Anaconda-devel September 2006

anaconda-devel@lists.fedoraproject.org

30 participants
32 discussions

Failures in grabber.py _do_grab function while running Anaconda
by Jasper O'neal Hartline 15 Sep '06

15 Sep '06

Hi. Would anyone be able to point me in the right direction for debugging an issue while: 1) Running Anaconda as invoked by Kadischi under a LiveCD environment The exceptions I recieve are along the lines of: OSError No such file or directory '/file/name/here' When the files (RPM headers) do actually exist in grabber.py _do_grab function. Is anyone aware of bizzarre failures with this module? I have an exception anaconda dump file if anyone would like to see it. J. Hartline

1 0

RFI: How to debug a particular problem with grub-install
by Dag Wieers 13 Sep '06

13 Sep '06

Hi, I have a particular problem, only verified with RHEL4AS U4, where Anaconda happily installs and reboots but there does not seem a bootloader installed (or it is installed at the wrong location). This happens on a HP Proliant DL580 G2 with the cciss driver. If I install the bootloader with grub-install myself (using a rescue cd and mount/chroot) then the grub device.map lists /dev/cciss/c0d0 as hd4 and sda as hd0. Not sure if that is the problem, but manually overwriting the device.map and making /dev/cciss/c0d0 hd0 works as expected. Now, instead of asking if this is indeed the problem, I was wondering how someone would go and debug this. Is there some way to know exactly what anaconda has been doing at the end of the installation, without modifying anaconda, without even rebooting and kickstarting again ? I would expect something like a complete 'set -x' and command output of *everything* it did. Or an strace like output. Something that can match behaviour with expectation. Or something you can grep for things, during or after the installation without loosing time or iterating over things. The ultimate goal is to provide sufficient information (and maybe even a patch) to put in a bugzilla report, which I do not have right now. Thanks in avdance for any help pursuiing this :) -- dag wieers, dag(a)wieers.com, http://dag.wieers.com/ -- [all I want is a warm bed and a kind word and unlimited power]

6 11

Re: Anaconda-devel-list Digest, Vol 31, Issue 10
by Steve Barnhart 12 Sep '06

12 Sep '06

On 9/12/06, anaconda-devel-list-request(a)redhat.com > > Hi Steve > Do you have any solution for this ? > > Thanks, No I do not. I know someone on here has to know. Please anyone with some information? > On 9/2/06, Steve Barnhart <stb52988(a)gmail.com> wrote: > > Hello, > > > > I am working on a custom anaconda installation and was wondering how > > (with the new codebase) if it is still possible to add slides and also > > the first 'splash' screen that always appeared before installation. > > Are these features still in the code but perhaps disabled? Thanks for > > your help. > > > -- > http://vnoss.org > Vietnamese Open Source Software Community > > > > ------------------------------ > > _______________________________________________ > Anaconda-devel-list mailing list > Anaconda-devel-list(a)redhat.com > https://www.redhat.com/mailman/listinfo/anaconda-devel-list > > End of Anaconda-devel-list Digest, Vol 31, Issue 10 > *************************************************** > -- Steve

1 0

Custom distro - adding slides, splash
by Steve Barnhart 11 Sep '06

11 Sep '06

Hello, I am working on a custom anaconda installation and was wondering how (with the new codebase) if it is still possible to add slides and also the first 'splash' screen that always appeared before installation. Are these features still in the code but perhaps disabled? Thanks for your help. -- Steve

2 1

anaconda debugging and recovery (was Re: RFI: How to debug a particular problem with grub-install)
by Ed Brown 08 Sep '06

08 Sep '06

Dag Wieers wrote: >>> How would I be able to see from anaconda what command is being run >>> and what output it got back. I am not looking for a solution to >>> this problem directly, I am looking for a way to make obvious what >>> is going on from (during or after) the installation. If something >>> was available to see exactly what happens, what decisions are >>> being made and what programs are run (even without the output) it >>> would positively affect bug-reports and help from the community to >>> improve Anaconda. ... > I'm worried about anaconda development in the sense that many people > use it, but few people improve it. And anaconda is definitely > something that could be improved wrt. debug output. ... > I was not looking for interim solutions really, I know how I can fix > this or work around it from a kickstart file. (Using %pre etc...) I've been making similar suggestions on the kickstart list. So much dialogue on both lists revolves around how to get information about what is going on, correctly or incorrectly. A 'debug mode' framework would not only improve the ability to quickly troubleshoot, it could also be used to provide a recovery capacity that is sorely lacking now. The user experience is of something primitive and broken (like a long phone answering menu that abruptly disconnects you in the end) when an install aborts after investing much time in it. I've done many hundreds of RedHat installs over the last 5 years or so, but it took 4 tries, from scratch, to get FC5-x86_64 installed on a new home computer recently. Many folks strive for minimal %post scripts, preferring to do their post-install configuration 'post-post-install', because it is can be so hard to develop, and maintain, %post (and %pre) scripts. If it were possible to enable 'breakpoints' during anaconda's install phase, possibly on error conditions, with a chance to interact, such as attempt to query state, or manually run failed commands or work around them otherwise, and then continue, it could save an enormous amount of time iterating through install attempts. And why should it be impossible to return to the end of the info gathering phase, with 'back' button opportunity from there, on errors during the install phase (for non-kickstart installs)? Could anaconda provide a 'debugProblem' function that could be called from any interpreter during %post (and perhaps %pre) scripts for kickstart installs, that could display a passed string argument, and provide a shell prompt for debugging and recovery, that on exiting, returned control to %post to be able to continue? How else could anaconda be made more friendly, and flexible? -Ed

