate to work with. Longer term, we probably need to move away from the fixed window size stuff, even though it'll be painful --- gui.py | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/gui.py b/gui.py index 4def03f..3d75ae1 100755 --- a/gui.py +++ b/gui.py @@ -1513,6 +1513,12 @@ class InstallControlWindow: else: self.window.set_size_request(800, 600) self.window.set_position(gtk.WIN_POS_CENTER_ALWAYS) + # this is kind of poor, but if we're running in the live mode + # and the dpi is something weird, give ourselves as much + # space as we can. this gets things to fit with a dpi + # of up to 147 + if not runningMiniWm(): + i.hide() if flags.debug: self.mainxml.get_widget("debugButton").show_now() -- 1.6.2.5

3 4

[PATCH] Fix obvious typo in font name
by Jeremy Katz 23 Jul '09

23 Jul '09

--- lang-table | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/lang-table b/lang-table index 04e49bd..2378b8d 100644 --- a/lang-table +++ b/lang-table @@ -23,7 +23,7 @@ Hindi hi none hi_IN.UTF-8 us Asia/Kolkata Hungarian hu latarcyrheb-sun16 hu_HU.UTF-8 hu Europe/Budapest Icelandic is latarcyrheb-sun16 is_IS.UTF-8 is-latin1 Atlantic/Reykjavik Iloko ilo latarcyrheb-sun16 ilo_PH.UTF-8 us Asia/Manila -Indonesian id latarcryheb-sun16 id_ID.UTF-8 us Asia/Jakarta +Indonesian id latarcyrheb-sun16 id_ID.UTF-8 us Asia/Jakarta Italian it latarcyrheb-sun16 it_IT.UTF-8 it Europe/Rome Japanese ja none ja_JP.UTF-8 jp106 Asia/Tokyo Kannada kn none kn_IN.UTF-8 us Asia/Kolkata -- 1.6.2.5

2 1

[PATCH] Log non-upgradable upgrade candidate roots.
by Radek Vykydal 23 Jul '09

23 Jul '09

This can be helpful in determining the cause of failed upgrade from log. I used it when working on #499321. Can be kind of corner-case, but can't hurt. --- storage/__init__.py | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/storage/__init__.py b/storage/__init__.py index 2370f0c..348a155 100644 --- a/storage/__init__.py +++ b/storage/__init__.py @@ -1045,6 +1045,9 @@ def findExistingRootDevices(anaconda, upgradeany=False): if upgradeany or \ anaconda.id.instClass.productUpgradable(product, version): rootDevs.append((device, "%s %s" % (product, version))) + else: + log.info("product %s version %s found on %s is not upgradable" + % (product, version, device.name)) # this handles unmounting the filesystem device.teardown(recursive=True) -- 1.6.0.6

2 1

Building anaconda for F11 fix
by Nicholas Veeser 22 Jul '09

22 Jul '09

Sorry to bug you guys on your mailing list, but it seemed related to the development process. Trying to understand how all the pieces of anaconda work to that I can apply a fix for my own setup of F11. Seems there is a fix in Rawhide for the following RAID issue. https://bugzilla.redhat.com/show_bug.cgi?id=503310 I am having the issue in F11 I found this unofficial fix in the bug http://rvykydal.fedorapeople.org/updates.clearpart.img Put the updates.clearpart.img in <repo>/images/updates.img That fixes it ------ For completeness, I would like to be able to reproduce, from source, what Mr. Vykydal did to produce the updates.img - Pulled the source with git - Read the building / packaging instructions - I found info about moving to autoconf here: http://www.linux-archive.org/debian-development/314101-moving-anaconda-auto… http://dcantrel.fedorapeople.org/anaconda/autognu/READ_ME.txt - Can't seem to find the ./bootstrap file mentioned Help? Please don't make me understand the voodoo of autoconf (I have been avoiding it for 10 years..;)

2 1

[PATCH 1/2] Add a method to determine whether a block device is read-only.
by Chris Lumens 22 Jul '09

22 Jul '09

--- isys/isys.c | 24 ++++++++++++++++++++++++ isys/isys.py | 6 ++++++ 2 files changed, 30 insertions(+), 0 deletions(-) diff --git a/isys/isys.c b/isys/isys.c index fcec6fd..90a697c 100644 --- a/isys/isys.c +++ b/isys/isys.c @@ -124,6 +124,7 @@ static PyObject * doGetBiosDisk(PyObject * s, PyObject * args); static PyObject * doSegvHandler(PyObject *s, PyObject *args); static PyObject * doAuditDaemon(PyObject *s); static PyObject * doPrefixToNetmask(PyObject *s, PyObject *args); +static PyObject * doDeviceReadOnly(PyObject *s, PyObject *args); static PyMethodDef isysModuleMethods[] = { { "ejectcdrom", (PyCFunction) doEjectCdrom, METH_VARARGS, NULL }, @@ -180,6 +181,7 @@ static PyMethodDef isysModuleMethods[] = { { "handleSegv", (PyCFunction) doSegvHandler, METH_VARARGS, NULL }, { "auditdaemon", (PyCFunction) doAuditDaemon, METH_NOARGS, NULL }, { "prefix2netmask", (PyCFunction) doPrefixToNetmask, METH_VARARGS, NULL }, + { "deviceIsReadOnly", (PyCFunction) doDeviceReadOnly, METH_VARARGS, NULL }, { NULL, NULL, 0, NULL } } ; @@ -1416,4 +1418,26 @@ static PyObject * doAuditDaemon(PyObject *s) { return Py_None; } +static PyObject * doDeviceReadOnly(PyObject *s, PyObject *args) { + char *diskname = NULL; + int fd, is_ro; + + if (!PyArg_ParseTuple(args, "s", &diskname)) return NULL; + + fd = open(diskname, O_RDONLY); + if (fd == -1) { + Py_INCREF(Py_None); + return Py_None; + } + + if (ioctl(fd, BLKROGET, &is_ro)) { + close(fd); + PyErr_SetFromErrno(PyExc_SystemError); + return NULL; + } + + close(fd); + return PyBool_FromLong(is_ro); +} + /* vim:set shiftwidth=4 softtabstop=4: */ diff --git a/isys/isys.py b/isys/isys.py index 4063130..189a6c1 100644 --- a/isys/isys.py +++ b/isys/isys.py @@ -804,6 +804,12 @@ def driveUsesModule(device, modules): pass return rc +def deviceIsReadOnly(device): + if not device.startswith("/dev/"): + return _isys.deviceIsReadOnly("/dev/" + device) + else: + return _isys.deviceIsReadOnly(device) + def mediaPresent(device): try: fd = os.open("/dev/%s" % device, os.O_RDONLY) -- 1.6.1.3

2 3

[PATCH 1/2] Try to auto-find the CD even if stage2= is specified
by Jeremy Katz 21 Jul '09

21 Jul '09

Try to auto-find the CD even if stage2= is specified; this lets us add stage2= and then the isohybrid'd images can be dd'd to a usb stick and stage2 gets find there as well as on the CD. Next step is to be able to get rid of the auto CD finding by always explicitly specifying stage2= --- loader/loader.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/loader/loader.c b/loader/loader.c index 8d92ed4..b48b0a0 100644 --- a/loader/loader.c +++ b/loader/loader.c @@ -1247,7 +1247,7 @@ static char *doLoaderMain(struct loaderData_s *loaderData, * it. However if stage2= was given, use that value as an override here. * That will also then bypass any method selection UI in loader. */ - if (!FL_ASKMETHOD(flags) && !loaderData->stage2Data) { + if (!FL_ASKMETHOD(flags)) { url = findAnacondaCD("/mnt/stage2"); if (url) { setStage2LocFromCmdline(url, loaderData); @@ -1261,7 +1261,7 @@ static char *doLoaderMain(struct loaderData_s *loaderData, skipLangKbd = 1; flags |= LOADER_FLAGS_NOPASS; - } else if (loaderData->instRepo) { + } else if (!loaderData->stage2Data && loaderData->instRepo) { /* If no CD/DVD with a stage2 image was found and we were given a * repo=/method= parameter, try to piece together a valid setting * for the stage2= parameter based on that. -- 1.6.1

3 3

[PATCH] Remove unresolvable file devices from the devicetree. (#503830)
by David Lehman 21 Jul '09

21 Jul '09

--- storage/__init__.py | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/storage/__init__.py b/storage/__init__.py index efd21a0..18c7645 100644 --- a/storage/__init__.py +++ b/storage/__init__.py @@ -1541,6 +1541,8 @@ class FSSet(object): log.error("cannot determine which device contains " "directory %s" % device.path) device.parents = [] + self.devicetree._removeDevice(device) + continue else: device.parents = [parent] @@ -1616,6 +1618,8 @@ class FSSet(object): log.error("cannot determine which device contains " "directory %s" % device.path) device.parents = [] + self.devicetree._removeDevice(device) + continue else: device.parents = [parent] -- 1.6.0.6

2 2

some patches for rawhide
by David Cantrell 20 Jul '09

20 Jul '09

1) Add support for the --gecos option on the kickstart 'user' command. 2) Move 'make updates' target to script/makeupdates. This script can be run without going through autogen.sh. If you run it standalone, it'll generate an updates.img file containing only Python code that has changed. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

2 3

rhel5-branch fix for 473747
by David Cantrell 20 Jul '09

20 Jul '09

Correct the UI message to say '300' instead of '30' for the upper bound. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

2 2

[PATCH] rhel5-branch completely remove ext4dev
by Joel Granados Moreno 20 Jul '09

20 Jul '09

Hello list Remove leftovers from ext4dev. Regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

2 2

[PATCH] Fix selection of alternative iface in UI after fail (#507084).
by Radek Vykydal 17 Jul '09

17 Jul '09

Make ifcfg configuration files getting parsed properly after update. The used inotify configuration update mechanism (ifcfg-rh NM plugin) requires 1) writing of new files out of /etc/sysconfig/network-scripts dir so that it doesn't trigger parsing of the file too early (before all is written), and more importantly, 2) removing of the old files before moving the new ones in so that the new file gets parsed and the respective connection gets re-read and eventually activated. Also make sure that only device selected in UI has ONBOOT set to yes in case of selection of another device after fail. --- iw/netconfig_dialog.py | 8 ++++++-- network.py | 7 ++++++- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/iw/netconfig_dialog.py b/iw/netconfig_dialog.py index 29afc0e..ea59613 100644 --- a/iw/netconfig_dialog.py +++ b/iw/netconfig_dialog.py @@ -226,8 +226,12 @@ class NetworkConfigurator: combo = self.xml.get_widget("interfaceCombo") active = combo.get_active_iter() val = combo.get_model().get_value(active, 1) - netdev = self.network.available()[val] - netdev.set(('ONBOOT', 'yes')) + for v, dev in self.network.available().items(): + if v == val: + dev.set(('ONBOOT', 'yes')) + netdev = dev + else: + dev.set(('ONBOOT', 'no')) # FIXME: need to do input validation if self.xml.get_widget("dhcpCheckbutton").get_active(): diff --git a/network.py b/network.py index f597c9d..9a8d7d2 100644 --- a/network.py +++ b/network.py @@ -538,7 +538,7 @@ class Network: ipv6autoconf = dev.get('IPV6_AUTOCONF').lower() dhcpv6c = dev.get('DHCPV6C').lower() - newifcfg = "%s/ifcfg-%s.new" % (netscripts, device,) + newifcfg = "/tmp/ifcfg-%s.new" % (device,) f = open(newifcfg, "w") if len(dev.get("DESC")) > 0: f.write("# %s\n" % (dev.get("DESC"),)) @@ -596,6 +596,11 @@ class Network: # move the new ifcfg in place destcfg = "%s/ifcfg-%s" % (netscripts, device,) + try: + os.remove(destcfg) + except OSError as e: + if e.errno != 2: + raise shutil.move(newifcfg, destcfg) # XXX: is this necessary with NetworkManager? -- 1.6.0.6

3 2

[PATCH] Show MAC address of network device in text mode too.
by Radek Vykydal 17 Jul '09

17 Jul '09

I came accross this by the way when working on something different. It adds what was done with commit 2e8dfb72c0c296f24f99512e8fb00c377417a60a (#504216) for GUI to text ui. Althoug BZ suggests that it was already present in text ui, looking into code doesn't (I have no rawhide image to test if it is actually true). --- textw/netconfig_text.py | 7 ++++--- 1 files changed, 4 insertions(+), 3 deletions(-) diff --git a/textw/netconfig_text.py b/textw/netconfig_text.py index 8475f00..379bedd 100644 --- a/textw/netconfig_text.py +++ b/textw/netconfig_text.py @@ -103,9 +103,10 @@ class NetworkConfiguratorText: selected_interface = None for dev in devs: - desc = netdevs[dev].get("DESC") - if desc: - desc = "%s - %.50s" % (dev, desc) + hwaddr = netdevs[dev].get("HWADDR") + + if hwaddr: + desc = "%s - %.50s" % (dev, hwaddr) else: desc = dev -- 1.6.0.6

3 2

[PATCH] Stop the cdrom device before ejecting (#505067)
by Martin Sivak 17 Jul '09

17 Jul '09

--- storage/devices.py | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/storage/devices.py b/storage/devices.py index 02c6945..5bfd31f 100644 --- a/storage/devices.py +++ b/storage/devices.py @@ -2993,6 +2993,9 @@ class OpticalDevice(StorageDevice): if not self.exists: raise DeviceError("device has not been created", self.path) + #try to umount and close device before ejecting + self.teardown() + # Make a best effort attempt to do the eject. If it fails, it's not # critical. fd = os.open(self.path, os.O_RDONLY | os.O_NONBLOCK) -- 1.5.4.3

3 2

[PATCH] Add hipersockets to NETTYPE description (BZ 511962)
by Brad Hinson 16 Jul '09

