Anaconda-devel February 2012

anaconda-devel@lists.stg.fedoraproject.org

28 participants
121 discussions

Ability to install to loopmounted filesystem - mentor wanted
by Tirtha Chatterjee 02 Dec '12

02 Dec '12

Hello. I am trying to implement something like Wubi on Fedora. For this, I first need to make it possible to install to loopmounted filesystems through anaconda. I don't think a lot of modification to the existing source code is required. I tried going through the source code, but it seemed quite big to manage without some guidance. I would be extremely happy if someone guided me on this. Hoping for an affirmative reply :-) -- Regards Tirtha Chatterjee KDE developer http://wyuka.co.cc/

2 3

[newui] TimezoneMap widget
by Vratislav Podzimek 26 Mar '12

26 Mar '12

These are two patches introducing the TimezoneMap widget. It is a slightly modified code from the gnome-control-center with three major changes: 1) I've renamed the widget (and its methods) to AnacondaTimezoneMap. 2) I've replaced the "location-changed" signal with "timezone-changed" signal because the old one contained whole TZLocation struct that would cause problems with introspection. "timezone-changed" contains only the selected timezone as gchar* which is easy to handle by introspection and is actually everything we need. 3) For the same reasons I've replaced the get_location method by get_timezone method. You can see simple Python script for testing and the behaviour of the TimezoneMap in this video: http://vpodzime.fedorapeople.org/timezone_map.ogg I need to say that I like the map from s-c-date more, but it can be a matter of opinion. I will try to write a code for glade, so that this widget could be used in it and then I will start working on the DateTime spoke. -- Vratislav Podzimek

2 8

Re: GPT and Fedora 17
by Adam Williamson 02 Mar '12

02 Mar '12

On Mon, 2012-02-06 at 23:19 +0000, Pádraig Brady wrote: > On 02/06/2012 10:40 PM, Brian C. Lane wrote: > > In Fedora 16 we changed to using GPT as the default disklabel for new > > installs. In a few cases, mostly limited to Lenovo hardware, we found > > that some BIOS's would not boot from GPT. We blacklisted Lenovo, falling > > back to msdos labels in order to solve this. > > > > Thanks to Matthew Garrett we found that switching on the boot flag of > > the GPT's protective MBR these BIOS's would then boot from GPT. Matthew > > wrote a patch for parted to allow controlling this flag using the > > disk_set pmbr_boot command in parted. This is in parted-3.0-7 > > > > In anaconda-17.6 I have reverted the Lenovo blacklist and changed things > > so that pmbr_boot is always set on GPT labeled installs. This should > > ensure that thing boot correctly. > > > > If this still causes problems the symptom will be that grub never starts > > and the bios may complain about not being able to find an OS. If you > > have problems with this please open a bug with the output from dmidecode > > > > You can still force usage of msdos partitions by passing nogpt on the > > kernel cmdline. > > Hmm, I tried that workaround I think on my Lenovo T520 with BIOS 1.29, > and it didn't help. I.E. point (1.) from the link referenced here: > https://bugzilla.redhat.com/show_bug.cgi?id=735733#c31 > > Fingers crossed I just missed something at the time. > I'll try out again tomorrow maybe. Yes, I remember that. Please do test the change out and let us know the results, we'd definitely like to know. Brian, this change takes effect from Alpha TC2 (whenever we spin it), it's not in Alpha TC1, correct? -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora http://www.happyassassin.net

3 3

Packaging & installation
by Bill Nottingham 02 Mar '12

02 Mar '12

Prompted by David Lehman's patch set, I wanted to discuss how we're building what we're installing in Fedora and RHEL, and how we might need to represent that both in the media and in the UI. What we've got now exposed at the lowest level (yum/kickstart) and supported as first-level objects throughout the system: 1. packages 2. groups (of packages) 3. repos (which contain group lists, and packages) What we additionally expose only in anaconda, and lose post-install: 4. tasks in install classes (lists of groups) What we expose in Fedora: 5. spins (live ISO via kickstart file) 6. 'the install DVD' (a package set via a kickstart file) What we expose in RHEL: 7. variants (essentially, a spin that's a package set instead of a live ISO) 8. add-ons (repos, with groups, that apply to specific variants) All of these use the comps file, which is a list of groups. Now, as we rework anaconda's UI (and backend, per dlehman's patches), we need to make sure not to break everything horribly. In Fedora, spins and the install DVD are driven off of kickstart files. This was done because it's the only thing pungi and livecd-creator supports, and because each spin/install DVD is a distinct object that can have its own kickstart. Anaconda doesn't need to worry about separately consuming them, as there's only one repo to ever pull from, and the spins are live images. In RHEL, variants and add-ons are driven off of a different set of files, for assorted historical reasons. Kickstart files don't directly work here because they don't allow you to define intermediate objects (add-ons) that are presented in the installer. Anaconda processes variants and add-ons via code that reads .treeinfo. PROPOSAL GIVEN THE NEW UI For reference, the proposed new UI is at: http://linuxgrrl.com/fedora-ux/Projects/Anaconda/Live%20Prototypes/07-softw… As you'll notice, it contains something that looks like a spin or variant on the left, and things that could be considered 'add-ons' on the right. These apply to the choices on the left. This would seem to fit nicely into the variants + add-ons model that RHEL uses. However, there are a couple of problems: - The code that parses the variants/add-on information and creates the appropriate .treeinfo bits does not exist in Fedora This is obviously a simple matter of not-so-simple code. - It requires creating a separate set of metadata (the variants file) that does not currently exist. This is relatively trivial. - It can cause skew between the live and not-live spins if this variants file falls out of sync with the kickstarts in spin-kickstarts. This is ugly - how can we get these to use a single set of data? Unfortunately, kickstart's %packages model doesn't give us an easy way to define a variant and its add-ons in a kickstart file. Ideas? Bill