2 2

RHPL spec
by Paul Nasrat 08 Sep '06

08 Sep '06

Anyone object if I move make to be under %build rather than %prep section. Paul

2 1

Upgrading Taroon to Nahant
by John Summerfield 07 Sep '06

07 Sep '06

Quick question, has anyone booted Nahant media and upgraded Taroon (oe earlier) to Nahant? Quick second Q Was it difficult? Thanks, I'm just trying to resolve a dispute about whether it works. -- Cheers John -- spambait 1aaaaaaa(a)coco.merseine.nu Z1aaaaaaa(a)coco.merseine.nu Tourist pics http://portgeographe.environmentaldisasters.cds.merseine.nu/ Please do not reply off-list

2 1

Anaconda boot loader problem.
by Rich O'Hare 05 Sep '06

05 Sep '06

I am having trouble installing an updated Fedora Core 5 that I built using the anaconda-runtime buildinstall script. It is a straight Fedora Core 5 distribution, with the updates applied. Everything goes great up until it gets to the boot loader. I think I should mention that I am installing to a SuperMicro that has a Adaptec Zero Channel RAID controller, because it seams to me that the anaconda hardware mapping is incorrect. ****************** /usr/lib/anaconda/gui.py, 1253, handleRenderCall /usr/lib/anaconda/iw/progress-guie.py, 245, renderCallback /usr/lib/anaconda/gui.py, 951, nextclicked /usr/lib/anaconda/dispatch.py, 146, gotoNext /usr/lib/ancaconda/dispatch.py, 217, moveStep /usr/lib/anaconda/bootloader.py, 178, writeBootloader /usr/lib/booty/bootloaderInfo.py, 1193, in write /usr/lib/booty/bootloaderInfo,py, 951, in writeGrub /usr/lib/python2.4/site-packages/rhpl/executil.py, 81, in execWithRedirect OSERROR: [Errno 2] No such file or directory. ******************** Here is how I build the distro: cd $FCBASE/$arch # First createrepo is for the tree createrepo -g Fedora/base/comps.xml . cd .. buildinstall --comp dist-$release.$date \ --pkgorder $FCBASE/pkgfile.$date --version $release \ --product 'Fedora Core' --release "Fedora Core $release" \ --prodpath Fedora $FCBASE/$arch # Note: Had to add expat rpm to pkgfile manually, it was missing. splittree.py --arch=i386 \ --total-discs=5 --bin-discs=5 --src-discs=0 --srcdir=. \ --release-string="Fedora Core $release" \ --pkgorderfile=$FCBASE/pkgfile.$date --distdir=$FCBASE/$arch \ --productpath=Fedora # run createrepo a second time (for the media) cd_discinfo=`head -1 $FCBASE/$arch-disc1/.discinfo` createrepo -u "media://${cd_discinfo}#1" -g Fedora/base/comps.xml \ -o ${arch}-disc1 \ --split ${arch}-disc? # The first CD is special (because it needs to be bootable) mkisofs -q -r -R -J -T -no-emul-boot -boot-load-size 4 -boot-info-table \ -b isolinux/isolinux.bin -c isolinux/boot.cat \ -V "FC $release update$date $arch" \ -A "Fedora Core $release update$date $arch" \ -publisher "$publisher" -p "$publisher" -x lost+found \ -o FC-$release-update$date-$arch-disc1.iso $FCBASE/$arch-disc1 CDS="2 3 4 5" for num in $CDS do echo -n " ... CD$num" mkisofs -q -r -R -J -T \ -V "FC $release update$date $arch" \ -A "Fedora Core $release update$date $arch" \ -publisher "$publisher" -p "$publisher" -x lost+found \ -o FC-$release-update$date-$arch-disc$num.iso $FCBASE/$arch-disc $num done echo "" CDS="1 2 3 4 5" for num in $CDS do implantisomd5 FC-$release-update$date-$arch-disc$num.iso done fi

1 0

(no subject)
by nahum yakubu 05 Sep '06

05 Sep '06

__________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0

anaconda crash
by Nerazzurri 05 Sep '06

05 Sep '06

hi everyone: i am customing the anaconda-10.2.1.5(FC4) but when i install, anaconda crash at the step of turnOnFilesystems (savePartitions()): Traceback (most recent call list) File "/usr/lib/anaconda/gui.py" line 1137... ... File "/usr/lib/anaconda/packages.py" line 579, in turnOnFilesystems diskset.savePartitions () File "/usr/lib/anaconda/partedUtils.py" line 726, in savePartitions disk.commit error: Error: Error informing the kernel about modifications to partition /dev/hda5 - Device or resource busy. This means that Linux won't know about any changes made to /dev/hda5 until you reboot... i know this's a bug of FC4 anaconda, and partly caused by "fd leak" i think there is open fd of "/dev/hd*", and didn't close. details in: (http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=164360) or (http://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=160693) because i have had some works at this version of anaconda(FC4), the only way to me is fix bug, but until now, i still can not find the bugs resources, and i cannot find any patch for this bug too. thanks for anyone giving me any direction or informations

1 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Anaconda-devel September 2006