16 Jul '09

Cosmetic issue. Add "hipersockets" to help text for nettype question on s390. --- anaconda-11.1.2.185/loader2/linuxrc.s390.orig +++ anaconda-11.1.2.185/loader2/linuxrc.s390 @@ -1068,7 +1068,7 @@ function question_choices_nettype() { function helptext_nettype() { echo $" Help text for network type:" echo $" qeth: OSA-Express Fast Ethernet, Gigabit Ethernet (including 1000Base-T)," - echo $" High Speed Token Ring, and ATM (running Ethernet LAN emulation)" + echo $" High Speed Token Ring, Hipersockets, and ATM (running Ethernet LAN emulation)" echo $" features in QDIO mode." echo $" [default]" echo $" lcs: OSA-2 Ethernet/Token Ring, OSA-Express Fast Ethernet in non-QDIO mode,"

2 2

Long Formatting Time when Installing from livecd on netbooks- Bug #502771 - Investigation and patch
by Eric Kerin 16 Jul '09

16 Jul '09

I think the approval queue ate my first email, let's try this again... ---------- Forwarded message ---------- From: Eric Kerin <eric_kerin(a)us.jabil.com> Date: Thu, Jul 9, 2009 at 4:18 PM Subject: Installing livecd from USB stick - Bug #502771 - Investigation and patch To: anaconda-devel-list(a)redhat.com Cc: clumens(a)redhat.com I put this in bugzilla in the closed bug - it was marked as a duplicate of another, it was suggested that I post it here. Quick summary - mke2fs will finish it's work quickly, and anaconda will spend quite sometime reading in the output, and updating the progress dialog in certain cases - the attached patch adds a check to the read/progress loop to check to see if the process has finished, and jumps out early if that's the case. Long story - I'm working on a netbook livecd install for Fedora. With the screen size of the machine I have to test on, anaconda goes into full-screen mode, the progress dialog goes full-screen as well (and remains hidden too, have not checked into this yet), it looks like because of this, it takes a long time for the progress pulses to return. (FYI - my machine also happens to be running maximus, but anaconda is excluded from it's maximizing effects) I ran into the same problem as others have, where you can hit next early, and applied the patch to make the progress dialog modal, but still ran into long periods of time where it was "formatting" with no disk IO. A quick check, and mke2fs was defunct, and anaconda was using up CPU nicely. I did some digging, and found that it was still reading the stdout data from mke2fs (it was printed very slowly, one character at a time to /dev/tty5) What I found, was that since iutil.py's execWithPulseProgress method reads one character at a time, in cases where the progress.pulse() method takes a long time to return, it slows up the reading of data, and the eventual reaping of the process. Which manifests as a defunct mke2fs process. If you give it long enough, it will continue the install. Long enough being 10 minutes or so, depending on your system. There's a loop in gui.py that smooths the pulses if it took a long time between calls, that's where a lot of the delay is spent, doing processEvents(), and because of the loop and smoothing it can start to take even longer - the first call to pulse that waits is is for <1s, next one smoothing a delta of 5 seconds, the next call is for 20 s delta, then 80 s delta...Eventually, it reads every character, after this exponentially increasing wait state. The attached patch does two things, only the change to iutil.py is really necessary - it just checks to see if the child process has exited already, and breaks if it is, since we don't need the output of the child process anyway. The patch for gui.py just keeps the delay from getting exponentially longer. Thanks, Eric Kerin

2 1

[PATCH] Multiple fstab entries of a single nodev fstype
by David Lehman 16 Jul '09

16 Jul '09

This patch set handles cases like multiple tmpfs entries in /etc/fstab on upgrade/rescue.

2 4

Fix installation of grub on md raid 1 devices.
by Radek Vykydal 16 Jul '09

16 Jul '09

Comes in two patches. Behaviour is described in comment of 2/2 patch, and corresponds to 3) from https://www.redhat.com/archives/anaconda-devel-list/2009-July/msg00157.html I tested the patches for these configs: disks sda, sdb, sdc /boot: grub target: -> grub was installed on: sdb1 sda (MBR) sda sdb1 sdb1 sdb1 md0 (sda1+sdb1) sda (MBR) sda, sdb md0 (sda1+sdb1) md0 sda1, sdb1 md0 (sdb1+sdc1) sda (MBR) sda, sdb, sdc md0 (sdb1+sdc1) md0 sdb1, sdc1 without /boot partition, /: grub target: -> grub was installed on: md0 (sda1+sdb1) sda sda, sdb md0 (sda1+sdb1) md0 sda1, sdb1 all boot OK, but I didn't tested behaviour after disk removals much (I got grub> prompt in cases I tried) Note: - UI (grub target selection wording) should be perhaps fixed to reflect the changes - Update of grub should be fixed to reflect the changes: up to this patch (f11) boot=/dev/md0 we wrote into /etc/sysconfig/grub or grub.conf meant that bootloader was installed on mbrs, and after patching it would mean that it is on raid device (partitions). We use this value in upgradeGrub.

2 3

[PATCH] Handle editing of LUKS partitions for which we have no key. (#502310)
by David Lehman 16 Jul '09

16 Jul '09

--- iw/partition_dialog_gui.py | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/iw/partition_dialog_gui.py b/iw/partition_dialog_gui.py index d38aa4e..9002175 100644 --- a/iw/partition_dialog_gui.py +++ b/iw/partition_dialog_gui.py @@ -215,7 +215,10 @@ class PartitionEditor: # preexisting partition request = self.origrequest if request.format.type == "luks": - usedev = self.storage.devicetree.getChildren(request)[0] + try: + usedev = self.storage.devicetree.getChildren(request)[0] + except IndexError: + usedev = request else: usedev = request -- 1.6.0.6

2 2

[PATCH] Allow creation of four primary partitions on a disk. (#505269)
by David Lehman 16 Jul '09

16 Jul '09

--- storage/partitioning.py | 15 ++++++++------- 1 files changed, 8 insertions(+), 7 deletions(-) diff --git a/storage/partitioning.py b/storage/partitioning.py index d5a8ea7..7e7a05a 100644 --- a/storage/partitioning.py +++ b/storage/partitioning.py @@ -706,10 +706,6 @@ def allocatePartitions(disks, partitions): sectorSize = disk.device.physicalSectorSize best = None - # TODO: On alpha we are supposed to reserve either one or two - # MB at the beginning of each disk. Awesome. - # -- maybe we do not care about alpha... - log.debug("checking freespace on %s" % _disk.name) new_part_type = getNextPartitionType(disk) @@ -719,9 +715,14 @@ def allocatePartitions(disks, partitions): continue if _part.req_primary and new_part_type != parted.PARTITION_NORMAL: - # we need a primary slot and none are free on this disk - log.debug("no primary slots available on %s" % _disk.name) - continue + if disk.primaryPartitionCount < disk.maxPrimaryPartitionCount: + # don't fail to create a primary if there are only three + # primary partitions on the disk (#505269) + new_part_type = parted.PARTITION_NORMAL + else: + # we need a primary slot and none are free on this disk + log.debug("no primary slots available on %s" % _disk.name) + continue best = getBestFreeSpaceRegion(disk, new_part_type, -- 1.6.0.6

2 1

[PATCH 1/2] Call mdadm -I with --no-degraded for all disks but the last
by Hans de Goede 15 Jul '09

15 Jul '09

We do not want to bring arrays up (and esp not use, so not scan them) in degraded mode, so add --no-degraded to the mdadm invocation. Note we do not add --no-degraded to the mdadm invocation for the last disk of the set, because doing so assembles the array in an inactive (and thus unusable) state. --- storage/devices.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/storage/devices.py b/storage/devices.py index b46479b..f3f29a4 100644 --- a/storage/devices.py +++ b/storage/devices.py @@ -2426,7 +2426,7 @@ class MDRaidArrayDevice(StorageDevice): device.setup() udev_settle(timeout=10) try: - mdraid.mdadd(device.path) + mdraid.mdadd(device.path, len(self.devices) < self.memberDevices) except MDRaidError as e: log.warning("failed to add member %s to md array %s: %s" % (device.path, self.path, e)) -- 1.6.2.2

4 6

[PATCH 1/3] Call mdadm -I with --no-degraded for all disks but the last
by Hans de Goede 15 Jul '09

15 Jul '09

We do not want to bring arrays up (and esp not use, so not scan them) in degraded mode, so add --no-degraded to the mdadm invocation. Note we do not add --no-degraded to the mdadm invocation for the last disk of the set, because doing so assembles the array in an inactive (and thus unusable) state. --- storage/devices.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/storage/devices.py b/storage/devices.py index b46479b..f3f29a4 100644 --- a/storage/devices.py +++ b/storage/devices.py @@ -2426,7 +2426,7 @@ class MDRaidArrayDevice(StorageDevice): device.setup() udev_settle(timeout=10) try: - mdraid.mdadd(device.path) + mdraid.mdadd(device.path, len(self.devices) < self.memberDevices) except MDRaidError as e: log.warning("failed to add member %s to md array %s: %s" % (device.path, self.path, e)) -- 1.6.2.2

2 3

[PATCH 1/3] Fix lvm on top of dmraid
by Hans de Goede 15 Jul '09

15 Jul '09

The lvm rules were failing to identify PV's on mdraid because blkid/vol_id has not been run on md raidsets by previous rules. --- modules.d/90lvm/64-lvm-vol_id.rules | 1 + modules.d/90lvm/64-lvm.rules | 1 + 2 files changed, 2 insertions(+), 0 deletions(-) diff --git a/modules.d/90lvm/64-lvm-vol_id.rules b/modules.d/90lvm/64-lvm-vol_id.rules index d3a07c1..f5f3a6f 100644 --- a/modules.d/90lvm/64-lvm-vol_id.rules +++ b/modules.d/90lvm/64-lvm-vol_id.rules @@ -8,6 +8,7 @@ SUBSYSTEM!="block", GOTO="lvm_end" ACTION!="add|change", GOTO="lvm_end" KERNEL=="dm-*", IMPORT{program}="vol_id --export $tempnode" +KERNEL=="md*", IMPORT{program}="/sbin/blkid -o udev -p $tempnode" ENV{ID_FS_TYPE}=="LVM2_member", \ TEST!="/tmp/.lvm_scan-%k", \ diff --git a/modules.d/90lvm/64-lvm.rules b/modules.d/90lvm/64-lvm.rules index 3403020..a1c2d1f 100644 --- a/modules.d/90lvm/64-lvm.rules +++ b/modules.d/90lvm/64-lvm.rules @@ -8,6 +8,7 @@ SUBSYSTEM!="block", GOTO="lvm_end" ACTION!="add|change", GOTO="lvm_end" KERNEL=="dm-*", IMPORT{program}="/sbin/blkid -o udev -p $tempnode" +KERNEL=="md*", IMPORT{program}="/sbin/blkid -o udev -p $tempnode" ENV{ID_FS_TYPE}=="LVM2_member", \ TEST!="/tmp/.lvm_scan-%k", \ -- 1.6.2.2

2 4

[PATCH] Run firstaidkit-qs script instead of the shell (shows rescue menu) (#508512)
by Martin Sivak 14 Jul '09

14 Jul '09

--- rescue.py | 25 ++++++++++++++++++++++++- scripts/upd-instroot | 4 +++- 2 files changed, 27 insertions(+), 2 deletions(-) diff --git a/rescue.py b/rescue.py index 665c56c..ba8455c 100644 --- a/rescue.py +++ b/rescue.py @@ -34,6 +34,7 @@ import iutil import shutil import time import network +import subprocess import gettext _ = lambda x: gettext.ldgettext("anaconda", x) @@ -128,6 +129,22 @@ def makeMtab(instPath, fsset): finally: f.close() +def makeFStab(instPath = ""): + if os.access("/etc/mtab", os.R_OK): + f = open("/etc/mtab" %(instPath,), "r") + buf = f.read() + f.close() + else: + buf = "" + + try: + f = open(instPath + "/etc/fstab", "w+") + if buf: + f.write(buf) + f.close() + except IOError, e: + log.info("failed to write /etc/fstab: %s" % e) + # make sure they have a resolv.conf in the chroot def makeResolvConf(instPath): if not os.access("/etc/resolv.conf", os.R_OK): @@ -183,7 +200,10 @@ def runShell(screen = None, msg=""): "system will reboot.")) print - if os.path.exists("/bin/bash"): + if os.path.exists("/usr/bin/firstaidkit-qs") and os.path.exists("/usr/bin/dialog"): + proc = subprocess.Popen(["/bin/firstaidkit-qs"]) + proc.wait() + elif os.path.exists("/bin/bash"): iutil.execConsole() else: print(_("Unable to find /bin/sh to execute! Not starting shell")) @@ -436,6 +456,9 @@ def runRescue(anaconda, instClass): log.error("error making a resolv.conf: %s" %(e,)) msgStr = _("Your system is mounted under the %s directory.") % (anaconda.rootPath,) + #create /etc/fstab in ramdisk, so it is easier to work with RO mounted filesystems + makeFStab() + # run %post if we've mounted everything if anaconda.isKickstart: from kickstart import runPostScripts diff --git a/scripts/upd-instroot b/scripts/upd-instroot index 8a6ddca..31e5d89 100755 --- a/scripts/upd-instroot +++ b/scripts/upd-instroot @@ -235,7 +235,7 @@ PACKAGESRESCUE="bzip2 bzip2-libs dump libidn modutils mtools mt-st mtr net-tools open smartmontools openssh openssh-clients pciutils rsh traceroute tar rsync device-mapper device-mapper-libs dmraid ntfsprogs samba-client - firstaidkit-plugin-all" + firstaidkit firstaidkit-plugin-all dialog" if [ $ARCH = i386 -o $ARCH = i586 -o $ARCH = x86_64 ]; then PACKAGESRESCUE="$PACKAGESRESCUE gpart grub" @@ -815,6 +815,8 @@ usr/sbin/smartctl usr/sbin/traceroute usr/$LIBDIR/firstaidkit/plugins/ usr/lib/python?.?/site-packages/pyfirstaidkit/ +usr/bin/dialog +usr/bin/firstaidkit-qs EOF if [ $ARCH = i386 -o $ARCH = i586 -o $ARCH = x86_64 ]; then -- 1.5.4.3