3 3

[PATCH rhel6-branch] iscsi: interface binding support
by Radek Vykydal 02 Mar '12

02 Mar '12

iSCSI interface binding support (bz #500273) Notes: * Requires changes in dracut: https://bugzilla.redhat.com/show_bug.cgi?id=797158 and further in iscsi-initiator-utils, fresh build: https://brewweb.devel.redhat.com/taskinfo?taskID=4085771 New parameters passed to dracut are described in commit message. * support for GUI, TUI, and kickstart (iscsi --iface option) * All devices must be either bound or using default. More fine-grained implementation would require deeper and invasive changes in GUI and libiscsi. Moreover, combining binding and default doesn't seem as useful or legitmate use case. * Discovery is done for all active interfaces. If user wants to activate additional interface in GUI (activating all needed interfaces in kickstart using network --activate is expected/recommended) and access a target via added interface it must be done before first login to a node of given target. We don't allow re-discovery of targets with nodes logged in as it can corrupt their authentication info. To offer activating of additional devices in GUI "Configure Network" button is added to Add Advanced Target dialog. It just runs nm-c-e as in hostname screen so user has to check Connect Automatically to activate the device (document). * No iBFT support yet. We need to figure out what to pass to dracut and whether dracut can handle it. I have quite simple patch to support iBFT in anaconda ready. 6.4 stuff? * I may squash some of the patches before pushing, I am leaving them split for review. Updates image for RHEL6.3-20120226.n.0 nightly (and hopefuly some later): http://rvykydal.fedorapeople.org/updates.iscsibind.img Screencast of GUI session: http://rvykydal.fedorapeople.org/iscsibind.ogg Screencast of TUI session: http://rvykydal.fedorapeople.org/iscsibind_tui.ogg

3 21

[rhel7-alpha-branch] fix dasd device type detection
by David Cantrell 29 Feb '12

29 Feb '12

This is probably also applicable to master. Tested with latest nightly, three installs in a row too. -- David Cantrell <dcantrell(a)redhat.com> Supervisor, Installer Engineering Team Red Hat, Inc. | Westford, MA | EST5EDT

2 2

[rhel7-alpha-branch] make sure s390x network devices are up
by David Cantrell 29 Feb '12

29 Feb '12

Make sure the network devices used at install time are up on reboot. Hack for alpha 1. -- David Cantrell <dcantrell(a)redhat.com> Supervisor, Installer Engineering Team Red Hat, Inc. | Westford, MA | EST5EDT

2 2

[rhel6] Fix sense of disklabel size check and add some logging (#671230)
by Brian C. Lane 29 Feb '12

29 Feb '12

From: "Brian C. Lane" <bcl(a)redhat.com> cherry-picked from commit 0254ad8d8c6635d67c08a2ad28a84bfae9c880f9 and tested against RHEL6.3 nightly Resolves: rhbz#671230 --- platform.py | 8 +++++++- 1 files changed, 7 insertions(+), 1 deletions(-) diff --git a/platform.py b/platform.py index 9904220..0ce3b70 100644 --- a/platform.py +++ b/platform.py @@ -160,14 +160,20 @@ class Platform(object): """The best disklabel type for the specified device.""" # if there's a required type for this device type, use that labelType = self.requiredDiskLabelType(device.partedDevice.type) + log.debug("required disklabel type for %s (%s) is %s" + % (device.name, device.partedDevice.type, labelType)) if not labelType: # otherwise, use the first supported type for this platform # that is large enough to address the whole device labelType = self.defaultDiskLabelType + log.debug("default disklabel type for %s is %s" % (device.name, + labelType)) for lt in self.diskLabelTypes: l = parted.freshDisk(device=device.partedDevice, ty=lt) - if l.maxPartitionStartSector < device.partedDevice.length: + if l.maxPartitionStartSector > device.partedDevice.length: labelType = lt + log.debug("selecting %s disklabel for %s based on size" + % (labelType, device.name)) break return labelType -- 1.7.7.6

