Anaconda-devel December 2007

anaconda-devel@lists.stg.fedoraproject.org

24 participants
35 discussions

Kickstart installation via network -- not able to override network settings?
by Ray Van Dolson 14 Jan '08

14 Jan '08

This is re-post from the kickstart list -- wasn't sure which one was more appropriate. In any case, I am doing network-based kickstart installs with RHEL 5.1 and running into an issue where anaconda is not prompting me for network card information. We specify the location of the kickstart file as located on an NFS share on bootup. The installation URL is via HTTP. Back in the RHEL4 days we used this same method -- we essentially use the kickstart file as "defalts" and do an interactive install so the installer may override things on an as-needed basis. Often times I machine requires a static IP address so the installer would simply provide this information when prompted to by anaconda. However, in RHEL5 we are _never_ prompted for network information at all. From reading the kickstart documentation it seems that if you are doing a network installation and do not provide any network configuration information in the kickstart file then it is assumed that you will be acquring your IP address via DHCP and the installed system is also configured in this same way. It would seem that my alternatives would be to either specify the full static IP information via the network keyword in the kickstart file or rely on a CD-based installation so that I am presented with the network settings dialog during installation. Neither is a great option as this means I will have to maintain special copies of the kickstart file for machines requiring static IP's or require the user to configure networking post installation to override the default DHCP setup. I realize I could also pass IP information via the bootup options, but I am trying to make this as trivial as possible for installers, and all that extra typing is prone to typo's. I have tried modifying my kickstart file to include only "network --bootproto=static" to no avail. Apparently it needs all of the options for this to work. And I still don't get a screen in the installer to manually configure networking. I've enabled debugging during the installation and don't see anything pointing to why the networking screen isn't brought up. Am I doing something wrong? Is this just "how it works" now with kickstart / anaconda? Thanks in advance. Ray

3 4

Change to GPLv2+
by David Cantrell 23 Dec '07

23 Dec '07

Update license boilerplates to GPLv2+ ("or any later version"). This was discussed by myself and Jeremy a while ago off list. My concern is needing anaconda to link with things that may be or will become incompatible with our current GPLv2 only license for anaconda. Perfect example is GNU parted, which is now GPLv3 (not my fault! I'm just the messenger!). We seemed to agree it was doable for anaconda, but someone needed to sit down and actually update all of the source files with new license boilerplates. I have now done that. What to expect besides a huge number of patches: 1) All Makefiles, shell scripts, *.py, *.c, and *.h source files now have a uniform GPLv2+ license boilerplate. 2) The physical address of the FSF is removed and replaced with an HTTP URL. The FSF now prefers this to the physical address in the boiler plate text. 3) Year ranges on copyright statements are expanded to list each individual year. Ranges aren't valid per FSF legalites, so each year has to be listed explicitly. I don't really understand the need to list more than one year, but I'm not here to ask that question. 4) Any author names and email addresses have been moved to the Red Hat Author(s) list under the license boilerplate. 5) The license boilerplate notes we are licensed under GPL version 2 or at your option any later version. 6) Code that we are using that did not write I just left alone. For example, the gptsync directory. Questions or comments, send them my way.

4 18

want to use "badblocks -wsvf -p 2" in anaconda installer 10.4.X.X on CentOS 4.4
by Parimal Sardar 23 Dec '07

23 Dec '07

hi I want to use *badblocks -wvf -p 2 . * That is badblocks with read &write destructive test & passes 2 times. my ks.cfg part --ondrive=sda --badblocks / --fstype ext3 --size=400 --asprimary I tried to change fsset.py. fsset.py 176 def badblocksDevice(self, entry, windowCreator, chroot='/'): 177 if windowCreator: 178 w = windowCreator(_("Checking for Bad Blocks"), 179 _("Checking for bad blocks on /dev/%s...") 180 % (entry.device.getDevice(),), 100) 181 else: 182 w = None 183 184 devicePath = entry.device.setupDevice(chroot) 185 args = [ "/usr/sbin/badblocks", "-svwf", devicePath ] 186 ________________________________________________________________ **throw me following error. How to fix it ? Need help ----------------------------------------------------- Traceback (most recent call last ): File "/usr/lib/anaconda/gui.py", line 1024, in handldeRenderCallback self.currentWindow.renderCallback() File "/usr/lib/anaconda/iw/progress_gui.py", line 242, in renderCallback self.inf.icw.nextClicked() File "/usr/lib/anaconda/gui.py", line 789, in nextClicked self.dispatch.gotoNext() File "/usr/lib/anaconda/dispatch.py", line 171, in gotoNext self.moveStep() File "/usr/lib/anaconda/dispatch.py", line 239, in moveStep rc = apply(func, self.bindArgs(args)) File "/usr/lib/anaconda/dispatch.py", line 568, in turnOnFilesystem thefsset.checkBadblocks(insPath) File "/usr/lib/anaconda/fsset.py", line 1409, in checkBadblocks self.badblocksEntry(entry, chroot) File "/usr/lib/anaconda/fsset.py", line 1386, in badblocksEntry entry.fsystem.badblocksDevice(entry, self.progressWindow, chroot) File "/usr/lib/anaconda/fsset.py", line 230,in badblocksDevice val = (long l[0] * 100) / long(l[1]) ValueErro:invalid literal for long () : Reading and comparing : 0 ---------------------------------------------------------------------------- Thanks & regards, Parimal -- -----BEGIN PGP PUBLIC KEY BLOCK----- Version: GnuPG v1.2.6 (GNU/Linux) mIsERzf5dgEEAOETNneNJpRr9jDQWhPOO56Fd1VkD7HsSJMUJEnHDLex1lZhwCDq A8G71H5T0Ccr752RSitD686K9zc59K6oMHEKs5DPPYA6JSNF/5/wIfHPdhy9IXwG ga+SiopBHDXbA/g+9IYo88yaDBaaRWN0sdCoArAIw9PJutMbsyivtDMvAAYptEJQ YXJpbWFsIChUaGlzIGlzIFBhcmltYWwnIGdwZyBrZXkuKSA8cGFyaW1hbC5zYXJk YXJAdWZvbW92aWV6LmNvbT6IugQTAQIAJAUCRzf5dgIbAwUJABJ1AAYLCQgHAwID FQIDAxYCAQIeAQIXgAAKCRBIe8snWGuNWSL4BADCUQ+NodPQR/a9IoiSU5P94nFm LNNKbwiIzzI1iiUUvDQcIvCGRIzfyd+Mnsf/Cq5Hx7VpQuwm7/4vxUEjL908JLXa cf1bF/LOaPc2gyp6QNOH+moSGu1zGWbPZ5HeNpNTk3K0exxAY6bYiOSSc7QKQVoz uvieByhlWKd8pVhn6w== =tIAO -----END PGP PUBLIC KEY BLOCK-----