3 5

breaking exn handling out of rhpl & anaconda
by Chris Lumens 13 Jul '09

13 Jul '09

I've spent the last few days trying to break the exception handling out of both rhpl and anaconda, combining them into their own project. The benefits for this are obvious: helping to get rid of rhpl, and letting tons of other programs make use of save-to-bugzilla. Anyway, my current work in progress is here: http://clumens.fedorapeople.org/exn/ Making this generic enough is a little annoying but I think I've managed it. The UI is probably the most complicated part. My reasoning for doing it like you see is so we can still support text and graphical interfaces, and so that anaconda can provide its own wacky UI classes that override what I've done. That's why the abstract interfaces look awful familiar. test.py is obviously not a part of this module, but it does show how to use the thing from a programmer's point of view. This is likely to get a little more complicated when I work on bugzilla support but that's just the nature of working with bugzilla. Things still left to do: - test save to bugzilla - it should "compile", but probably doesn't do anything - work up a patch for anaconda, rhpl, firstboot, system-config-*, etc. - do all the typical project stuff (translations, spec file, etc.) - add wait dialogs as well - there's a bug about this somewhere - write docs describing how to make this easier to use - make sure everything's robust - your pet requirement here Any questions? - Chris

1 0

ADP Anaconda Patches, 7/9/2009 Edition
by Alford, Seth 13 Jul '09

13 Jul '09

The ADP Anaconda Patches, 7/9/2009 Edition $Id: README /main/10 2009/07/10 00:12:59 alfords Exp $ Copyright (c) 2001-2009 Automatic Data Processing Inc. NOTICE: This is a README which describes a series of anaconda patches which ADP is redistributing. We are publishing this software to fulfill the requirements of the GPL. This software is offered with NO WARRANTY and NO SUPPORT of any kind. If you are reading this and you are not an ADP employee, see subsection 1.2, "Licensing, Intended Audience, and a Note for Non-ADP Readers", below. CHANGELOG: 6/30/2009 Completely rewrote this README from a previous version. I deleted a lot of extraneous information, including syslinux changes, and some suggestions for how to debug anaconda. The syslinux changes are described in a separate posting I made to the syslinux mailing list. Other features or debugging techniques are described in documentation from RedHat. I also included information specific to individual patches with those specific patches, rather than having all that documentation here. That makes this README much shorter. Since it's shorter, people might actually read it. Since this is a complete rewrite, I restarted the CHANGELOG. TABLE OF CONTENTS: 1. INTRODUCTION 1.1 What is this? 1.2 Licensing, Intended Audience, and a Note for Non-ADP Readers 1.3 What ADP Needed versus Anaconda's Capabilities 2. HELPFUL RESOURCES 2.1 What to Read 2.2 Preparing for Anaconda Development 1. INTRODUCTION 1.1 What is this? Some years ago, ADP modified the anaconda which came with RedHat 7.0, and, later, RedHat 8.0. To comply with the GPL, we redistributed changes that we made to those versions of anaconda. To explain what we did, we wrote a large README which we posted with the modifications. ADP has now modified the anaconda which came with CentOS 5.3. The newer version of anaconda is more powerful than the older versions. The newer anaconda is much closer to what ADP needs. But, we still had to modify anaconda itself to get it to do what we wanted. We're now sharing those modifications. That's so that: 1. We fulfill the requirements of the GPL 2. Our modifications might be picked up and included in the base code. That way we don't have to make the same modifications again, later. Also, rather than produce a single large README, I split up the information between this smaller README and each patch. 1.2 Licensing, Intended Audience, and a Note for Non-ADP Readers An intended audience for this README are developers at ADP, who might have to further modify anaconda in the future. Note that I said "might". ADP makes no promise that ADP will make any further modifications to this software. As I said above, this software is offered with NO WARRANTY and NO SUPPORT of any kind. A secondary intended audience for this README are people who are not at ADP. We could have written two versions of this README file: one for ADP readers, and one for non-ADP readers. But, we did not want to have to maintain internal and external versions of this README and/or the accompanying software. Hence these patches may contain some references to non-GPL'd software, or ADP's build environment, which are "independent and separate works" under section 2 of the GPL, and therefore not covered by the GPL. THEREFORE, WE ARE NOT PUBLISHING ANYTHING ELSE UNDER THE GPL, INCLUDING BUT NOT LIMITED TO: * files which do not appear in the patches, * ADP's proprietary backup and restore software, * our build environment, * other ADP proprietary software referenced or included by this README, including but not limited to DSDA. That means: * this README may contain some references which won't make sense to non-ADP readers. These include, but are not limited to, our build environment, and build process. We don't provide explanation for these references, either. * other parts of anaconda, as modified, may also be non-functional. * some readers may find themselves saying, "yes, I know about that already." Other readers may say, "I don't understand that reference, it must be an ADP-ism." 1.3 What ADP Needed versus Anaconda's Capabilities For ADP's purposes, anaconda, even with a customized kickstart file, lacked several features, including: * simplified messages for the user * completely automatic partitioning on a variety of disk configurations, and support for some specific-to-ADP disk partitioning requirements * support for tape restore * --userinput flag to pre and post install scripts * -s flag passed to pre and post install scripts when running on a serial console * support for a restore install type 2. HELPFUL RESOURCES 2.1 What to Read Here is an incomplete list of resources that I think will help you in learning how to modify anaconda and kickstart: * Beazley's _Python Essential Reference_. Remember, anaconda is written in Python. Check which version of Python comes with your version of anaconda. Make sure that your version of Beazley addresses a matching version of Python. Otherwise, you may find yourself trying to take advantage of a new Python feature that doesn't exist in your environment. Or, you might find you are puzzled by a Python feature in one of the anaconda sources that Beazley does not explain. * The RedHat Enterprise Linux Installation Guide, available at http://www.redhat.com/docs/en-US/Red_Hat_Enterprise_Linux/5/html/Installati… * The kickstart-list and anaconda-dev mailing lists. You can join at www.redhat.com. The list archives are located there, too. * Most importantly, read Chris Lumens' pykickstart documentation. You should be able to get this by installing the pykickstart rpm then looking in /usr/share/doc/pykickstart*/programmers-guide . 2.2 Preparing for Anaconda Development Install at least the python, python-devel, pykickstart, anaconda, and anaconda-runtime packages and their dependencies on your Linux development system. This message and any attachments are intended only for the use of the addressee and may contain information that is privileged and confidential. If the reader of the message is not the intended recipient or an authorized representative of the intended recipient, you are hereby notified that any dissemination of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by e-mail and delete the message and any attachments from your system.

2 15

[PATCH][RHEL5.4] update support for ext4 in anaconda
by Jeffrey Bastian 13 Jul '09

13 Jul '09

ext4 support is being updated in RHEL 5.4 from an early development release to the stable code from Fedora 10. See https://bugzilla.redhat.com/show_bug.cgi?id=485315 As a result, the kernel module has changed names from ext4dev.ko to just ext4.ko, and anaconda no longer lists ext4/ext4dev as an available filesystem (if the 'ext4' command line option is used). See https://bugzilla.redhat.com/show_bug.cgi?id=510634 The attached patch updates anaconda to use the new module name. More work is probably needed to drop ext4dev. Jeff

2 1

[PATCH] Support for bootloader on md raid devices.
by Radek Vykydal 13 Jul '09

13 Jul '09

In f10, for /boot on md raid 1 we used to have only option to install bootloader on partition but in fact, we installed on MBR. In f11 we added option to instal to MBR in UI, and we are still installing on MBR for both options. In rhel 5 we are always installing bootloader on MBR too. Also grub-install /dev/md0 installs on MBR, so maybe this is not something we want to do at all, but in that case, we should probably remove the option to install bootloader on /dev/md0 raid device from UI. Two issues: * The patch assumes that 'mdX' bootloader target is md raid 1 device with members being partitions. Intel bios raid using mdraid (or whole-disk md devices in general?) need to be handled/considered. * If we allow (start to do) installing on md raid partitions, upgrade should be patched to count with this possibility. A little problem is that up to f11 boot=/dev/md0 we write into /etc/sysconfig/grub or grub.conf means that bootloader was installed on mbr, and after patching it would mean that it is on raid device (partitions). --- booty/x86.py | 17 ++++------------- 1 files changed, 4 insertions(+), 13 deletions(-) diff --git a/booty/x86.py b/booty/x86.py index 3890e69..8527442 100644 --- a/booty/x86.py +++ b/booty/x86.py @@ -112,14 +112,13 @@ class x86BootloaderInfo(efiBootloaderInfo): for bootDev in bootDevs: cmds = [] - gtPart = self.getMatchingPart(bootDev, grubTarget) - gtDisk = self.grubbyPartitionName(getDiskPart(gtPart, self.storage)[0]) + gtDisk = self.grubbyPartitionName(getDiskPart(bootDev, self.storage)[0]) bPart = self.grubbyPartitionName(bootDev) cmd = "root %s\n" % (bPart,) stage1Target = gtDisk if target == "partition": - stage1Target = self.grubbyPartitionName(gtPart) + stage1Target = bPart cmd += "install %s%s/stage1 d %s %s/stage2 p %s%s/grub.conf" % \ (args, grubPath, stage1Target, grubPath, bPart, grubPath) @@ -155,7 +154,8 @@ class x86BootloaderInfo(efiBootloaderInfo): if grubTarget[-2] == 'p' or \ (grubTarget[-2].isdigit() and grubTarget[-3] == 'p'): target = "partition" - elif grubTarget[-1].isdigit() and not path.startswith('md'): + # NOTE: this includes also mdX grub targets + elif grubTarget[-1].isdigit(): target = "partition" f = open(cf, "w+") @@ -362,15 +362,6 @@ class x86BootloaderInfo(efiBootloaderInfo): return 0 - def getMatchingPart(self, bootDev, target): - bootName, bootPartNum = getDiskPart(bootDev, self.storage) - devices = self.getPhysicalDevices(target) - for device in devices: - name, partNum = getDiskPart(device, self.storage) - if name == bootName: - return device - return devices[0] - def grubbyDiskName(self, name): return "hd%d" % self.drivelist.index(name) -- 1.6.0.6

2 3

[PATCH] Use tail process to ensure line buffered output to screen and log file at the same moment... (#506664)
by Martin Sivak 13 Jul '09

13 Jul '09

--- iutil.py | 84 ++++++++++++++++++++++++++++++++++++++++++++++++++------------ 1 files changed, 68 insertions(+), 16 deletions(-) diff --git a/iutil.py b/iutil.py index ba1d0b4..0a4adfa 100644 --- a/iutil.py +++ b/iutil.py @@ -33,6 +33,7 @@ from flags import flags from constants import * from rhpl.translate import _ import re +import select import logging log = logging.getLogger("anaconda") @@ -81,37 +82,88 @@ def execWithRedirect(command, argv, stdin = None, stdout = None, runningLog = open("/tmp/program.log", "a") runningLog.write("Running... %s\n" % ([command] + argv,)) + + #prepare os pipes for feeding tee proceses + pstdout, pstdin = os.pipe() + perrout, perrin = os.pipe() env = os.environ.copy() env.update({"LC_ALL": "C"}) + #prepare "empty" tee proceses so we can properly detect errors + proc_std = None + proc_err = None + try: - proc = subprocess.Popen([command] + argv, stdin=stdin, - stdout=subprocess.PIPE, + #create output pipes with logging (it needs to be different process to avoid deadlock and EOF detection..) + proc_std = subprocess.Popen(["tee", "-a", "/tmp/program.log"], stdin=pstdout, + stdout=stdout, stderr=subprocess.PIPE, + cwd=root, + env=env, bufsize = 1) + + proc_err = subprocess.Popen(["tee", "-a", "/tmp/program.log"], stdin=perrout, + stdout=stderr, + stderr=subprocess.PIPE, + cwd=root, + env=env, bufsize = 1) + + #run the command + proc = subprocess.Popen([command] + argv, stdin=stdin, + stdout=pstdin, + stderr=perrin, preexec_fn=chroot, cwd=root, - env=env) + env=env, bufsize = 1) + + proc.wait() + + #close the output pipes + if proc_std: + proc_std.terminate() + proc_std.wait() + proc_std = None + + if proc_err: + proc_err.terminate() + proc_err.wait() + proc_err = None + + ret = proc.returncode - while True: - (outStr, errStr) = proc.communicate() - if outStr: - os.write(stdout, outStr) - runningLog.write(outStr) - if errStr: - os.write(stderr, errStr) - runningLog.write(errStr) - if proc.returncode is not None: - ret = proc.returncode - break except OSError as e: errstr = "Error running %s: %s" % (command, e.strerror) log.error(errstr) + + #free descriptors + os.close(pstdout) + os.close(perrout) + os.close(pstdin) + os.close(perrin) + + #close the output pipes + if proc_std: + proc_std.terminate() + proc_std.wait() + proc_std = None + + if proc_err: + proc_err.terminate() + proc_err.wait() + proc_err = None + runningLog.write(errstr) runningLog.close() raise RuntimeError, errstr runningLog.close() + + #free descriptors + os.close(pstdout) + os.close(perrout) + os.close(pstdin) + os.close(perrin) + return ret ## Run an external program and capture standard out. -- 1.5.4.3

1 1

[PATCH] rhel5 include graphical libs.
by Joel Granados Moreno 13 Jul '09

13 Jul '09

Hello List The libdrm is needed to make ATI Technologies Inc M24 work on rhel5. make sure we include the sofile in the image. Review appreciated. Regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

