Anaconda-devel June 2008

anaconda-devel@lists.stg.fedoraproject.org

18 participants
18 discussions

[PATCH] removal of libdhcp
by David Cantrell 21 Dec '08

21 Dec '08

No patch attached because it's quite large. I apologize for the length, but considering this replaces a library and introduces a new struct it meant a new file and changing calls everywhere. http://dcantrel.fedorapeople.org/anaconda/iface/ iface.h - Header file explaining the iface API. iface_t structure replacing pumpNetIntf and networkDeviceConfig, and all of the functions. iface.patch - The changes, it's about 90% complete at this point. *.txt - Explanations and notes. This patch removes anaconda's use of libdhcp, which also means removing libdhcp4client and libdhcp6client. For static network configuration, I am using libnl. To gather current network interface information, I am using libnl. To control IPv6 autoconf, I read/ write /proc since that's all we can do right now. For DHCP and DHCPv6, I run dhclient and dhcp6c, respectively. I'd like everyone to have a look at iface.h and the patch file. It's not complete yet, so I probably know about the obvious things (the FIXMEs and the useless debugging printfs and the incomplete isys.py code and so on). The goal with iface.patch is to get us closer to using NetworkManager in stage 1 and stage 2. I envision the NM changeover to be just as large, but hopefully by then we will have decided to nuke certain parts of loader entirely. Completely eliminating libdhcp and friends will be a nice step. There are some other things that need discussion as well. mkinitrd currently uses libdhcp, for example. This had me thinking we could generate libisys in an anaconda-devel package for things like mkinitrd, but I'm just brainstorming. Maybe I should put the iface code elsewhere? I dunno. At any rate, mkinitrd needs to be not left out. Work in progress, comments welcome. -- David Cantrell <dcantrell(a)redhat.com> Red Hat / Honolulu, HI

8 28

fsprofile
by Curtis Doty 30 Jun '08

30 Jun '08

Is there any trick to tuning ext3 filesystems? In kickstart, I've tested a simple --fsprofile=largefile for both lv and md filesystems to no avail. It doesn't even appear to extend the args in formatDevice(). ../C

1 0

[PATCH] Fix a bug in partition_dialog_gui.py
by Tetsuya MORIMOTO 23 Jun '08

23 Jun '08

Hi, I found a keyboard binding bug. At "Add Partition" window, it's "Additional Size Options" table(frame). -- Additional Size Options ------- | ( ) Fixed size | ( ) Fill all space up to (MB): [ gtk.SpinButton ] | ( ) Fill to maximum allowable size --- When we operate that table with keyboard(Up/Down arrow key), The focus order is as follows. > 1. Fixed size > 3. Fill all space up to (MB): [ gtk.SpinButton ] > 2. Fill to maximum allowable size It's strange for users intuitively. This bug is creating gtk.HBox() container for second radio and spin buttons. It seems the GTK recognize another layer(or focus order) to pack radio and spin buttons. If its purpose is just for layout, we can specify the layout which is attached for gtk.Table(). # refer to attachment patch # I confirmed my patch works the intended operation for Fedora9 with anaconda-11.4.0.82-1 Could you review my patch, please ? and I'd appreciate it if you could give me the comment. thanks, Tetsuya MORIMOTO

1 0

RE: anaconda installing packages out of order (RHEL5.2)
by Moray Henderson (ICT) 19 Jun '08

19 Jun '08

>From: Brian Reichert <reichert(a)numachi.com> >Our ISO boots, and installs our RPMs, but out of order, and does >not preserve the dependancies. (_Mostly_ does, but some critical >misses.) I've been running into that problem too. I think Anaconda re-orders the packages as it works out dependencies, so (as far as I can tell) you can't rely on pkgorder or comps.xml to give the final install order. (If I'm wrong about that, I would be very grateful if someone would tell me how to do it.) The key to the solution is the difference between the rpm directives "Requires" and "PreReq" in the .spec files. "Requires: x" means "this package requires the x package in order to _run_". If both packages appear in the same rpm transaction, the transaction will succeed _but x may not be installed first_. "PreReq: x" means "this package requires x to be present _before it will install properly_". In this case, if both packages appear in the same transaction, rpm will ensure that x is installed first. Moray. "To err is human. To purr, feline" http://members.aol.com/edgwddirk

