Anaconda-devel May 2007

anaconda-devel@lists.fedoraproject.org

26 participants
25 discussions

Check the RPM deps
by Mirko Gmail 22 May '07

22 May '07

Finnaly i'm able to install my custom RHEL5 distro. But i have about 590 rpms in my cd and at the end Anaconda install only 298 rpms, i see much dependency problem in anaconda log during installation, but can i solve this problem befor burn the new cd? Before genhdlist warning me, but now? createrepo only takes the list of rpms and make a repodata... Please give me a little help!

2 1

Could not allocate partitions as primary -- forced to reboot
by Ray Van Dolson 16 May '07

16 May '07

Hello all, this is somewhat of an annoyance and I hope this is the correct mailing list on which to post regarding it. I will describe the most recent scenario in which this happened to me: I installed RHEL4 U5 x86_64 onto a blade within a PowerEdge 1955 via Kickstart. The system was brand new and contained Dell's utility partitions still. My kickstart config file's disk section looks like the following: autopart bootloader --location=mbr --append="rhgb quiet" The GUI installer uses my kickstart defaults and I was able to modify the layout the partitioning tool had given me to remove the Dell utility partitions and then to recreate the Linux filesystem layout. I did so without LVM. Ended up with partitions sda1, 2 and 3 for /boot, swap and / respectively. After a couple days we needed to reinstall this machine. I started up a kickstart install again using the same config file as above, but this time the GUI installer gave me the following when I got to the partitioning section at the beginning of installation: "Partitioning failed: Could not allocate partitions as primary partitions" The only option I was given was to reboot the system. I tried commenting out the "autopart" line in my kickstart config file, but met with the same results. The only way I could proceed was to either a) Not use kickstart at all (a manual installation let me partition by hand without error) b) Use fdisk to wipe out my partitions ahead of time. I feel like this situation should be handled better by anaconda -- there should be an option other than reboot as it would be trivial to clear the partitions out from within the partition editor. Yes, the obvious thing is to use clearpart --all, but in some cases I may not want to wipe out every partition. The installer should make its best effort to auto-partition with any free space available and present the user with the option to adjust if necessary. At the very least it does not seem necessary to force me to reboot. I have also run into this issue with existing Windows partitions on the system, etc. Is this a potential bug or am I misunderstanding something simple here? TIA, Ray

2 1

Changing MTU for pxeboot
by Olivier BONHOMME 15 May '07

15 May '07

Hello everybody, I am trying to install a Fedora Core 6 using pxeboot. The problem is that the server which contains the install images has a MTU set to 9000. Then when I try to download the install image on the system which I try to install, the transfer fails because I have a MTU 9000 on a side and a MTU 1500 on the other side. I saw on the Redhat bugzilla, that it was possible to send a new MTU parameter to /sbin/loader and anaconda using the command line. I tried to put mtu=9000 on my pxelinux.cfg file but it doesn't seem to work. Does somebody has this problem and does somebody know how to change the MTU parameter on /sbin/loader and anaconda in a automatically way ? Thank you for your answers Best Regards, Olivier BONHOMME

4 13

PATCH: Fix disable dmraid on text and graphical mode
by HARA Hiroshi 14 May '07

14 May '07

Hi, I found User always can not select "Disable dmraid device" on "Advanced Storage Options" This problem will happened on all arch. (this is strange bug.) I did some test on ICH6R with text and graphical mode after applying this attached patch (linx updates). (I can select "Disable dmraid device" and then /dev/mapper/ddf1_raid1 can be changed to /dev/sda and /dev/sdb) Would you review my Patch,please? and if you have a comment, would you give it to me ? thank you very much

3 5

Anaconda send signal to running Dogtail testcase
by Alexander Todorov 10 May '07

10 May '07

Hello list, Do you think it is a good thing anaconda to inform a running dogtail testcase before it exists? We can send SIGUSR1 to the dogtail testcase and let it handle this. An example what this can be used for: write output to dogtail logs and use %post script in kickstart to save the logs on the installed machine. Some questions come in mind: 1) Do we have to wait for the running dogtail process to exit normally before continuing to %post step in anaconda? If we wait and the other process hangs up? The solution that I see here is send SIGUSR1 first, wait a while (how long) then send SIGKILL to avoid hanging up. 2) What is an exception in anaconda occurs? Say while handling partitioning. Does it execute %post if present or directly shuts down? 3) The best place to implement this? I think adding a new step in dispatch.py and handling exceptions is enough. What's your opinion? Any feedback is welcome. Greetings, Alexander.

2 1

Editing xml files
by EXT-Rodgers, Marianne 09 May '07

09 May '07

How in the Sam Hill to you edit xml files on CentOS? We are supposed to edit the comps xml, where I hear the @groups and their rpms are, but I can't read the comps.xml. I've seen some directions to update the comps.xml, but I want to edit my kickstart file and get it right. Where are the lists of rpms (like in a @group) that correspond to the rpm's in the @group? If they are in an xml file, I'm still lost. I also think that somewhere there is a list of Red Hat rpms that correspond to CentOS rpms. That would be great too. How in the Sam Hill do you write to a CD on CentOS? I've looked at http://www.linuxheadquarters.com/howto/basic/cdrecord.shtml, but I still can't write to a CD. I've done all the things advised. I can burn DVDs all day, but not a CD. I've been working on these questions for over a week, Google and all. I am trying to made a customized install using kickstart. I'm kind of tired of trying, trying, and trying and getting no where. Sorry for the attitude, I'm just a little frustrated. While I'm at it, is there a way to completely automate the installation, so you don't have to type in "linux ks=cdrom:somenameks.cfg" I did it once before. I wrote the command to the hard drive, used it, then deleted it. My development HD got destroyed. I had no way to back it up. I don't work on a network. I have to spend hard earned real dollars to get a USB, or better yet a firewire HD for backups. Haven't been able to do that yet. Any guidance, places to go, etc, will be greatly greatly appreciated!!!!! Thanks so much! ;) The comments & Opinions expressed are my own & do not represent the views of United Space Alliance