2 2

[master] expand info in initialize disk dialog (#460697)
by David Cantrell 11 Jul '09

11 Jul '09

Originally filed for RHEL-5. I posted a patch earlier and Steffen Maier commented on it saying it would be useful to have more information for the drive in question, especially if you have a large number of the same type in the system. The dialog box could probably be improved, but I'm mostly just going for functionality with this patch. In addition to displaying the device model string and the size in megabytes (as originally requested in the BZ), I added a device details string if we can find it in /dev/disk/by-path on the system. Example output on my system: Error processing drive sda 305243MB (ATA ST3320620AS). Maybe it needs to be reinitialized. YOU WILL LOSE ALL DATA ON THIS DRIVE IN YOUR REINITIALIZE IT! Device details: pci-0000:00:1f.2-scsi-0:0:0:0 For DASD and zFCP, you'll get a slightly different clown barf string for the device details. I think the dialog wording could be improved, but since Peter and Andy are working on error string cleanups, I would like to leave that to them. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

2 3

[PATCH] Don't write removable devices to /etc/fstab (#505697).
by Chris Lumens 10 Jul '09

10 Jul '09

--- storage/__init__.py | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/storage/__init__.py b/storage/__init__.py index 03b9876..7996d00 100644 --- a/storage/__init__.py +++ b/storage/__init__.py @@ -1851,6 +1851,10 @@ class FSSet(object): if not device.format.mountable and device.format.type != "swap": continue + # Don't write removable devices, either. Let the desktop do it. + if device.removable: + continue + fstype = getattr(device.format, "mountType", device.format.type) if fstype == "swap": mountpoint = "swap" -- 1.6.1.3

4 4

[master] include comps.dtd (patch from RHEL 5.4)
by David Cantrell 10 Jul '09

10 Jul '09

This is another 5.4 patch of mine, updated for master. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

3 8

[PATCH] Clean yum caches following preupgrade, too (#503096).
by Chris Lumens 10 Jul '09

10 Jul '09

--- yuminstall.py | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index 0491fad..31ed4ea 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -1584,7 +1584,7 @@ reposdir=/etc/anaconda.repos.d,/tmp/updates/anaconda.repos.d,/tmp/product/anacon repo.dirCleanup() # expire yum caches on upgrade - if anaconda.id.getUpgrade() and os.path.exists("%s/var/cache/yum" %(anaconda.rootPath,)): + if (flags.cmdline.has_key("preupgrade") or anaconda.id.getUpgrade()) and os.path.exists("%s/var/cache/yum" %(anaconda.rootPath,)): log.info("Expiring yum caches") try: iutil.execWithRedirect("yum", ["clean", "all"], @@ -1595,7 +1595,7 @@ reposdir=/etc/anaconda.repos.d,/tmp/updates/anaconda.repos.d,/tmp/product/anacon pass # nuke preupgrade - if flags.cmdline.has_key("preupgrade") and anaconda.id.getUpgrade() and os.path.exists("%s/var/cache/yum/anaconda-upgrade" %(anaconda.rootPath,)): + if flags.cmdline.has_key("preupgrade") and os.path.exists("%s/var/cache/yum/anaconda-upgrade" %(anaconda.rootPath,)): try: shutil.rmtree("%s/var/cache/yum/anaconda-upgrade" %(anaconda.rootPath,)) except: -- 1.6.1.3

2 1

[PATCH] All other teardown methods take a "recursive" argument (#506166).
by Chris Lumens 10 Jul '09

10 Jul '09

--- storage/devices.py | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/storage/devices.py b/storage/devices.py index b46479b..e236710 100644 --- a/storage/devices.py +++ b/storage/devices.py @@ -2835,7 +2835,7 @@ class FileDevice(StorageDevice): for parent in self.parents: parent.format.setup() - def teardown(self): + def teardown(self, recursive=None): StorageDevice.teardown(self) if self.format and self.format.exists and not self.format.status: self.format.device = self.path -- 1.6.1.3

2 1

[PATCH] You must be this tall to ride.
by Bill Nottingham 10 Jul '09

10 Jul '09

Adjust memory requirements. Signed-off-by: Bill Nottingham <notting(a)redhat.com> --- isys/isys.h | 10 +++------- 1 files changed, 3 insertions(+), 7 deletions(-) diff --git a/isys/isys.h b/isys/isys.h index 43355ca..15e461e 100644 --- a/isys/isys.h +++ b/isys/isys.h @@ -20,13 +20,9 @@ #ifndef H_ISYS #define H_ISYS -#define MIN_RAM 64000 -#define MIN_GUI_RAM 192000 -#if defined(__x86_64__) || defined(__ia64__) || defined(__s390x__) || defined(__ppc64__) -#define EARLY_SWAP_RAM 400000 -#else -#define EARLY_SWAP_RAM 270000 -#endif +#define MIN_RAM 131072 +#define MIN_GUI_RAM 393216 +#define EARLY_SWAP_RAM 524288 #define OUTPUT_TERMINAL "/dev/tty5" -- 1.6.2.5

2 1

Installing livecd from USB stick - Bug #502771 - Investigation and patch
by Eric Kerin 09 Jul '09

09 Jul '09

I put this in bugzilla in the closed bug - it was marked as a duplicate of another, it was suggested that I post it here. - Not subscribed please CC. Quick summary - mke2fs will finish it's work quickly, and anaconda will spend quite sometime reading in the output, and updating the progress dialog in certain cases - the attached patch adds a check to the read/progress loop to check to see if the process has finished, and jumps out early if that's the case. Long story - I'm working on a netbook livecd install for Fedora. With the screen size of the machine I have to test on, anaconda goes into full-screen mode, the progress dialog goes full-screen as well (and remains hidden too, have not checked into this yet), it looks like because of this, it takes a long time for the progress pulses to return. (FYI - my machine also happens to be running maximus, but anaconda is excluded from it's maximizing effects) I ran into the same problem as others have, where you can hit next early, and applied the patch to make the progress dialog modal, but still ran into long periods of time where it was "formatting" with no disk IO. A quick check, and mke2fs was defunct, and anaconda was using up CPU nicely. I did some digging, and found that it was still reading the stdout data from mke2fs (it was printed very slowly, one character at a time to /dev/tty5) What I found, was that since iutil.py's execWithPulseProgress method reads one character at a time, in cases where the progress.pulse() method takes a long time to return, it slows up the reading of data, and the eventual reaping of the process. Which manifests as a defunct mke2fs process. If you give it long enough, it will continue the install. Long enough being 10 minutes or so, depending on your system. There's a loop in gui.py that smooths the pulses if it took a long time between calls, that's where a lot of the delay is spent, doing processEvents(), and because of the loop and smoothing it can start to take even longer - the first call to pulse that waits is is for <1s, next one smoothing a delta of 5 seconds, the next call is for 20 s delta, then 80 s delta...Eventually, it reads every character, after this exponentially increasing wait state. The attached patch does two things, only the change to iutil.py is really necessary - it just checks to see if the child process has exited already, and breaks if it is, since we don't need the output of the child process anyway. The patch for gui.py just keeps the delay from getting exponentially longer. Thanks, Eric Kerin

1 0

[PATCH] mdmon missing in install.img
by Jacek Danecki 09 Jul '09

09 Jul '09

mdmon added to install.img mdmon application is needed to switch imsm md array to read/write mode. diff --git a/scripts/upd-instroot b/scripts/upd-instroot index 1ee3084..a560733 100755 --- a/scripts/upd-instroot +++ b/scripts/upd-instroot @@ -405,6 +405,7 @@ sbin/ldconfig sbin/lspci sbin/lvm* sbin/mdadm +sbin/mdmon sbin/mkdosfs sbin/mke2fs sbin/mkfs.btrfs

2 1

[PATCH] rhel5 include graphical libs.
by Joel Granados Moreno 09 Jul '09

09 Jul '09

Hello List The libdrm is needed to make ATI Technologies Inc M24 work on rhel5. Review appreciated. Regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

4 4

PATCH [grub]: add support for partitionable md
by Hans de Goede 09 Jul '09

09 Jul '09

2 1

[PATCH 1/2] Write out NM_CONTROLLED=no for NICs used for FCoE
by Hans de Goede 09 Jul '09

09 Jul '09

Write out NM_CONTROLLED=no for NICs used for FCoE, note that unlike with iSCSI we do not blindly write out NM_CONTROLLED=no for all NICs, but just for the NIC which is used for FCoE. The iSCSI behaviour is undesirable, but the whole writing of NM_CONTROLLED=no for iSCSI will go away as soon as NetworkManager is fixed to not down devices when it takes over control, which should be fixed soon. --- network.py | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/network.py b/network.py index e3ce2c1..b4a8c89 100644 --- a/network.py +++ b/network.py @@ -587,8 +587,9 @@ class Network: # FIXME: use device.host_address to only add "NM_CONTROLLED=no" # for interfaces actually used enroute to the device for d in anaconda.id.storage.devices: - if rootdev.dependsOn(d) and isinstance(d, - storage.devices.NetworkStorageDevice): + if rootdev.dependsOn(d) and \ + isinstance(d, storage.devices.NetworkStorageDevice) and\ + (not d.nic or d.nic == device): f.write("NM_CONTROLLED=no\n") break -- 1.6.2.2

3 7

[PATCH 1/4] 70-anaconda.rules: always import blkid output
by Hans de Goede 09 Jul '09

09 Jul '09

We are currently not calling blkid when a block device is not an mdraid set or a device mapper device, resulting in us not identifying mdraid members as such. This probably worked when still using vol_id because some other rules file was doing an import vol_id before our rules. --- 70-anaconda.rules | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/70-anaconda.rules b/70-anaconda.rules index 0efd536..0fd7647 100644 --- a/70-anaconda.rules +++ b/70-anaconda.rules @@ -4,6 +4,8 @@ SUBSYSTEM!="block", GOTO="anaconda_end" ENV{ANACBIN}="/sbin" TEST!="$env{ANACBIN}/dmsetup", ENV{ANACBIN}="/usr/sbin" +IMPORT{program}="$env{ANACBIN}/blkid -o udev -p $tempnode" + KERNEL!="dm-*", GOTO="anaconda_mdraid" IMPORT{program}="$env{ANACBIN}/dmsetup info -c --nameprefixes --unquoted --rows --noheadings -o name,uuid,suspended,readonly,major,minor,open,tables_loaded -j%M -m%m" @@ -14,7 +16,6 @@ ENV{DM_UUID}=="?*", SYMLINK+="disk/by-id/dm-uuid-$env{DM_UUID}" ENV{DM_STATE}=="SUSPENDED", GOTO="anaconda_end" -IMPORT{program}="blkid -o udev -p $tempnode" OPTIONS="link_priority=-100" ENV{ID_FS_USAGE}=="filesystem|other|crypto", ENV{ID_FS_UUID_ENC}=="?*", SYMLINK+="disk/by-uuid/$env{ID_FS_UUID_ENC}" @@ -37,7 +38,6 @@ ENV{DEVTYPE}=="partition", ENV{MD_UUID}=="?*", SYMLINK+="disk/by-id/md-uuid-$env ENV{DEVTYPE}=="partition", ENV{MD_DEVNAME}=="*[^0-9]", SYMLINK+="md/$env{MD_DEVNAME}%n" ENV{DEVTYPE}=="partition", ENV{MD_DEVNAME}=="*[0-9]", SYMLINK+="md/$env{MD_DEVNAME}p%n" -IMPORT{program}="blkid -o udev -p $tempnode" OPTIONS+="link_priority=100" OPTIONS+="watch" ENV{ID_FS_USAGE}=="filesystem|other|crypto", ENV{ID_FS_UUID_ENC}=="?*", SYMLINK+="disk/by-uuid/$env{ID_FS_UUID_ENC}" -- 1.6.2.2

2 8

[PATCH] Remove _catchallCategory. yum handles this for us now.
by Chris Lumens 09 Jul '09

09 Jul '09

--- yuminstall.py | 25 ------------------------- 1 files changed, 0 insertions(+), 25 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index d1cb626..623911d 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -1097,8 +1097,6 @@ reposdir=/etc/anaconda.repos.d,/tmp/updates/anaconda.repos.d,/tmp/product/anacon self.ayum._setGroups(None) continue - self._catchallCategory() - def doRepoSetup(self, anaconda, thisrepo = None, fatalerrors = True): self.__withFuncDo(anaconda, lambda r: self.ayum.doRepoSetup(thisrepo=r.id), thisrepo=thisrepo, fatalerrors=fatalerrors) @@ -1186,29 +1184,6 @@ reposdir=/etc/anaconda.repos.d,/tmp/updates/anaconda.repos.d,/tmp/product/anacon self.ayum.repos.callback = None - def _catchallCategory(self): - # FIXME: this is a bad hack, but catch groups which aren't in - # a category yet are supposed to be user-visible somehow. - # conceivably should be handled by yum - grps = {} - for g in self.ayum.comps.groups: - if g.user_visible: - grps[g.groupid] = g - - for cat in self.ayum.comps.categories: - for g in cat.groups: - if grps.has_key(g): - del grps[g] - - if len(grps.keys()) == 0: - log.info("no groups missing") - return - c = yum.comps.Category() - c.name = _("Uncategorized") - c._groups = grps - c.categoryid = "uncategorized" - self.ayum.comps._categories[c.categoryid] = c - def getDefaultGroups(self, anaconda): langs = anaconda.id.instLanguage.getCurrentLangSearchList() rc = map(lambda x: x.groupid, -- 1.6.1.3

2 1

[PATCH] Use a method yum provides, rather than inventing our own.
by Chris Lumens 08 Jul '09