2 1

Do we read the same books?
by osamabin 21 Dec '07

21 Dec '07

I just joined Shelfari to connect with other book lovers. Come see the books I love and see if we have any in common. Then pick my next book so I can keep on reading. Click below to join my group of friends on Shelfari! http://www.shelfari.com/Register.aspx?ActivityId=59888148&InvitationCode=39… osamabin Shelfari is a free site that lets you share book ratings and reviews with friends and meet people who have similar tastes in books. It also lets you build an online bookshelf, join book clubs, and get good book recommendations from friends. You should check it out. -------- You have received this email because osamabin (parimal.sardar(a)gmail.com) directly invited you to join his/her community on Shelfari. It is against Shelfari's policies to invite people who you don't know directly. Follow this link (http://www.shelfari.com/actions/emailoptout.aspx?email=anaconda-devel-list@…) to prevent future invitations to this address. If you believe you do not know this person, you may view (http://www.shelfari.com/osamabin) his/her Shelfari page or report him/her in our feedback (http://www.shelfari.com/Feedback.aspx) section. Shelfari, 616 1st Ave #300, Seattle, WA 98104

1 0

[PATCH] anaconda dekudzu-fication, final bits
by Bill Nottingham 20 Dec '07

20 Dec '07

This set of patches adds a minimal hardware probing library to anaconda for the loader to use, and switches to it. It would be simpler to not have to maintain this code ourselves, but.... Notes: - Useful descriptions for network devices are not hooked up yet - Driver disks/manual driver selection is still broken Tested with CD installs. Network installs seem screwy with network config (unrelated to this - rawhide is similarly broken) Bill

2 5

[PATCH]Fix gtk.MessageDialog Exception
by HARA Hiroshi 19 Dec '07

19 Dec '07

Hi, When I tested the latest anaconda in my PC,I encountered the following exception. [exception] -------- anaconda None exception report Traceback (most recent call first): File "/usr/lib/anaconda/gui.py", line 788, in __init__ self.dialog = gtk.MessageDialog(mainWindow, 0, style, buttons, text) File "/usr/lib/anaconda/gui.py", line 1067, in messageWindow custom_buttons, custom_icon, run=True, parent=parent).getrc() File "/usr/lib/anaconda/iw/autopart_type.py", line 213, in addZfcpDrive self.intf.messageWindow(_("Error"), e) File "/usr/lib/anaconda/iw/autopart_type.py", line 240, in addDrive rc = self.addZfcpDrive() TypeError: GtkMessageDialog.__init__() argument 5 must be string or None, not exceptions.ValueError -------- anaconda sometimes calls anaconda.intf.messageWindow with exceptions.ValueError like this. But gtk.MessageDialog in messageWindow can not accept it. so I added str() to gtk.MessageDialog like the attached patch. ---------------------- self.dialog = gtk.MessageDialog(mainWindow, 0, style, buttons, str(text)) ---------------------- or should I change all of calling side like the following? ---------------------- self.intf.messageWindow(_("Error"), str(e)) or self.intf.messageWindow(_("Error"), "%s" %(e,)) ---------------------- Would you review the attached patch and give me a comment, please? thanks, Hiroshi HARA