2 1

[lorax] Add yum and anaconda to requires
by Martin Gracik 29 Feb '12

29 Feb '12

--- lorax.spec | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/lorax.spec b/lorax.spec index d4c78f5..b9c0c59 100644 --- a/lorax.spec +++ b/lorax.spec @@ -30,6 +30,8 @@ Requires: gzip Requires: xz Requires: squashfs-tools >= 4.2 Requires: e2fsprogs +Requires: yum +Requires: anaconda %ifarch %{ix86} x86_64 Requires: syslinux >= 4.02-5 -- 1.7.5.4

1 0

[LORAX 1/4] livemedia-creator: Add support for making ami images
by Brian C. Lane 28 Feb '12

28 Feb '12

From: "Brian C. Lane" <bcl(a)redhat.com> AMI images are un-partitioned filesystem images with a grub.conf that is read by the pv-grub bootloader used by EC2. Most of the actual work making the AMI is done in the kickstart. This just creates the image file. --- src/sbin/livemedia-creator | 41 ++++++++++++++++++++++++++++++----------- 1 files changed, 30 insertions(+), 11 deletions(-) diff --git a/src/sbin/livemedia-creator b/src/sbin/livemedia-creator index 23c9b3f..e91c050 100755 --- a/src/sbin/livemedia-creator +++ b/src/sbin/livemedia-creator @@ -46,7 +46,7 @@ from pykickstart.version import makeVersion from pylorax.base import DataHolder from pylorax.treebuilder import TreeBuilder, RuntimeBuilder, udev_escape from pylorax.sysutils import joinpaths, remove, linktree -from pylorax.imgutils import PartitionMount, mksparse +from pylorax.imgutils import PartitionMount, mksparse, mkext4img from pylorax.executils import execWithRedirect, execWithCapture @@ -406,6 +406,23 @@ def get_kernels( boot_dir ): return [f[8:] for f in files if f.startswith("vmlinuz-")] +def make_ami( disk_img, ami_img="ami-root.img", ami_label="AMI" ): + """ + Copy the / partition to an un-partitioned disk image + + ami_img is the filename to write, defaults to ami-root.img + ami_label is the FS label to apply to the image + + All other AMI setup is handled by the kickstart's %post + """ + with PartitionMount( disk_img ) as img_mount: + work_dir = tempfile.mkdtemp() + log.info("working dir is {0}".format(work_dir)) + log.info("creating {0}".format(ami_img)) + mkext4img(img_mount.mount_dir, joinpaths(work_dir, ami_img), label=ami_label) + return work_dir + + def make_livecd( disk_img, squashfs_args="", templatedir=None, title="Linux", project="Linux", releasever=16 ): """ @@ -423,6 +440,9 @@ def make_livecd( disk_img, squashfs_args="", templatedir=None, """ with PartitionMount( disk_img ) as img_mount: + if not img_mount or not img_mount.mount_dir: + return None + kernel_list = get_kernels( joinpaths( img_mount.mount_dir, "boot" ) ) log.debug( "kernel_list = {0}".format(kernel_list) ) if kernel_list: @@ -592,7 +612,7 @@ if __name__ == '__main__': log.error("You need to run this as root") sys.exit( 1 ) - if not os.path.exists( opts.lorax_templates ): + if opts.make_iso and not os.path.exists( opts.lorax_templates ): log.error( "The lorax templates directory ({0}) doesn't" " exist.".format( opts.lorax_templates ) ) sys.exit( 1 ) @@ -614,10 +634,6 @@ if __name__ == '__main__': log.error( "--make-appliance is not yet implemented." ) sys.exit( 1 ) - if opts.make_ami: - log.error( "--make-ami is not yet implemented." ) - sys.exit( 1 ) - if not opts.no_virt and not opts.iso and not opts.disk_image: log.error( "virt-install needs an install iso." ) sys.exit( 1 ) @@ -701,13 +717,16 @@ if __name__ == '__main__': result_dir = make_livecd( opts.disk_image or disk_img, opts.squashfs_args, opts.lorax_templates, opts.title, opts.project, opts.releasever ) + elif opts.make_ami and not opts.image_only: + result_dir = make_ami(opts.disk_image or disk_img) - if not opts.keep_image and not opts.disk_image: - os.unlink( disk_img ) + # cleanup the mess + if disk_img and not opts.keep_image and not opts.disk_image: + os.unlink( disk_img ) - if opts.result_dir: - shutil.copytree( result_dir, opts.result_dir ) - shutil.rmtree( result_dir ) + if opts.result_dir and result_dir: + shutil.copytree( result_dir, opts.result_dir ) + shutil.rmtree( result_dir ) log.info("SUMMARY") log.info("-------") -- 1.7.7.6

1 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 February 2012