08 Jul '09

--- yuminstall.py | 17 ++++++----------- 1 files changed, 6 insertions(+), 11 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index 623911d..478b22b 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -743,17 +743,12 @@ class AnacondaYum(YumSorter): downloadpkgs = [] totalSize = 0 totalFiles = 0 - for txmbr in self.tsInfo.getMembers(): - if txmbr.ts_state in ['i', 'u']: - po = txmbr.po - else: - continue - - if po: - totalSize += int(po.returnSimple("installedsize")) / 1024 - for filetype in po.returnFileTypes(): - totalFiles += len(po.returnFileEntries(ftype=filetype)) - downloadpkgs.append(po) + for txmbr in self.tsInfo.getMembersWithState(output_states=TS_INSTALL_STATES): + if txmbr.po: + totalSize += int(txmbr.po.returnSimple("installedsize")) / 1024 + for filetype in txmbr.po.returnFileTypes(): + totalFiles += len(txmbr.po.returnFileEntries(ftype=filetype)) + downloadpkgs.append(txmbr.po) return (downloadpkgs, totalSize, totalFiles) -- 1.6.1.3

2 1

[PATCH 1/2] pykickstart: Add fcoe command (take 2)
by Hans de Goede 08 Jul '09

08 Jul '09

Add an command to activate FCoE attached SAN's from kickstart --- pykickstart/commands/__init__.py | 4 +- pykickstart/commands/fcoe.py | 73 ++++++++++++++++++++++++++++++++++++++ pykickstart/handlers/control.py | 2 + 3 files changed, 77 insertions(+), 2 deletions(-) create mode 100644 pykickstart/commands/fcoe.py diff --git a/pykickstart/commands/__init__.py b/pykickstart/commands/__init__.py index 7ce21f7..0952bf8 100644 --- a/pykickstart/commands/__init__.py +++ b/pykickstart/commands/__init__.py @@ -18,8 +18,8 @@ # with the express permission of Red Hat, Inc. # import authconfig, autopart, autostep, bootloader, clearpart, device -import deviceprobe, displaymode, dmraid, driverdisk, firewall, firstboot, group -import ignoredisk, interactive, iscsi, iscsiname, key, keyboard, lang +import deviceprobe, displaymode, dmraid, driverdisk, fcoe, firewall, firstboot +import group, ignoredisk, interactive, iscsi, iscsiname, key, keyboard, lang import langsupport, lilocheck, logging, logvol, mediacheck, method, monitor import mouse, multipath, network, partition, raid, reboot, repo, rescue, rootpw import selinux, services, skipx, timezone, updates, upgrade, user, vnc diff --git a/pykickstart/commands/fcoe.py b/pykickstart/commands/fcoe.py new file mode 100644 index 0000000..e3ffce8 --- /dev/null +++ b/pykickstart/commands/fcoe.py @@ -0,0 +1,73 @@ +# +# Hans de Goede <hdegoede(a)redhat.com> +# +# Copyright 2009 Red Hat, Inc. +# +# This copyrighted material is made available to anyone wishing to use, modify, +# copy, or redistribute it subject to the terms and conditions of the GNU +# General Public License v.2. This program is distributed in the hope that it +# will be useful, but WITHOUT ANY WARRANTY expressed or implied, including the +# implied warranties of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# See the GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License along with +# this program; if not, write to the Free Software Foundation, Inc., 51 +# Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. Any Red Hat +# trademarks that are incorporated in the source code or documentation are not +# subject to the GNU General Public License and may only be used or replicated +# with the express permission of Red Hat, Inc. +# +from pykickstart.base import * +from pykickstart.options import * + +class F12_FcoeData(BaseData): + removedKeywords = BaseData.removedKeywords + removedAttrs = BaseData.removedAttrs + + def __init__(self, *args, **kwargs): + BaseData.__init__(self, *args, **kwargs) + self.nic = kwargs.get("nic", None) + + def __str__(self): + retval = BaseData.__str__(self) + retval += "fcoe" + + if self.nic: + retval += " --nic=%s" % self.nic + + return retval + "\n" + + +class F12_Fcoe(KickstartCommand): + removedKeywords = KickstartCommand.removedKeywords + removedAttrs = KickstartCommand.removedAttrs + + def __init__(self, writePriority=71, *args, **kwargs): + KickstartCommand.__init__(self, writePriority, *args, **kwargs) + self.op = self._getParser() + self.fcoe = kwargs.get("fcoe", []) + + def __str__(self): + retval = "" + for fcoe in self.fcoe: + retval += fcoe.__str__() + + return retval + + def _getParser(self): + op = KSOptionParser() + op.add_option("--nic", dest="nic", required=1) + return op + + def parse(self, args): + zd = self.handler.FcoeData() + (opts, extra) = self.op.parse_args(args=args, lineno=self.lineno) + if len(extra) > 0: + mapping = {"command": "fcoe", "options": extra} + raise KickstartValueError, formatErrorMsg(self.lineno, msg=_("Unexpected arguments to %(command)s command: %(options)s") % mapping) + + self._setToObj(self.op, opts, zd) + return zd + + def dataList(self): + return self.fcoe diff --git a/pykickstart/handlers/control.py b/pykickstart/handlers/control.py index 8e1ab66..8f78c3e 100644 --- a/pykickstart/handlers/control.py +++ b/pykickstart/handlers/control.py @@ -541,6 +541,7 @@ commandMap = { "deviceprobe": deviceprobe.FC3_DeviceProbe, "dmraid": dmraid.FC6_DmRaid, "driverdisk": driverdisk.FC3_DriverDisk, + "fcoe": fcoe.F12_Fcoe, "firewall": firewall.F10_Firewall, "firstboot": firstboot.FC3_Firstboot, "graphical": displaymode.FC3_DisplayMode, @@ -873,6 +874,7 @@ dataMap = { "DriverDiskData": driverdisk.F12_DriverDiskData, "DeviceData": device.F8_DeviceData, "DmRaidData": dmraid.FC6_DmRaidData, + "FcoeData": fcoe.F12_FcoeData, "GroupData": group.F12_GroupData, "IscsiData": iscsi.F10_IscsiData, "LogVolData": logvol.F9_LogVolData, -- 1.6.2.2

3 3

[PATCH] Use the yum preconf object to do $releasever substitution.
by Chris Lumens 08 Jul '09

08 Jul '09

--- yuminstall.py | 16 ++++++---------- 1 files changed, 6 insertions(+), 10 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index 5fdca80..5170e94 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -555,11 +555,11 @@ class AnacondaYum(YumSorter): # We need to make sure $releasever gets set up before .repo files are # read. Since there's no redhat-release package in /mnt/sysimage (and # won't be for quite a while), we need to do our own substutition. - def getReposFromConfig(self): - def _getReleasever(): - from ConfigParser import ConfigParser - c = ConfigParser() + def _getReleasever(): + from ConfigParser import ConfigParser + c = ConfigParser() + try: if os.access("%s/.treeinfo" % self.anaconda.methodstr, os.R_OK): ConfigParser.read(c, "%s/.treeinfo" % self.anaconda.methodstr) else: @@ -569,13 +569,8 @@ class AnacondaYum(YumSorter): ConfigParser.read(c, "/tmp/.treeinfo") return c.get("general", "version") - - try: - self.yumvar["releasever"] = _getReleasever() except: - self.yumvar["releasever"] = productVersion - - YumSorter.getReposFromConfig(self) + return productVersion # Override this method so yum doesn't nuke our existing logging config. def doLoggingSetup(self, *args, **kwargs): @@ -611,6 +606,7 @@ class AnacondaYum(YumSorter): if hasattr(self, "preconf"): self.preconf.fn = fn self.preconf.root = root + self.preconf.releasever = self._getReleasever() self.preconf.enabled_plugins = ["whiteout", "blacklist"] YumSorter._getConfig(self) else: -- 1.6.1.3

2 1

More arch fixes.
by Bill Nottingham 08 Jul '09

08 Jul '09

iutil.py | 16 +++++++++++++--- storage/devicelibs/lvm.py | 2 +- 2 files changed, 14 insertions(+), 4 deletions(-)

2 4

[master] Fix many or all 'cannot commit to disk after 5 attempts' problems
by David Cantrell 08 Jul '09

08 Jul '09

The problem I have been able to reliably reproduce is active logical volumes on the system when the storage code goes to commit changes to the disk. Added a status() property to LVMLogicalVolumeDevice that calls lvm.lvs() and reads the lv_attr field looking for an 'a' which indicates the logical volume is active. This patch causes the rest of the LVM device status code to fall in to place and the devices are torn down/deactivated correctly. -- David Cantrell <dcantrell(a)redhat.com> Red Hat, Inc. / Honolulu, HI

2 3

[PATCH] pykickstart: Add fcoe command
by Hans de Goede 07 Jul '09

07 Jul '09

Add an command to activate FCoE attached SAN's from kickstart --- pykickstart/commands/fcoe.py | 69 +++++++++++++++++++++++++++++++++++++++ pykickstart/handlers/control.py | 1 + 2 files changed, 70 insertions(+), 0 deletions(-) create mode 100644 pykickstart/commands/fcoe.py diff --git a/pykickstart/commands/fcoe.py b/pykickstart/commands/fcoe.py new file mode 100644 index 0000000..b3c01bd --- /dev/null +++ b/pykickstart/commands/fcoe.py @@ -0,0 +1,69 @@ +# +# Hans de Goede <hdegoede(a)redhat.com> +# +# Copyright 2009 Red Hat, Inc. +# +# This copyrighted material is made available to anyone wishing to use, modify, +# copy, or redistribute it subject to the terms and conditions of the GNU +# General Public License v.2. This program is distributed in the hope that it +# will be useful, but WITHOUT ANY WARRANTY expressed or implied, including the +# implied warranties of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# See the GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License along with +# this program; if not, write to the Free Software Foundation, Inc., 51 +# Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. Any Red Hat +# trademarks that are incorporated in the source code or documentation are not +# subject to the GNU General Public License and may only be used or replicated +# with the express permission of Red Hat, Inc. +# +from pykickstart.base import * +from pykickstart.options import * + +class F12_FcoeData(BaseData): + removedKeywords = BaseData.removedKeywords + removedAttrs = BaseData.removedAttrs + + def __init__(self, *args, **kwargs): + BaseData.__init__(self, *args, **kwargs) + self.nic = kwargs.get("nic", None) + + def __str__(self): + retval = BaseData.__str__(self) + retval += "fcoe" + + if self.nic: + retval += " --nic=%s" % self.nic + + return retval + "\n" + + +class F12_Fcoe(KickstartCommand): + removedKeywords = KickstartCommand.removedKeywords + removedAttrs = KickstartCommand.removedAttrs + + def __init__(self, writePriority=0, *args, **kwargs): + KickstartCommand.__init__(self, writePriority, *args, **kwargs) + self.op = self._getParser() + self.fcoe = kwargs.get("fcoe", []) + + def __str__(self): + retval = "" + for fcoe in self.fcoe: + retval += fcoe.__str__() + + return retval + + def _getParser(self): + op = KSOptionParser() + op.add_option("--nic", dest="nic", required=1) + return op + + def parse(self, args): + zd = self.handler.FcoeData() + (opts, extra) = self.op.parse_args(args=args, lineno=self.lineno) + self._setToObj(self.op, opts, zd) + return zd + + def dataList(self): + return self.fcoe diff --git a/pykickstart/handlers/control.py b/pykickstart/handlers/control.py index 8e1ab66..be6d6f3 100644 --- a/pykickstart/handlers/control.py +++ b/pykickstart/handlers/control.py @@ -541,6 +541,7 @@ commandMap = { "deviceprobe": deviceprobe.FC3_DeviceProbe, "dmraid": dmraid.FC6_DmRaid, "driverdisk": driverdisk.FC3_DriverDisk, + "fcoe": fcoe.F12_Fcoe, "firewall": firewall.F10_Firewall, "firstboot": firstboot.FC3_Firstboot, "graphical": displaymode.FC3_DisplayMode, -- 1.6.2.2

2 3

Installer device Filtering...
by Joel Granados 07 Jul '09

07 Jul '09

Hello list. There was a discussion in irc about device filtering. I just wanted to summarize and post what was discussed so we can maybe get some valuable insight from the mailing list. This is more like an irc dump than a summary. But I hope it shows some of the topics that should/could be discussed * What information to give to the filter? This is related to the UI. And this was the reason to begin the discussion. - Give a regular expression that identifies the devices - Specify with a checkbox if you want anaconda to use the regular expression. - Specify what that regular expression should do (Ignore or Include) * On what would those regular expressions operate? - canonicalized path of the device link under /sys/block. - []$ readlink -f /sys/block/sda/device /sys/devices/pci0000:00/0000:00:05.0/host0/target0:0:0/0:0:0:0 - /dev/disk/by-path version is also possible. Could be useful. * Why not have the possibility to refresh? (Refresh would work with the information contained in the kernel.) - The user tweaks the scanning list and hits refresh. - User does this until he/she is satisfied with the device list. * Three possible scanning stages where identified. Each of these might be controlled by the filter. 1. Kernel scan 2. udev scan 3. devicetree.py:scan * We actually want to be able to tell the kernel not to scan certain LUN's. * Wish people would learn how to configure fibrechannel switch acl's so we would only see the lun's they actually want us to touch. Would it be a good idea to document the configuration setup that anaconda expects? * How would udev filters aid us with the overall filtering. Regards. -- Joel Andres Granados Brno, Czech Republic, Red Hat.