2 1

anaconda installing packages out of order (RHEL5.2)
by Brian Reichert 18 Jun '08

18 Jun '08

I'm tasked with building a custom kickstart ISO for Red Hat 5 update 2. This is using anaconda-11.1.2.113-1 anaconda-runtime-11.1.2.113-1 I've been successful for months, but today, ran into a weird symptom. Our ISO boots, and installs our RPMs, but out of order, and does not preserve the dependancies. (_Mostly_ does, but some critical misses.) During the creation of the ISO, I direct the build to: - copy our RPMs into the approriate directory for mkisofs - generate a test RPM database, and from that, generate a list of installed packages rpm --dbpath ${pkg_db_dir} -qa --queryformat='%{NAME}\n' >> all_packages.txt - And I supply that text file in my ks.cfg file %include /mnt/source/all_packages.txt - I create the on-disk repo: createrepo -p -q ${RPMBASE} I observe: - comparing the messages in /var/log/anaconda.log makes it seem that the packages 'seen' by anaconda are in the right order. That is to say, this command: awk '/Adding/ {print $6}' /var/log/anaconda.log yeilds a list that is in the same order as all_packages.txt from the ISO. - but those lists are not the same order as anaconda installed: awk '/Installing/ {print $2}' /root/install.log Some of those are out-of-order, and breakage occurs. :/ Does anyone have any suggestions as to how to debug this? I can provide logs (or snippets thereof), if that would help in the diagnosis... -- Brian Reichert <reichert(a)numachi.com> 55 Crystal Ave. #286 Daytime number: (603) 434-6842 Derry NH 03038-1725 USA BSD admin/developer at large

1 0

die, methods, die
by Chris Lumens 18 Jun '08

18 Jun '08

Well here it is, a giant pile of patches that completely reworks all the installation repo selection and stage2 finding. Some of these patches are probably a little too small, and some are probably a little too large. But this is a huge body of work, so that's just how it goes. Current status: - The entire purpose of method selection in loader now is trying to find the stage2 image. If the stage2 image can be found on an attached CD, that's what will get used. - You can specify where the installation repo is located with repo= or method=. The intention is that method= is deprecated and will go away in the future, but we need to preserve it at least for now. - stage2= still exists and will continue to exist. preupgrade probably doesn't work, though. If you give a stage2=, that takes priority over any image that might be found on the CD. - If you don't give a method= or repo=, we first look for any CD/DVDs. If one is found, we look for packages. If those are found, we use them. If not, we default to using the mirrorlist provided by the /etc/yum.repos.d files. This means that for the most common installation method, you'll just toss your boot.iso in the drive, run straight into stage2, and be up using the mirrorlist. Hooray! - Since very few people will be getting the method selection screen in loader, most people won't have networking up when we get to stage2. That means in order to use the mirrorlist, we have to use the NetworkConfigurator. This isn't quite ideal, but David's going to have a giant networking patch one of these days so I'm not really going to bother. Current problems: - Media check is commented out. My intention here is to move this to stage2 if possible. - yum logs to tty1. - HDISO and NFSISO are completely untested and probably don't work, but I'll make them work later. - We don't write out a method to /root/anaconda-ks.cfg. - Error handling in AnacondaYum.__init__ is rather poor, but I need a better repo editor to fix that, and that's on a whole different branch I need to return to. Anyway this isn't 100% working but it seems good enough for me to consider committing. I've tested boot.iso and pxebooting with HTTP, NFS, and a DVD and those all worked decently. The big problem for me right now is the error handling. But that can be beefed up once this stuff goes in and I have an easier time testing. Questions? - Chris

3 25

Debugging kickstart
by Johnson, Brian 18 Jun '08

18 Jun '08

I have a kickstart script that apparently loads and it goes to the "Setting up kickstart" part but then just goes to the language selection screen, there is obviously something wrong with my kickstart. How do I get more debug information about where anaconda is failing Brian Johnson Manager Online Operations EAMythic bjjohnson(a)ea.com 703-877-8499