2 1

Problem with RHEL5 customization: buildinstall error
by Mirko Gmail 08 May '07

08 May '07

Hi everyone! I use to build our custom distro on one cd with RHEL3 adn RHEL4, now i'm trying on RHEL5... I understand the use of createrepo but i have some problem to run buildinstall, exactly i don't understand the parameters... what is prodpath? Well i try to run buildinstall: /usr/lib/anaconda-runtime/buildinstall --pkgorder /home/RH5_CUSTOM/pkgorder.txt --version "5.0" --product "Red Hat Enterprise" --release "5.0" --prodpath /home/comps /home/RH5_CUSTOM And i have a couple of error... /home/RH5_CUSTOM/buildinstall.tree.13558 /home/RH5_CUSTOM/i386/Server rpm2cpio: /home/RH5_CUSTOM//home/comps/anaconda-runtime-[0-9]*: No such file or directory cpio: premature end of archive /home/RH5_CUSTOM/i386/Server `/home/RH5_CUSTOM/buildinstall.tree.13558/usr/lib/anaconda-runtime/./upd-ins troot*': No such file or directory `/home/RH5_CUSTOM/buildinstall.tree.13558/usr/lib/anaconda-runtime/./mk-imag es*': No such file or directory `/home/RH5_CUSTOM/buildinstall.tree.13558/usr/lib/anaconda-runtime/./makesta mp.py*': No such file or directory `/home/RH5_CUSTOM/buildinstall.tree.13558/usr/lib/anaconda-runtime/./buildin stall*': No such file or directory Building images... /usr/lib/anaconda-runtime/buildinstall: line 134: /home/RH5_CUSTOM/buildinstall.tree.13558/upd-instroot: No such file or directory Creating repository metadata... Getting package order... PYTHONPATH=/tmp/treedir.13558/instimage/usr/lib/anaconda /tmp/treedir.13558/instimage/usr/lib/anaconda-runtime/pkgorder /home/RH5_CUSTOM /home/comps /usr/lib/anaconda-runtime/buildinstall: line 142: /tmp/treedir.13558/instimage/usr/lib/anaconda-runtime/pkgorder: No such file or directory Making images... /usr/lib/anaconda-runtime/buildinstall: line 146: /home/RH5_CUSTOM/buildinstall.tree.13558/mk-images: No such file or directory Writing .discinfo file /usr/lib/anaconda-runtime/buildinstall: line 149: /home/RH5_CUSTOM/buildinstall.tree.13558/makestamp.py: No such file or directory Ok guys can u give me some suggest? I know that something is going wrong and it's me...!

2 1

R: Problem with RHEL5 customization: buildinstall error
by Mirko Gmail 08 May '07

08 May '07

> /home/RH5_CUSTOM/buildinstall.tree.13558 /home/RH5_CUSTOM/i386/Server > rpm2cpio: /home/RH5_CUSTOM//home/comps/anaconda-runtime-[0-9]*: No such file or directory Ok i have not anaconda rpm in my directory... But my pkgorder file have only these files: kernel-doc-2.6.18-8.el5.noarch.rpm kernel-PAE-devel-2.6.18-8.el5.i686.rpm kernel-devel-2.6.18-8.el5.i686.rpm kernel-headers-2.6.18-8.el5.i386.rpm kernel-xen-devel-2.6.18-8.el5.i686.rpm kernel-PAE-2.6.18-8.el5.i686.rpm kernel-2.6.18-8.el5.i686.rpm kernel-xen-2.6.18-8.el5.i686.rpm iscsi-initiator-utils-6.2.0.742-0.5.el5.i386.rpm Frustrated...

2 1

Re: Upgrade failure
by xxxxxx xxxxx 07 May '07

07 May '07

> I updated rescue.py and forgot to update network.py. Committed change > to anaconda CVS. Replace pumpNetDevice() call with dhcpNetDevice(dev) > and only pass in the dev object to the configNetDevice() function. Thanks for that. Any idea when the new anaconda RPM will be uploaded to the fedora repository?

2 1

Upgrade failure
by xxxxxx xxxxx 07 May '07

07 May '07

Hi, I have updated my distro with the latest anaconda, hoping to get rid of the annoying error: https://www.redhat.com/archives/anaconda-devel-list/2007-April/msg00060.html I was hoping the new release (11.2.0.57) would fix it, since noone replied to my post. Anyway, I added that, and updated some deps, (libdhcp, booty, pykickstart, etc), but probably missed one. I constantly get the following error: -- File "/usr/lib/anaconda/network.py", line 339, in lookupHostname self.gateway) --- etc etc it also mentions the following files: yuminstall.py, backend.py, dispatch.py, text.py ends with: -- TypeError: configNetDevice() takes exactly 2 arguments (4 given) Does anyone have any idea why I am getting this, and how I can fix it? Am I missing a dependency? Thanks.

2 1

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 May 2007