6 5

[PATCH] Clean up some arch code.
by Bill Nottingham 07 Jul '09

07 Jul '09

Just use rpmUtils's definition of multilib arches, and we no longer set ppc64iseries/ppc64pseries. Signed-off-by: Bill Nottingham <notting(a)redhat.com> --- yuminstall.py | 5 ++--- 1 files changed, 2 insertions(+), 3 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index 70ed98a..acc2fba 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -762,8 +762,7 @@ class AnacondaYum(YumSorter): return (downloadpkgs, totalSize, totalFiles) def setColor(self): - if (rpmUtils.arch.canonArch.startswith("ppc64") or - rpmUtils.arch.canonArch in ("s390x", "sparc64", "x86_64", "ia64")): + if rpmUtils.arch.isMultiLibArch(): self.ts.ts.setColor(3) def run(self, instLog, cb, intf, id): @@ -1285,7 +1284,7 @@ reposdir=/etc/anaconda.repos.d,/tmp/updates/anaconda.repos.d,/tmp/product/anacon foundkernel = False - if (isys.smpAvailable() or isys.htavailable()) and not rpmUtils.arch.canonArch.startswith("ppc64"): + if (isys.smpAvailable() or isys.htavailable()) and rpmUtils.arch.canonArch != "ppc64": if selectKernel("kernel-smp"): foundkernel = True -- 1.6.2.5

2 1

[PATCH rhel 5.4] Make buildinstall error-out on failed loop mount, propagate status (#472552)
by Radek Vykydal 07 Jul '09

07 Jul '09

I forgot to check mk-images exit status in buildinstall in previous patch. --- scripts/buildinstall | 2 ++ scripts/mk-images | 6 ++++-- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/scripts/buildinstall b/scripts/buildinstall index 072f8bf..efb754f 100755 --- a/scripts/buildinstall +++ b/scripts/buildinstall @@ -161,6 +161,8 @@ $MK_TREEINFO --family="$PRODUCTSTR" ${VARIANT:+--variant="$VARIANT"} --version=$ echo "Making images..." $MK_IMAGES $DEBUGSTR $NOGRSTR $PKGDIR $p $TREEDIR/image-template $TREEDIR/instimage $BUILDARCH "$PRODUCTSTR" $VERSION $PRODUCTPATH "$BUGURL" +rc=$? +[ $rc = 0 ] || exit $rc echo "Writing .discinfo file" $MK_STAMP --releasestr="$RELEASESTR" --arch=$BUILDARCH --discNum="$DISCSTR" --baseDir=$PRODUCTPATH/base --packagesDir=$PKGDIR --pixmapsDir=$PRODUCTPATH/pixmaps --outfile=$p/.discinfo diff --git a/scripts/mk-images b/scripts/mk-images index 48ba567..8bd4790 100755 --- a/scripts/mk-images +++ b/scripts/mk-images @@ -746,7 +746,8 @@ makemainimage () { mke2fs -q -F $mmi_tmpimage > /dev/null tune2fs -c0 -i0 $mmi_tmpimage >/dev/null mount -o loop $mmi_tmpimage $mmi_mntpoint - [ $? = 0 ] || return $? + rc=$? + [ $rc = 0 ] || return $rc (cd $IMGPATH; find . | fgrep -v "./usr/lib/anaconda-runtime" | cpio -H crc -o) | (cd $mmi_mntpoint; cpio -iumd) @@ -783,7 +784,8 @@ makeSecondStage() { [ $? = 0 ] || exit 1 echo "Building stage2.img" makemainimage "stage2" "squashfs" - [ $? = 0 ] || exit 1 + rc=$? + [ $rc = 0 ] || exit $rc } doPostImages() { -- 1.6.0.6

2 1

Assorted patches for rawhide
by David Cantrell 07 Jul '09

07 Jul '09

Some RHEL-5 forward ports and some other fixes. -- David Cantrell <dcantrell(a)redhat.com> Red Hat, Inc. / Honolulu, HI

3 12

[PATCH] no "rhgb quiet" on s390 to enable visible boot progress and system automation (#509881)
by Steffen Maier 06 Jul '09

06 Jul '09

Anaconda appends "rhgb quiet" to the bootloader's kernel arguments. On s390x, there is not graphical console. A console of type line mode is most common. Even more so, it's not uncommon to do system automation based on the guest OS's console output. Therefore, allow full output of kernel boot messages on the console. --- yuminstall.py | 9 +++++---- 1 files changed, 5 insertions(+), 4 deletions(-) diff --git a/yuminstall.py b/yuminstall.py index fb89281..d96dae6 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -1614,10 +1614,11 @@ reposdir=/etc/anaconda.repos.d,/tmp/updates/anaconda.repos.d,/tmp/product/anacon w = anaconda.intf.waitWindow(_("Post Install"), _("Performing post install configuration...")) - if len(self.ayum.tsInfo.matchNaevr(name='rhgb')) > 0: - anaconda.id.bootloader.args.append("rhgb quiet") - elif len(self.ayum.tsInfo.matchNaevr(name='plymouth')) > 0: - anaconda.id.bootloader.args.append("rhgb quiet") + if not iutil.isS390(): + if len(self.ayum.tsInfo.matchNaevr(name='rhgb')) > 0: + anaconda.id.bootloader.args.append("rhgb quiet") + elif len(self.ayum.tsInfo.matchNaevr(name='plymouth')) > 0: + anaconda.id.bootloader.args.append("rhgb quiet") if self.ayum.tsInfo.getProvides("service(graphical-login)") != {} and anaconda.id.displayMode == 'g' and not flags.usevnc: anaconda.id.desktop.setDefaultRunLevel(5) -- 1.6.3.1

2 1

[PATCH] fix backtrace in s390 reipl support due to missing anaconda.id.fsset (#509877)
by Steffen Maier 06 Jul '09

06 Jul '09

anaconda.id.fsset does no longer exist and the reipl code did a back trace. Assuming anaconda.id.bootloader.device is always a disk device (not a partition) and it always contains the device /boot and hence the zipl bootloader is located on, this fixes the back trace. --- iutil.py | 26 +------------------------- 1 files changed, 1 insertions(+), 25 deletions(-) diff --git a/iutil.py b/iutil.py index 8649224..2457d4e 100644 --- a/iutil.py +++ b/iutil.py @@ -764,12 +764,6 @@ def reIPLonCCW(iplsubdev, reipl_path): device = "<unknown>" try: - iplbits = re.split ('([0-9]+)', iplsubdev) - if len (iplbits) != 3: - message = _("Error: %s splits into %s but not like we expect" % (iplsubdev,iplbits,)) - log.warning(message) - raise Exception (message) - device = os.readlink("/sys/block/" + iplbits[0] + "/device").split('/')[-1] writeReiplMethod(reipl_path, 'ccw') @@ -816,12 +810,6 @@ def reIPLonFCP(iplsubdev, reipl_path): fcpvalue = { "device": "<unknown>", "wwpn": "<unknown>", "lun": "<unknown>" } try: - iplbits = re.split ('([0-9]+)', iplsubdev) - if len (iplbits) != 3: - message = _("Error: %s splits into %s but not like we expect" % (iplsubdev,iplbits,)) - log.warning(message) - raise Exception (message) - syspath = "/sys/block/" + iplbits[0] + "/device" fcpprops = [ ("hba_id", "device"), ("wwpn", "wwpn"), ("fcp_lun", "lun") ] @@ -883,19 +871,7 @@ def reIPL(anaconda, loader_pid): reipl_path = "/sys/firmware/reipl" - iplfs = anaconda.id.fsset.getEntryByMountPoint("/boot") - if iplfs is None: - iplfs = anaconda.id.fsset.getEntryByMountPoint("/") - - if iplfs is None: - message = _("Could not get information for mount point /boot or /") - log.warning(message) - return (message, instruction) - - try: - ipldev = iplfs.device.device - except: - ipldev = None + ipldev = anaconda.id.bootloader.device if ipldev is None: message = _("Error determining mount point type") -- 1.6.3.1

2 2

[PATCH] de-busyboxify
by Chris Lumens 06 Jul '09

06 Jul '09

This set of patches removes all users of busybox from anaconda. If you're curious about the reason for this (besides the obvious), check out bug 505639. I'm tired of dealing with nonsense like this, so it's time for busybox to go. busybox.anaconda had about 250 programs linked to it. I have replaced them with the real versions of commands where appropriate, and removed at least 100 programs entirely. I've checked that all the programs we call in anaconda are still present, but these sorts of changes have the potential to break lots of things. If everyone likes this direction, I'll spend the day or two required to do a scratch build, then a scratch tree, then some testing. - Chris

2 6

We need to start thinking about writing kernel cmdline options to grub.conf for dracut
by Hans de Goede 05 Jul '09

05 Jul '09

Hi All, When we switch to dracut we will need to specify certain kernel commandline options to dracut. For example when / is on iscsi dracut will need information for configuring the nic and about the iscsi target present on the kernel cmdline. The easiest way to handle this would be to add a new attribute to all devices which returns a list of kernel cmdline options to add to bring up that device (which will be empty in most cases). The problem here is that we can need (for example) multiple iscsi disks, and the question then becomes how to merge the cmdlines returned from the various devices. Harald I've included you in the CC, because I think this is something which dracut cannot handle at all atm. The parameters like iscsi_target_name, etc. All only take one value AFAIK, but what then if we need to bring up 2 different iscsi targets, for example for doing a mirror on 2 different iscsi disks ? I'm thinking that it would be better to have an iscsi= option, which will then contain the rootdev= iscsi syntax from dhcp (so all info for one target in one string) and allow having this option multiple times, does that sound like a plan ? Regards, Hans

4 5

Re: Features/YumLangpackPlugin and anaconda consequences
by Jens Petersen 05 Jul '09

05 Jul '09

----- "Bill Nottingham" <notting(a)redhat.com> wrote: > Hans de Goede (hdegoede(a)redhat.com) said: > > If we do this through %_install_langs, we gain nothing as currently > that > > is set to "all languages" and not changable (during the install > atleast). I agree with others %_install_langs is probably not the way: I think it is not intuitive for users and hard to find. > Well, the other way is keying off $LANG/$LC_MESSAGES. That doesn't > allow for 'I want multiple languages' as well, though. My plan which I mean(t) to blog about is to have a meta-package per language for langpack support. So if you want Japanese and Spanish langpacks say, you would just install japanese-langpacks and spanish-langpacks (from the lang support groups) and yum would start listening. So if the yum plugin is available to anaconda it should be ok. Jens

2 2

[PATCH] Allow creating new groups through kickstart.
by Chris Lumens 02 Jul '09

02 Jul '09

This is executed before users are created, so new users may be added to the new groups immediately. --- instdata.py | 6 ++++++ kickstart.py | 1 + users.py | 35 +++++++++++++++++++++++++++++++++++ 3 files changed, 42 insertions(+), 0 deletions(-) diff --git a/instdata.py b/instdata.py index 07984d5..7e44177 100644 --- a/instdata.py +++ b/instdata.py @@ -182,6 +182,12 @@ class InstallData: stdout="/dev/tty5", stderr="/dev/tty5", root=self.anaconda.rootPath) + for gd in self.ksdata.group.groupList: + if not self.users.createGroup(name=gd.name, + gid=gd.gid, + root=self.anaconda.rootPath): + log.error("Group %s already exists, not creating." % gd.name) + for ud in self.ksdata.user.userList: if not self.users.createUser(name=ud.name, password=ud.password, diff --git a/kickstart.py b/kickstart.py index ce99855..5147fa5 100644 --- a/kickstart.py +++ b/kickstart.py @@ -985,6 +985,7 @@ commandMap = { "firewall": Firewall, "firstboot": Firstboot, "graphical": commands.displaymode.FC3_DisplayMode, + "group": commands.group.F12_Group, "halt": Reboot, "harddrive": commands.method.FC6_Method, "ignoredisk": IgnoreDisk, diff --git a/users.py b/users.py index ec9aa05..6534a00 100644 --- a/users.py +++ b/users.py @@ -84,6 +84,41 @@ class Users: def __init__ (self): self.admin = libuser.admin() + def createGroup (self, name=None, gid=None, root="/mnt/sysimage"): + childpid = os.fork() + + if not childpid: + os.chroot(root) + + del(os.environ["LIBUSER_CONF"]) + self.admin = libuser.admin() + + try: + if self.admin.lookupGroupByName(name): + os._exit(1) + + groupEnt = self.admin.initGroup(name) + + if gid >= 0: + groupEnt.set(libuser.GIDNUMBER, gid) + + self.admin.addGroup(groupEnt) + os._exit(0) + except Exception, e: + log.critical("Error when creating new group: %s" % str(e)) + os._exit(1) + + try: + (pid, status) = os.waitpid(childpid, 0) + except OSError as e: + log.critical("exception from waitpid while creating a group: %s %s" % (e.errno, e.strerror)) + return False + + if os.WIFEXITED(status) and (os.WEXITSTATUS(status) == 0): + return True + else: + return False + def createUser (self, name=None, password=None, isCrypted=False, groups=[], homedir=None, shell=None, uid=None, algo=None, lock=False, root="/mnt/sysimage"): -- 1.6.1.3