3 2

[PATCH] check values of kickstart option timezone
by Radek Vykydal 18 Jun '08

18 Jun '08

Hi, this patch should solve unhandled exception caused by invalid value of kickstart option timezone (bug #404321). Why this way: The timezone value read from ks can be checked (by checking the presence of respective timezone file in usr/share/zoneinfo): 1) when reading ks 2) in setuptime step (where it actually is checked now) 3) in writeconfig step (where it raises IOError which is subj of this bug) 4) in firstboot date module I think the cleanest way is to check it in 1), because then, in case of invalid value, you can ask the user for a valid one in timezone step. This is what the patch of kickstart.py does. Other possibility is to defer the check to 3) with an advantage of wider choice of installed timezone files (/mnt/sysimage/usr/share/zoneinfo). Then, in case of invalid value, there would be 2 options: - some default ('America/New_York') value would be set in /etc/sysconfig/clock and /etc/localtime (otherwise the exception would be raised by system-config-date during firstboot date configuration) - or the invalid value from ks would be saved into /etc/sysconfig/clock, /etc/localtime would not be copied, and firstboot module date would check these values and in case of being invalid display also timezone tab in firstboot date selection window. The patch adds also checking in 3) (timezone.py patch), but only to defer the raise of exception (which should never happen with checking in 1) and asking) to firstboot phase where it can eventually be 'solved' with reconfig. Well, I am not sure if this second patch is necessary/justifiable. If the ability to set the timezone to some value available only in installed packages (ignore invalid value in 1), check in 3)) was really desired, I'd add an option to ks for it? Radek Vykydal Patch against rhel-5 branch: diff --git a/kickstart.py b/kickstart.py index 3c18853..4340d72 100644 --- a/kickstart.py +++ b/kickstart.py @@ -569,6 +569,10 @@ class AnacondaKSHandlers(KickstartHandlers): def doTimezone(self, args): KickstartHandlers.doTimezone(self, args) dict = self.ksdata.timezone + tzfile = "/usr/share/zoneinfo" + dict["timezone"] + if not os.access(tzfile, os.R_OK): + log.warning("Can't read timezone file set in kickstart, will ask") + return self.id.instClass.setTimezoneInfo(self.id, dict["timezone"], dict["isUtc"]) self.skipSteps.append("timezone") diff --git a/timezone.py b/timezone.py index d63fcfb..7bab9f9 100644 --- a/timezone.py +++ b/timezone.py @@ -13,6 +13,7 @@ import shutil import iutil +import os from flags import flags import logging @@ -37,10 +38,13 @@ class Timezone: fromFile = instPath + "/usr/share/zoneinfo/" + self.tz - try: - shutil.copyfile(fromFile, instPath + "/etc/localtime") - except OSError, (errno, msg): - log.error("Error copying timezone (from %s): %s" % (fromFile, msg)) + if not os.access(fromFile, os.R_OK): + log.error("Timezone to be copied (%s) doesn't exist" % fromFile) + else: + try: + shutil.copyfile(fromFile, instPath + "/etc/localtime") + except OSError, (errno, msg): + log.error("Error copying timezone (from %s): %s" % (fromFile, msg)) f = open(instPath + "/etc/sysconfig/clock", "w") Patch against rawhide: diff --git a/kickstart.py b/kickstart.py index 7b0ea3d..d71d251 100644 --- a/kickstart.py +++ b/kickstart.py @@ -705,6 +705,11 @@ class Timezone(commands.timezone.FC6_Timezone): def parse(self, args): commands.timezone.FC6_Timezone.parse(self, args) + tzfile = "/usr/share/zoneinfo" + self.timezone + if not os.access(tzfile, os.R_OK): + log.warning("Can't read timezone file set in kickstart, will ask") + return + self.handler.id.timezone.setTimezoneInfo(self.timezone, self.isUtc) self.handler.skipSteps.append("timezone") diff --git a/timezone.py b/timezone.py index 3960d81..1ed7491 100644 --- a/timezone.py +++ b/timezone.py @@ -19,6 +19,7 @@ import shutil import iutil +import os from flags import flags import logging @@ -43,10 +44,13 @@ class Timezone: fromFile = instPath + "/usr/share/zoneinfo/" + self.tz - try: - shutil.copyfile(fromFile, instPath + "/etc/localtime") - except OSError, (errno, msg): - log.error("Error copying timezone (from %s): %s" % (fromFile, msg)) + if not os.access(fromFile, os.R_OK): + log.error("Timezone to be copied (%s) doesn't exist" % fromFile) + else: + try: + shutil.copyfile(fromFile, instPath + "/etc/localtime") + except OSError, (errno, msg): + log.error("Error copying timezone (from %s): %s" % (fromFile, msg)) f = open(instPath + "/etc/sysconfig/clock", "w")

2 1

Customizing Text Mode Install
by niall el-assaad 17 Jun '08

17 Jun '08

Hi, I'm trying to build a customized text mode install image and want to change some of the text in the installer, such as the "Welcome to Fedora Core" that is displayed in the top left of the screen. I've looked at the code and can't see where I would change this, any ideas? thanks, niall

2 2

[PATCH] Add a mirrorlist option.
by Jesse Keating 16 Jun '08

16 Jun '08

This option allows us to pass a mirrorlist into buildinstall instead of or in addition to baseurls. This will allow the yum object to make use of a list of potential mirrors should it have difficulties downloading something. --- scripts/buildinstall | 17 +++++++++++++++++ 1 files changed, 17 insertions(+), 0 deletions(-) diff --git a/scripts/buildinstall b/scripts/buildinstall index 4266d83..829d33c 100755 --- a/scripts/buildinstall +++ b/scripts/buildinstall @@ -67,6 +67,10 @@ while [ $# -gt 0 ]; do UPDATES=$2 shift; shift ;; + --mirrorlist) + MIRRORLIST="$MIRRORLIST $2" + shift; shift + ;; *) if [ -z "$REPO" ]; then @@ -142,6 +146,19 @@ EOF let n++ done +n=1 +for l in $MIRRORLIST; do + cat >> $yumconf <<EOF + +[anaconda-mirrorlistrepo-$n] +name=anaconda mirrorlist repo $n +mirrorlist=$l +enabled=1 +EOF + let n++ +done + + BUILDINSTDIR=`mktemp -d ${TMPDIR:-/tmp}/buildinstall.tree.XXXXXX` TREEDIR=`mktemp -d ${TMPDIR:-/tmp}/treedir.XXXXXX` -- 1.5.5.2

2 1

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Anaconda-devel June 2008