1 0

[PATCH (stage1)] Rework module loading
by Bill Nottingham 19 Dec '07

19 Dec '07

One of the things that anaconda uses kudzu for is determining the list of modules to load. This patch fixes that, along with doing a variety of other things, determined by the following logic: - We want to move to a model where we just load modules from the normal information and aliases exported in sysfs - This means that we are loading (generally) by modalias, not by module name - Ergo, we need to either duplicate all the alias handling, sysfs walking, etc. in anaconda... - Or, we use the system modprobe and udev coldplug. (That is what this implements.) - To use the system modprobe, we either need to hack it to understand the cpio module ball... - Or, we move to just shipping a normal tree of modules. (That is what this implements.) - To do so, either we need a special static version of modprobe... - Or, we switch to a dynamic stage 1 (That is what this implements.) The way module loading now works: - Under normal circumstances, modules are loaded via standard udev coldplug mechanisms - Early module loading is still supported - Passing module options on the commandline is still supported - blacklist=<foo> is now supported, to blacklist automatic loading of a particular module - nofirewire, nousb, etc. are ignored. The only noXX handled is 'noprobe', which disables udev coldplug entirely. - Manual driver selection is still allowed, if necessary. Things not yet modified/fixed in this patchset: - Manual driver selection is aborting in newt somewhere. May be unrelated. - Driver disks are not modified to the new tree layout. Basically, they should now be extracted and copied to /lib/modules/$(whatever)/updates/. Some tweaks with depmod may be necessary to get the aliases in the right place. Size concerns: Before: 7.2MB After: 8.6MB Could be reduced by weaning the glib2 dependencies, not linking zlib statically in modutils, and a couple of other things. Zee diffstat: b/Makefile.inc | 2 b/loader2/Makefile | 8 b/loader2/cdinstall.c | 14 b/loader2/cdinstall.h | 8 b/loader2/driverdisk.c | 14 b/loader2/driverdisk.h | 1 b/loader2/driverselect.c | 9 b/loader2/hardware.c | 213 +------ b/loader2/hardware.h | 22 b/loader2/hdinstall.c | 20 b/loader2/hdinstall.h | 4 b/loader2/loader.c | 136 ---- b/loader2/loader.h | 3 b/loader2/method.h | 5 b/loader2/modules.c | 1146 +++++---------------------------------- b/loader2/modules.h | 42 - b/loader2/nfsinstall.c | 8 b/loader2/nfsinstall.h | 4 b/loader2/telnetd.c | 4 b/loader2/telnetd.h | 4 b/loader2/urlinstall.c | 7 b/loader2/urlinstall.h | 4 b/scripts/Makefile | 1 b/scripts/buildinstall.functions | 120 ++++ b/scripts/mk-images | 291 ++------- b/scripts/upd-instroot | 127 ---- b/utils/Makefile | 11 loader2/firewire.c | 93 --- loader2/firewire.h | 7 loader2/modstubs.c | 263 -------- loader2/modstubs.h | 9 loader2/moduledeps.c | 196 ------ loader2/moduledeps.h | 15 loader2/pcmcia.c | 148 ----- loader2/pcmcia.h | 10 loader2/usb.c | 142 ---- loader2/usb.h | 10 utils/moddeps.c | 91 --- utils/trimmodalias | 23 39 files changed, 491 insertions(+), 2744 deletions(-) Bill

3 28

[PATCH]Validation of root password with cracklib
by HARA Hiroshi 18 Dec '07

18 Dec '07

Hi, Current anaconda validates only length of root password. but passwd command validates more things for password like the following... ------------------------------------------- # passwd Changing password for user root. New UNIX password: BAD PASSWORD: it is based on a dictionary word Retype new UNIX password: ------------------------------------------- so I added the validation of root password to anaconda using cracklib same as passwd. Would you review the patches, please ? and I'd appreciate it if you could give me the comment. thank you,

6 21

[PATCH] don't immediately retry to download a package
by Chris Lumens 18 Dec '07

18 Dec '07

The following patch changes our failure callback to wait before retrying to download a package. We have to do this because urlgrabber doesn't provide a mechanism for this which I guess could be considered a feature because of the flexibility it allows. Anyway it starts by waiting 0.25s and waits twice as much each time. The use of the window is obvious. Thoughts? - Chris diff --git a/yuminstall.py b/yuminstall.py index 6af8b65..aae0037 100644 --- a/yuminstall.py +++ b/yuminstall.py @@ -556,6 +556,14 @@ class AnacondaYum(YumSorter): def urlgrabberFailureCB (self, obj, *args, **kwargs): log.warning("Try %s/%s for %s failed" % (obj.tries, obj.retry, obj.url)) + delay = 0.25*(2**(obj.tries-1)) + if delay > 1: + w = anaconda.intf.waitWindow(_("Retrying"), _("Retrying package download...")) + time.sleep(delay) + w.pop() + else: + time.sleep(delay) + def getDownloadPkgs(self): downloadpkgs = [] totalSize = 0

2 2

%post --name
by Christopher Boumenot 18 Dec '07

18 Dec '07

It would be nice to know what post script is executing, but do to their nature that is impossible. Would you accept a patch that added --name argument to %post (and %pre) to tell Anaconda what the scriptlet's name is? I am open to other suggestions. Thanks, Christopher

4 11

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Anaconda-devel December 2007