2 1

[PATCH 1/2] Initial FCoE support
by Hans de Goede 02 Jul '09

02 Jul '09

This patch adds support for using FCoE during the installation. This patch merely lays the initial ground work, there is more work todo: - The system will not boot without manual help after the install, as dracut / mkinitrd do not support FCoE yet - If FCoE is not used for / but for example for /srv, then information about the nic used for FCoE needs to be written in a to be defined config file in the system, and rc.sysinit needs to be thought to read this file and bring up FCoE SAN's / Fabrics not used for / - kickstart support for FCoE still needs to be done --- iw/autopart_type.py | 84 ++++++++++++++++++++++++++- storage/__init__.py | 5 ++ storage/fcoe.py | 92 +++++++++++++++++++++++++++++ textw/partition_text.py | 29 +++++++++- ui/adddrive.glade | 19 ++++++ ui/fcoe-config.glade | 149 +++++++++++++++++++++++++++++++++++++++++++++++ 6 files changed, 376 insertions(+), 2 deletions(-) create mode 100644 storage/fcoe.py create mode 100644 ui/fcoe-config.glade diff --git a/iw/autopart_type.py b/iw/autopart_type.py index 8a12689..9c04e52 100644 --- a/iw/autopart_type.py +++ b/iw/autopart_type.py @@ -32,6 +32,7 @@ from iw_gui import * from flags import flags import network from storage import iscsi +from storage import fcoe from storage.deviceaction import * import gettext @@ -325,6 +326,81 @@ class PartitionTypeWindow(InstallWindow): return rc + def addFcoeDrive(self): + (dxml, dialog) = gui.getGladeWidget("fcoe-config.glade", + "fcoeDialog") + + # Populate the combo + combo = dxml.get_widget("fcoeNicCombo") + cell = gtk.CellRendererText() + combo.pack_start(cell, True) + combo.set_attributes(cell, text = 0) + cell.set_property("wrap-width", 525) + combo.set_size_request(480, -1) + store = gtk.TreeStore(gobject.TYPE_STRING, gobject.TYPE_STRING) + combo.set_model(store) + + netdevs = self.anaconda.id.network.available() + keys = netdevs.keys() + keys.sort() + selected_interface = None + for dev in keys: + # Skip NIC's which are connected (iow in use for a net install) + if dev in network.getActiveNetDevs(): + continue + + i = store.append(None) + + desc = netdevs[dev].get("DESC") + if desc: + desc = "%s - %s" %(dev, desc) + else: + desc = "%s" %(dev,) + + mac = netdevs[dev].get("HWADDR") + if mac: + desc = "%s - %s" %(desc, mac) + + if selected_interface is None: + selected_interface = i + + store[i] = (desc, dev) + + if selected_interface: + combo.set_active_iter(selected_interface) + else: + combo.set_active(0) + + # Show the dialog + gui.addFrame(dialog) + dialog.show_all() + sg = gtk.SizeGroup(gtk.SIZE_GROUP_HORIZONTAL) + sg.add_widget(dxml.get_widget("fcoeNicCombo")) + + while 1: + rc = dialog.run() + + if rc == gtk.RESPONSE_CANCEL: + break; + + iter = combo.get_active_iter() + if iter is None: + self.intf.messageWindow(_("Error"), + "Must select a NIC to use.", + type="warning", custom_icon="error") + continue; + + try: + self.storage.fcoe.addSan(store.get_value(iter, 1), self.intf) + except IOError, e: + self.intf.messageWindow(_("Error"), str(e)) + rc = gtk.RESPONSE_CANCEL + + break + + dialog.destroy() + return rc + def addZfcpDrive(self): (dxml, dialog) = gui.getGladeWidget("zfcp-config.glade", "zfcpDialog") @@ -366,6 +442,10 @@ class PartitionTypeWindow(InstallWindow): dxml.get_widget("iscsiRadio").set_sensitive(False) dxml.get_widget("iscsiRadio").set_active(False) + if not fcoe.has_fcoe(): + dxml.get_widget("fcoeRadio").set_sensitive(False) + dxml.get_widget("fcoeRadio").set_active(False) + #figure out what advanced devices we have available and set sensible default group = dxml.get_widget("iscsiRadio").get_group() for button in group: @@ -379,6 +459,8 @@ class PartitionTypeWindow(InstallWindow): return if dxml.get_widget("iscsiRadio").get_active() and iscsi.has_iscsi(): rc = self.addIscsiDrive() + elif dxml.get_widget("fcoeRadio").get_active() and fcoe.has_fcoe(): + rc = self.addFcoeDrive() elif dxml.get_widget("zfcpRadio") is not None and dxml.get_widget("zfcpRadio").get_active(): rc = self.addZfcpDrive() dialog.destroy() @@ -498,7 +580,7 @@ class PartitionTypeWindow(InstallWindow): self.xml.get_widget("bootDriveCombo").set_sensitive(False) self.xml.get_widget("encryptButton").set_sensitive(False) - if not iutil.isS390() and not iscsi.has_iscsi(): + if not iutil.isS390() and not iscsi.has_iscsi() and not fcoe.has_fcoe(): self.xml.get_widget("addButton").set_sensitive(False) sigs = { "on_partitionTypeCombo_changed": self.comboChanged, diff --git a/storage/__init__.py b/storage/__init__.py index 623eff1..6fdb086 100644 --- a/storage/__init__.py +++ b/storage/__init__.py @@ -46,6 +46,7 @@ from devicelibs.lvm import safeLvmName from devicelibs.dm import name_from_dm_node from udev import * import iscsi +import fcoe import zfcp import gettext @@ -204,6 +205,7 @@ class Storage(object): self.__luksDevs = {} self.iscsi = iscsi.iscsi() + self.fcoe = fcoe.fcoe() self.zfcp = zfcp.ZFCP() self._nextID = 0 @@ -269,6 +271,7 @@ class Storage(object): w = self.anaconda.intf.waitWindow(_("Finding Devices"), _("Finding storage devices...")) self.iscsi.startup(self.anaconda.intf) + self.fcoe.startup(self.anaconda.intf) self.zfcp.startup() if self.anaconda.id.getUpgrade(): clearPartType = CLEARPART_TYPE_NONE @@ -911,6 +914,7 @@ class Storage(object): def write(self, instPath): self.fsset.write(instPath) self.iscsi.write(instPath, self.anaconda) + self.fcoe.write(instPath, self.anaconda) self.zfcp.write(instPath) def writeKS(self, f): @@ -978,6 +982,7 @@ class Storage(object): f.write("\n") self.iscsi.writeKS(f) + self.fcoe.writeKS(f) self.zfcp.writeKS(f) diff --git a/storage/fcoe.py b/storage/fcoe.py new file mode 100644 index 0000000..46c5bcb --- /dev/null +++ b/storage/fcoe.py @@ -0,0 +1,92 @@ +# +# fcoe.py - fcoe class +# +# Copyright (C) 2009 Red Hat, Inc. All rights reserved. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License as published by +# the Free Software Foundation; either version 2 of the License, or +# (at your option) any later version. +# +# This program is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +# GNU General Public License for more details. +# +# You should have received a copy of the GNU General Public License +# along with this program. If not, see <http://www.gnu.org/licenses/>. +# + +import os +import iutil +import logging +import time +log = logging.getLogger("anaconda") + +import gettext +_ = lambda x: gettext.ldgettext("anaconda", x) + +_fcoe_module_loaded = False + +def has_fcoe(): + global _fcoe_module_loaded + if not _fcoe_module_loaded: + iutil.execWithRedirect("modprobe", [ "fcoe" ], + stdout = "/dev/tty5", stderr="/dev/tty5", + searchPath = 1) + _fcoe_module_loaded = True + + return os.access("/sys/module/fcoe", os.X_OK) + +class fcoe(object): + def __init__(self): + self.started = False + self.nics = [] + + def _stabilize(self, intf = None): + if intf: + w = intf.waitWindow(_("Connecting to FCoE SAN"), + _("Connecting to FCoE SAN")) + + # I have no clue how long we need to wait, this ought to do the trick + time.sleep(10) + iutil.execWithRedirect("udevadm", [ "settle" ], + stdout = "/dev/tty5", stderr="/dev/tty5", + searchPath = 1) + if intf: + w.pop() + + def startup(self, intf = None): + if self.started: + return + + if not has_fcoe(): + return + + # Place holder for adding autodetection of FCoE setups based on + # firmware tables (like iBFT for iSCSI) + + self.started = True + + def addSan(self, nic, intf=None): + if not has_fcoe(): + raise IOError, _("FCoE not available") + + log.info("Activating FCoE SAN attached to %s" % nic) + + f = open("/sys/module/fcoe/parameters/create", "w") + f.write(nic) + f.close() + + self._stabilize(intf) + self.nics.append(nic) + + def writeKS(self, f): + # fixme plenty (including add ks support for fcoe in general) + return + + def write(self, instPath, anaconda): + # erm, do we need todo anything here ? + return + +# vim:tw=78:ts=4:et:sw=4 diff --git a/textw/partition_text.py b/textw/partition_text.py index fdf3496..f14a273 100644 --- a/textw/partition_text.py +++ b/textw/partition_text.py @@ -162,6 +162,9 @@ class PartitionTypeWindow: newdrv.append("Add iSCSI target") if iutil.isS390(): newdrv.append( "Add zFCP LUN" ) + from storage import fcoe + if fcoe.has_fcoe(): + newdrv.append("Add FCoE SAN") if len(newdrv) == 0: return INSTALL_BACK @@ -176,12 +179,18 @@ class PartitionTypeWindow: if button == TEXT_BACK_CHECK: return INSTALL_BACK - if choice == 1: + if newdrv[choice] == "Add zFCP LUN": try: return self.addZFCPDriveDialog(screen) except ValueError, e: ButtonChoiceWindow(screen, _("Error"), str(e)) return INSTALL_BACK + elif newdrv[choice] == "Add FCoE SAN": + try: + return self.addFcoeDriveDialog(screen) + except ValueError, e: + ButtonChoiceWindow(screen, _("Error"), str(e)) + return INSTALL_BACK else: try: return self.addIscsiDriveDialog(screen) @@ -206,6 +215,24 @@ class PartitionTypeWindow: return INSTALL_OK + def addFcoeDriveDialog(self, screen): + (button, entries) = EntryWindow(screen, + _("Add FCoE SAN"), + _("Enter the device name for the NIC which is connected to the FCoE SAN. For example \"eth0\"."), + prompts = [ _("NIC device name") ] ) + if button == TEXT_CANCEL_CHECK: + return INSTALL_BACK + + nic = entries[0].strip() + if nic not in self.anaconda.id.network.available(): + ButtonChoiceWindow(screen, _("Error"), + _("%s is not a valid NIC device name.") % nic) + return INSTALL_BACK + + self.anaconda.id.storage.fcoe.addSan(nic) + + return INSTALL_OK + def addIscsiDriveDialog(self, screen): if not network.hasActiveNetDev(): ButtonChoiceWindow(screen, _("Error"), diff --git a/ui/adddrive.glade b/ui/adddrive.glade index 599ed79..2d991ef 100644 --- a/ui/adddrive.glade +++ b/ui/adddrive.glade @@ -202,6 +202,25 @@ </child> <child> + <widget class="GtkRadioButton" id="fcoeRadio"> + <property name="visible">True</property> + <property name="label" translatable="yes" context="yes">Add _FCoE SAN</property> + <property name="use_underline">True</property> + <property name="relief">GTK_RELIEF_NORMAL</property> + <property name="focus_on_click">True</property> + <property name="active">False</property> + <property name="inconsistent">False</property> + <property name="draw_indicator">True</property> + <property name="group">iscsiRadio</property> + </widget> + <packing> + <property name="padding">0</property> + <property name="expand">True</property> + <property name="fill">True</property> + </packing> + </child> + + <child> <widget class="GtkRadioButton" id="radiobutton1"> <property name="visible">True</property> <property name="sensitive">False</property> diff --git a/ui/fcoe-config.glade b/ui/fcoe-config.glade new file mode 100644 index 0000000..7034064 --- /dev/null +++ b/ui/fcoe-config.glade @@ -0,0 +1,149 @@ +<?xml version="1.0"?> +<glade-interface> +  +  + <widget class="GtkDialog" id="fcoeDialog"> + <property name="title" translatable="yes">Configure FCoE Parameters</property> + <property name="window_position">center</property> + <property name="type_hint">dialog</property> + <child internal-child="vbox"> + <widget class="GtkVBox" id="dialog-vbox1"> + <property name="visible">True</property> + <child> + <widget class="GtkVBox" id="vbox1"> + <property name="visible">True</property> + <property name="border_width">12</property> + <property name="spacing">12</property> + <child> + <widget class="GtkLabel" id="label3"> + <property name="visible">True</property> + <property name="label" translatable="yes">Please select the network interface which is connected to +your FCoE switch.</property> + <property name="wrap">True</property> + </widget> + <packing> + <property name="expand">False</property> + <property name="fill">False</property> + <property name="position">0</property> + </packing> + </child> + <child> + <placeholder/> + </child> + <child> + <widget class="GtkTable" id="fcoeTable"> + <property name="visible">True</property> + <property name="border_width">12</property> + <property name="n_columns">2</property> + <property name="column_spacing">6</property> + <property name="row_spacing">6</property> + <child> + <widget class="GtkLabel" id="label1"> + <property name="visible">True</property> + <property name="label" translatable="yes">NIC</property> + </widget> + </child> + <child> + <widget class="GtkComboBox" id="fcoeNicCombo"> + <property name="visible">True</property> + </widget> + <packing> + <property name="left_attach">1</property> + <property name="right_attach">2</property> + </packing> + </child> + </widget> + <packing> + <property name="position">2</property> + </packing> + </child> + <child> + <placeholder/> + </child> + </widget> + <packing> + <property name="position">2</property> + </packing> + </child> + <child internal-child="action_area"> + <widget class="GtkHButtonBox" id="dialog-action_area1"> + <property name="visible">True</property> + <property name="layout_style">end</property> + <child> + <widget class="GtkButton" id="button1"> + <property name="label">gtk-cancel</property> + <property name="response_id">-6</property> + <property name="visible">True</property> + <property name="can_focus">True</property> + <property name="can_default">True</property> + <property name="receives_default">False</property> + <property name="use_stock">True</property> + </widget> + <packing> + <property name="expand">False</property> + <property name="fill">False</property> + <property name="position">0</property> + </packing> + </child> + <child> + <widget class="GtkButton" id="button2"> + <property name="response_id">-10</property> + <property name="visible">True</property> + <property name="can_focus">True</property> + <property name="can_default">True</property> + <property name="receives_default">False</property> + <child> + <widget class="GtkAlignment" id="alignment1"> + <property name="visible">True</property> + <property name="xscale">0</property> + <property name="yscale">0</property> + <child> + <widget class="GtkHBox" id="hbox1"> + <property name="visible">True</property> + <property name="spacing">2</property> + <child> + <widget class="GtkImage" id="image1"> + <property name="visible">True</property> + <property name="stock">gtk-add</property> + <property name="icon-size">4</property> + </widget> + <packing> + <property name="expand">False</property> + <property name="fill">False</property> + <property name="position">0</property> + </packing> + </child> + <child> + <widget class="GtkLabel" id="label6"> + <property name="visible">True</property> + <property name="label" translatable="yes">_Add FCoE Disk(s)</property> + <property name="use_underline">True</property> + </widget> + <packing> + <property name="expand">False</property> + <property name="fill">False</property> + <property name="position">1</property> + </packing> + </child> + </widget> + </child> + </widget> + </child> + </widget> + <packing> + <property name="expand">False</property> + <property name="fill">False</property> + <property name="position">1</property> + </packing> + </child> + </widget> + <packing> + <property name="expand">False</property> + <property name="pack_type">end</property> + <property name="position">0</property> + </packing> + </child> + </widget> + </child> + </widget> +</glade-interface> -- 1.6.2.2

