Anaconda-devel April 2009

anaconda-devel@lists.fedoraproject.org

30 participants
169 discussions

[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] Changes for using md on top isw format
by Jacek Danecki 18 Jun '09

18 Jun '09

Hi, This series of 6 patches obsoletes patches sent in topic "Changes for using mdadm for isw_raid_member". All patches were rebased to the version anaconda-11.5.0.42.1. These patches need pyparted >= 2.0.11 and patch for mdadm which add support for -E --export option. MD containers and arrays are created in directory /dev/md as links to devices in /dev Activation of MD arrays is done by adding disks to the array using command mdadm -I. MDRaidArrayDevice._devDir value was changed to "/dev/md" MD arrays are visible as HARD Drives, and can be partitioned. MD containers are displayed as RAID Devices. Support for new kernel parameter iswmd was changed. Please review. Regards

5 58

[PATCH] Suppress warnings about changing to text-mode if not in graphics-mode
by Niels de Vos 05 May '09

05 May '09

If the user requested text-mode or cmdline-mode it is not needed to display a message about missing X (minstg2.img) or too little memory. Signed-off-by: Niels de Vos <niels.devos(a)wincor-nixdorf.com> --- This patch is against rhel5-base and tested with SRPM of anaconda-11.1.2.168-1.el5. anaconda | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/anaconda b/anaconda index 58f3865..4ccc9a1 100755 --- a/anaconda +++ b/anaconda @@ -448,7 +448,7 @@ def checkMemory(opts): # override display mode if machine cannot nicely run X if not flags.test: - if iutil.memInstalled() < isys.MIN_GUI_RAM: + if opts.display_mode == 'g' and iutil.memInstalled() < isys.MIN_GUI_RAM: stdoutLog.warning(_("You do not have enough RAM to use the graphical " "installer. Starting text mode.")) opts.display_mode = 't' @@ -806,7 +806,7 @@ if __name__ == "__main__": # now determine if we're going to run in GUI or TUI mode # # if no X server, we have to use text mode - if not (flags.test or flags.rootpath) and (rhpl.getArch() != "s390" and not os.access("/mnt/runtime/usr/bin/Xorg", os.X_OK)): + if opts.display_mode == 'g' and not (flags.test or flags.rootpath) and (rhpl.getArch() != "s390" and not os.access("/mnt/runtime/usr/bin/Xorg", os.X_OK)): stdoutLog.warning(_("Graphical installation not available... " "Starting text mode.")) time.sleep(2) -- 1.5.5.6

2 3

misc fixes for rawhide
by David Cantrell 05 May '09

05 May '09

1) Add %{?dist} to release number in anaconda.spec. I'm tired of not having this in the RPM name. 2) Fix problem in loader where it was prompting again to bring up the network even if we already had an active connection. 3) Remove minihal.py and get list of NICs from NetworkManager. -- David Cantrell <dcantrell(a)redhat.com> Red Hat, Inc. / Honolulu, HI

4 11

[PATCH RHEL-5.4 fips] mkinitrd: make sure all crypto modules are loaded before use
by Hans de Goede 04 May '09

04 May '09

We need to have verified *all* crypto modules before *any* of them may be used in fips mode. Since module verification happen at load time this patch simply loads them all early on in the initrd. Regards, Hans

4 3

[PATCH] mdraid incomplete arrays
by Joel Granados Moreno 04 May '09

04 May '09

Hello list. This patch fixed the realated bug, but I am unsure if it breaks something elsewhere. What do ppl think? Is it sane? Regards. Joel Andres Granados Brno, Czech Republic, Red Hat.

3 3

[PATCH 1/4] Handle clearpart in the early kickstart pass too.
by Chris Lumens 01 May '09

01 May '09

--- kickstart.py | 16 ++++++++++++++-- 1 files changed, 14 insertions(+), 2 deletions(-) diff --git a/kickstart.py b/kickstart.py index c1098ed..a95762e 100644 --- a/kickstart.py +++ b/kickstart.py @@ -239,11 +239,17 @@ class ClearPart(commands.clearpart.FC3_ClearPart): if self.type is None: self.type = CLEARPART_TYPE_NONE - hds = map(lambda x: x.name, self.handler.id.storage.disks) + hds = map(udev_device_get_name, udev_get_block_devices()) for disk in self.drives: if disk not in hds: raise KickstartValueError, formatErrorMsg(self.lineno, msg="Specified nonexistent disk %s in clearpart command" % disk) + # If doing the early kickstart processing, we will not yet have an + # instdata attribute. That's okay because we pull the lists right + # out of this class instead of the instdata. + if not self.handler.id: + return retval + self.handler.id.storage.clearPartType = self.type self.handler.id.storage.clearPartDisks = self.drives if self.initAll: @@ -1024,7 +1030,7 @@ class EarlyKSHandler(superclass): self.id = self.anaconda.id self.maskAllExcept(["vnc", "displaymode", "text", "cmdline", - "graphical", "rescue", "ignoredisk"]) + "graphical", "rescue", "ignoredisk", "clearpart"]) class AnacondaPreParser(KickstartParser): # A subclass of KickstartParser that only looks for %pre scripts and @@ -1148,6 +1154,12 @@ def fullCommandPass(anaconda, file, earlyKS): import storage anaconda.id.storage.ignoredDisks = earlyKS.ignoredisk.ignoredisk anaconda.id.storage.exclusiveDisks = earlyKS.ignoredisk.onlyuse + + anaconda.id.storage.clearPartType = earlyKS.clearpart.type + anaconda.id.storage.clearPartDisks = earlyKS.clearpart.drives + if earlyKS.clearpart.initAll: + anaconda.id.storage.reinitializeDisks = earlyKS.clearpart.initAll + storage.storageInitialize(anaconda) handler = AnacondaKSHandler(anaconda) -- 1.6.1.3

2 5

[PATCH] Be more aggressive unmounting before install starts (#498260)
by Jeremy Katz 01 May '09

01 May '09

In the live install, having things mounted leads to tracebacks. This isn't really fool-proof, but it should help --- liveinst/liveinst.sh | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/liveinst/liveinst.sh b/liveinst/liveinst.sh index 844ca06..badb3f5 100644 --- a/liveinst/liveinst.sh +++ b/liveinst/liveinst.sh @@ -58,6 +58,7 @@ fi # devkit-disks is now mounting lots of stuff. for now, let's just try to unmount it all umount /media/* +for mnt in $(cat /proc/mounts |grep ^/dev |grep -v live | awk {'print $2;'}) ; do umount $mnt 2>/dev/null ; done /sbin/swapoff -a /sbin/lvm vgchange -an --ignorelockingfailure -- 1.6.1

2 1

Testing Anaconda
by Clyde E. Kunkel 01 May '09

01 May '09

Will there be an opportunity to test anaconda again before Fedora 11 goes gold? TIA

2 3

(no subject)
by Chris Lumens 30 Apr '09

30 Apr '09

I don't know why we do an arch upgrade check in two places. One really needs to go. Until we figure that mess out, this patch at least fixes the second check to understand i386 and i586 are probably the same for upgrades. - Chris

2 3

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Anaconda-devel April 2009