2 5

[PATCH 1/3] Add FCoE support to storage/udev.py
by Hans de Goede 02 Jul '09

02 Jul '09

Add the ability to identify if a disk is an FCoE device and to get the identifier of the disk and the name of the NIC used to connect to an FCoE disk --- storage/udev.py | 27 +++++++++++++++++++++++++++ 1 files changed, 27 insertions(+), 0 deletions(-) diff --git a/storage/udev.py b/storage/udev.py index af3ee12..d057cf2 100644 --- a/storage/udev.py +++ b/storage/udev.py @@ -386,3 +386,30 @@ def udev_device_get_iscsi_port(info): path_components = info["ID_PATH"].split("-") return path_components[1].split(":")[1] + +# fcoe disks have ID_PATH in the form of: +# pci-eth#-fc-${id} +# fcoe parts look like this: +# pci-eth#-fc-${id}-part# +def udev_device_is_fcoe(info): + try: + path_components = info["ID_PATH"].split("-") + + if info["ID_BUS"] == "scsi" and len(path_components) >= 4 and \ + path_components[0] == "pci" and path_components[2] == "fc" and \ + path_components[1][0:3] == "eth": + return True + except KeyError: + pass + + return False + +def udev_device_get_fcoe_nic(info): + path_components = info["ID_PATH"].split("-") + + return path_components[1] + +def udev_device_get_fcoe_identifier(info): + path_components = info["ID_PATH"].split("-") + + return path_components[3] -- 1.6.2.2

2 3

[PATCH] Update /etc/hosts with hostname for loopback IP address (#506384)
by Radek Vykydal 02 Jul '09

02 Jul '09

Note: Compared to writing out of our own /etc/hosts which was removed in bug 491808, we do not write line for localhosthost IP address. --- network.py | 23 +++++++++++++++++++++++ 1 files changed, 23 insertions(+), 0 deletions(-) diff --git a/network.py b/network.py index 5c08e18..1c06fb6 100644 --- a/network.py +++ b/network.py @@ -670,6 +670,29 @@ class Network: if domainname: self.domains = [domainname] + # /etc/hosts + # update lines for 127.0.0.1 and ::1 with hostname if the file exists + if instPath and os.path.isfile(instPath + "/etc/hosts"): + of = open(instPath + "/etc/hosts", "r") + updatedlines = [] + update = False + for line in of: + line = line.strip() + if line.startswith('127.0.0.1') or line.startswith('::1'): + if self.hostname not in line.split(): + line += " %s" % self.hostname + update = True + updatedlines.append(line) + of.close() + + if update: + nf = open(instPath + "/etc/hosts", "w") + upd_comment = "# hostname %s added to /etc/hosts by anaconda\n" % self.hostname + nf.write(upd_comment + '\n'.join(updatedlines) + '\n') + nf.close() + log.info("/etc/hosts updated with hostname %s" % self.hostname) + + # /etc/resolv.conf if (not instPath) or (not os.path.isfile(instPath + '/etc/resolv.conf')) or flags.livecdInstall: if os.path.isfile('/etc/resolv.conf') and instPath != '': -- 1.6.0.6

2 1

RFC: writing kernel cmdline options to grub.conf for dracut
by Hans de Goede 02 Jul '09

02 Jul '09

Hi, This morning I've been talking to Harald Hoyer about what sort of commandline options dracut will be needing to find the / filesystem beside root=UUID=1234567890 . In most cases (normal disks, dmraid, mdraid, lvm, dmcrypt) root=UUID=1234567890 should suffice. However in certain cases for example dracut will need additional info to find the disks. We've come to the following plan for iscsi targets: 1) Extend the dhcp_root dhcp variable iscsi syntax to be able include a username password, so: iscsi:192.168.50.2::::iqn.2009-06.dracut:target66 Can become: iscsi:user:pass@192.168.50.2::::iqn.2009-06.dracut:target66 Or: iscsi:user:pass:reverse_user:reverse_pass@192.168.50.2::::iqn.2009-06.dracut:target66 2) Pass root-path=iscsi:... on the kernel cmdline, for each needed iscsi target, so if necessary this will be passed multiple times, dracut will be modified to be able handle multiple root-path arguments being passed in 3) chmod /proc/cmdline 400, so that it cannot be read by ordinary users, plugging the passwork leak problem Now the remaining question is how to implement the adding of the needed cmdline options to grub.conf. For the iscsi case (and atleast fcoe will be similar) I see the following 2 options, both of which are feasible: 1) Add code to iscsi.py to get the needed cmdline options 2) Add a dracutCmdlineOptions property to all Device classes, which will return the needed options (if any) to make the device in question available inside dracut. One of the issues here is that onlining an iscsi target, may get us multiple disks as the target may have multiple LUN's. If more then one LUN of the same target is needed we might end up with having the same root-path=iscsi:... option twice on the cmdline, filtering this will be is easy though so this is not really an issue. Looking at this from the iscsi pov, doing this from iscsi.py is easier, as that already has a list of targets instead of a list of disks (so LUN's) as the rest of the storage code has. Looking at this from the pov of the rest of the code it might be cleaner in the long run to make this a device property though. So I'm not sure what to do, advice is / ideas are very welcome ? Regards, Hans

3 3

RHEL-5 patch to fix #448006
by David Cantrell 02 Jul '09

02 Jul '09

Regression introduced by me for the fix for #432011, this patch makes sure we grab the bootfile from the DHCP response. Tested locally using a RHEL 5.3 tree. The 'ks' argument works with this patch in place. -- David Cantrell <dcantrell(a)redhat.com> Red Hat, Inc. / Honolulu, HI

2 4

[PATCH] Rework tracking of devices containing install media (#497087)
by David Lehman 02 Jul '09

02 Jul '09

This set of patches changes the way we track devices containing installation media. It is a two-stage approach: First, we save the device specifications exactly as provided to us. Second, when populating the device tree, we set the protected attribute of devices matching the provided specs. This frees us from worrying about devices being renamed since the attribute indicating that a device is protected is part of the device instance itself.

3 9

[PATCH - 497334] change the basic assumption about kickstart network
by Chris Lumens 01 Jul '09

01 Jul '09

These two patches change one of the basic operating assumptions about the network command in kickstart. Before: No "network" command implies dhcp Now: No "network" command implies stop and ask in stage1 This brings the network command in line with how everything else in kickstart, and also gets rid of the obnoxious and short lived --bootproto=query, though it will still have to be recognized for compatibility reasons. - Chris

2 3

[PATCH] Set focus on hostname entry in network UI screen (#494135)
by Radek Vykydal 01 Jul '09

01 Jul '09

--- iw/network_gui.py | 5 ++++- 1 files changed, 4 insertions(+), 1 deletions(-) diff --git a/iw/network_gui.py b/iw/network_gui.py index 9f5ced7..ce2c879 100644 --- a/iw/network_gui.py +++ b/iw/network_gui.py @@ -44,13 +44,16 @@ class NetworkWindow(InstallWindow): self.hostnameEntry.set_text(self.hostname) self.hostnameEntry.connect("activate", lambda w: self.ics.setGrabNext(1)) - self.hostnameEntry.grab_focus() + self.hostnameEntry.connect("map-event", self.setFocus) # load the icon gui.readImageFromFile("network.png", image=self.icon) return self.align + def setFocus(self, area, data): + self.hostnameEntry.grab_focus() + def hostnameError(self): self.hostnameEntry.grab_focus() raise gui.StayOnScreen -- 1.6.0.6

2 1

[PATCH] two patches for rawhide, the second depends on the first
by Radek Vykydal 01 Jul '09

01 Jul '09

The 2/2 patch is "TAKE III", this one depends on the 1/2 patch.

1 5

RFC: proposal for handling isw dmraid -> mdraid upgrade path
by Hans de Goede 01 Jul '09

01 Jul '09

Hi, I've been thinking about $SUBJECT lately, and I would like to propose the following way to handle this: 1) Make the use of mdraid or dmraid for BIOS-raid supported by both configurable both in anaconda (it already is) and in the installed OS (so in relevant udev rules and / or in rc.sysinit). anaconda will have a cmdline option to override the defaults, and the installed system will have a /etc/sysconfig/biosraid file, with a single setting: PREFERRED_BIOSRAID_HANDLER Which can be either dmraid or mdraid 2) On a fresh install anaconda will use mdraid and will write "PREFERRED_BIOSRAID_HANDLER=mdraid" to /etc/sysconfig/biosraid (unless forced to use dmraid through the command line) 3) On an update anaconda will look for /etc/sysconfig/biosraid, and use what ever is there, if the file does not exist (as it will not an an upgrade from F-11), it will use dmraid as we did in previous releases and write "PREFERRED_BIOSRAID_HANDLER=dmraid" to /etc/sysconfig/biosraid, so that future upgrades will still use dmraid. This will result in an it just works experience for users, and offers the ability to keep using dmraid for BIOS-raid supported by path, which is something several people have indicated they want. The only hole I can poke into this is yum update's from F-11, but we can document what to do on the yum upgrade wiki page, which people doing yum upgrade's should always read. Regards, Hans

3 7

Features/YumLangpackPlugin and anaconda consequences
by Bill Nottingham 01 Jul '09

01 Jul '09

Crrently, in anaconda and yum, we install language specific addons via conditional packages in comps. It's generally agreed that this sucks as an intereface, but it's what we have. There is a proposal to move this to a yum plugin for Fedora 12. Whenever a package has language-specific extensions (such as 'openoffice.org-core'; these packages would be described in some metadata), the corresponding language pack (determined via %_install_langs, most likely) would be installed. However, this has two conseqences for anaconda. 1) Plugins Anaconda doesn't currently enable them. We would want this to work in some way in anaconda, which would require enabling at least this plugin. How possible is this? 2) The actual installation experience Right now, if you pick an install language, that language group is automatically enabled. Additional languages are offered via their comps groups through customized package selection. However, that has two issues: - We never show these groups if there are only conditional packages in it, at least according to various bugs I've seen filed (oops) - If we're doing this via a plugin only, how can we better set our install languages if these comps groups don't exist? For the latter question, we may want to run some ideas past design. Bill

4 4

More on the GUI (post FUDCon)
by Joel Granados 01 Jul '09

01 Jul '09

Hey list. Met with Mo at FUDCon and got lots of great ideas. This mail is intended to share some of these. 1. First of all she gave me a cool reference that can be used for general Human Interface design purposes. http://library.gnome.org/devel/hig-book/stable/. This is the most recent one I found, I believe its from 2008, if there is a more recent one, pls scream. 2. Help information. This refers to text that explains stuff that is not be trivial to understand. Instead of doing a help mechanism we could create a small explanation label, where needed, that goes under the concept that needs explaining. Much like http://library.gnome.org/misc/release-notes/2.26/figures/rnusers.brasero.pn… or http://library.gnome.org/misc/release-notes/2.26/figures/rnusers.epiphany.p… where the buttons have the name and the explanation in some other size and are slightly clear. 3. Have simplified device names. To have "sda" instead of "/dev/sda" and we would place the whole path "/dev/sda" as a small comment on the bottom of the device name. I'm guessing this would ease the user experience with dmraid type devices. 4. Use verbs for the actions. We have already been discussing this in the list. 5. Use Gig for the sizes. I had missed this in previous posts. But it be a good idea to have the size in Gb after 999 Mb. We can have a function that returns the show-able string. 6. Merge tree and bar view (This is my personal favorite, And I'll probably include it in a glade file in future posts. :). To have the bar information on the tree view. It would be on the far right and it would just have minimal information as most of it is contained in the tree view. I have to give Mo all the credit for this one :) 7. Put the type of installation (where we select "use entire drive", "Replace existing Linux system"...) in a separate window with additional comments on the options. This has two advantages, IMO, 1. We give the user more info so he/she can make an informed decision. 2. We don't show the encrypted check box when it is not needed (less confusion for the user). I still don't know exactly where to put the advanced configuration button in this case though. 8. Giving the user the possibility to select more than one device to shrink by using a checkbox instead of a drop down menu. Don't really know how much of a good idea this is, since I don't really know how that is coded, but its worth considering IMO. 9. The location of the action buttons would be better on the bottom left. This goes better with the look and feel of other gnome applications. So we would have the tree view, then the action buttons and then the "continue" & "Back" buttons. 10. We should pop up a window with all the "create" possibilities [1] for the customize partition "create" button. They would be handled with a comboboxe and would have a little explanation beneath the choice type. The user would select what he/she wants and a screan that creates the device would pop up. Mo: If I am forgetting something, pls chime in. Finally, considering that this will not be a complete re-write of the GUI for the storage section. I will probably start cleaning up the GUI files to get a better modularization and to easily move things around for the changes that we are planning. As always, suggestions are very much appreciated. [1] -> "LVM - physical volume", "LVM - Logical Volume" .... -- Joel Andres Granados Brno, Czech Republic, Red Hat.

4 6

[PATCH] Two unrelated patches for anaconda master
by David Lehman 30 Jun '09

30 Jun '09

2 3

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Anaconda-devel July 2009