I'm currently trying to build a CentOS 5 live CD with the livecd-creator
script (version 007). I modified the creator script to avoid
try/except/finally structures which are unsupported by python 2.4.3. I
also had to force the creation of /etc/sysconfig in install_root to
avoid errors during the language and timezone configuration.
Now, I am stuck with this error:
[root@localhost centos-live]# livecd-creator
Package cadaver - 0.22.5-1.el5.rf.i386 already installed and latest version
Package cadaver - 0.22.5-1.el5.rf.i386 already installed and latest version
Traceback (most recent call last):
File "/usr/bin/livecd-creator", line 1082, in ?
File "/usr/bin/livecd-creator", line 1063, in main
File "/usr/bin/livecd-creator", line 866, in install
File "/usr/bin/livecd-creator", line 689, in configureSystem
subprocess.call(["/usr/bin/passwd", "-d", "root"],
File "/usr/lib/python2.4/subprocess.py", line 412, in call
return Popen(*args, **kwargs).wait()
File "/usr/lib/python2.4/subprocess.py", line 542, in __init__
File "/usr/lib/python2.4/subprocess.py", line 975, in _execute_child
OSError: [Errno 2] No such file or directory
At this point during the livecd construction, only boot, dev, etc,
lost+found, proc, selinux, sys and var directories are present in
Any help would be appreciated,
To users of the LiveCD and LiveDVD creation utility: Kadischi
There is now a subpackage for kadischi called kadischi-tools.
This subpackage contains find-live-cd, eject-live-cd and scanswap.
Until Kadishci and Kadischi-tools makes it into the main Fedora repositories
it will be required to include package zlib into your repository
transaction if using
a local disk or networked repository source with Kadischi whether you
are using the
CVS compiled and installed method or RPM methods.
To do this, simply modify your comps.xml file and add zlib as a
For RPM users, simply add kadischi-tools as a mandatory package, it
already requires zlib.
If you do not include zlib as a package in the transaction set, booting
the LiveCD or LiveDVD
will likely fail due to it's requirements on libz.so.1 from zlib package.
If you are using a stock Fedora repository, those users will simply need
until kadischi and kadischi-tools is added to the main repository.
Alternatively, you can set up a local repository for inclusion into
still use a stock Fedora repository along with a localized disk
repository with kadischi-tools
in it and listed.
The purpose of creating the subpackage was to alleviate the current
binaries being compiled against
zlib, which pciutils requires and pciutils-devel is used to compile
This was a better approach than statically compiling the binaries.
If you have any questions, feel free to ask. This is a simple notice for
Kadischi utility users.
>From: Cain, Brian (GE Healthcare) [mailto:Brian.Cain@ge.com]
>Sent: Friday, May 11, 2007 10:31 AM
>To: Aaron Hanson
>Subject: FW: [Fedora-livecd-list] Netboot
>> -----Original Message-----
>> From: fedora-livecd-list-bounces(a)redhat.com
>> [mailto:firstname.lastname@example.org] On Behalf Of Aaron
>> Sent: Monday, May 07, 2007 12:24 PM
>> To: fedora-livecd-list(a)redhat.com
>> Subject: [Fedora-livecd-list] Netboot
>> Hi All -
>> I though this would have been discussed by now but I
>> any relevant threads in the archive. I would like to boot the entire
>> livecd image via network. To my basic understanding, the
>> to do this might be:
> I think you've got the right idea on this. I was
>considering doing something similar -- I was thinking about
>doing an install to an NFS root using some of the code in
>livecd-creator. What I like about your idea is that once the
>image has been copied over, you don't really need the network anymore.
> Have you made any progress? Would you be willing to
>keep me (and/or the list) in the loop on how this goes?
Hi Brian -
Thanks for your suggestions and yes, I will continue to post
progress on this issue to the list. That is, unless someone feels it is
I found my first mistake. The ramfs is a simple compressed cpio;
I forgot to use the "-H newc" option to cpio, which is necessary in this
case because the image is very big.
Now I can net-boot and get to a bash shell. Unfortunately the
"init" script produced by mayflower is still confused by the absence of
any root device. I'm now trying to come up with a small patch to get
Hi All -
I didn't get any responses on my question earlier in the week, so I
just forged ahead. I'm trying to get a net-boot of the complete livecd
image. I've been merging the isolinux/initrd.img with the entire
squashfs os.image and sending it to the target device with 'pxelinux'.
The kernel loads, and the initrd is dutifully transferred, but I
think I'm building the initrd wrong, because the kernel can't figure it
out (see below). Does anybody have some tips on the correct way to
construct the initrd? Thanks...
Linux version 2.6.21-1.3116.fc7
(brewbuilder(a)hs20-bc1-7.build.redhat.com) (gcc 7
BIOS-provided physical RAM map:or
checking if image is initramfs...<6>Switched to high resolution mode on
it isn't (no cpio magic); looks like an initrd
Freeing initrd memory: 98747k freed
md: ... autorun DONE.
RAMDISK: Compressed image found at block 0
RAMDISK: incomplete write (-28 != 32768) 16777216
isofs_fill_super: bread failed, dev=md1, iso_blknum=16, block=32
No filesystem could mount root, tried: iso9660
Kernel panic - not syncing: VFS: Unable to mount root fs on
I am using the latest kadischi to build a LiveCD on FC6. When booting
the kernel, the LiveCD reports errors as follows:
Loading ATA modules
FATAL: Module ata_generic not found
FATAL: Module pata_ali not found
And about 30 other PATA modules.
The error seems benign as it finishes booting successfully. Is there a
way to get rid of these error messages?
I was excited to see the release of the Fedora LiveCD as I'm looking
around at different distributions but for some reason I can't get it to
see my internet connection at all.
If anyone could point me in the right direction I'd appreciate it.
NIC: Cnet PRO200WL PCI (not wireless)
Router: Linksys RTP300
Output of ifconfig:
eth0 Link encap:Ethernet HWaddr 00:80:AD:79:95:90
inet addr: 192.168.15.102 Bcast: 192.168.15.255 Mask: 255.255.255.0
inet6 addr: fe80::280:adiff:fe79:9590/64 Scope: Link
UP BROADCAST MULTICAST MTU:1500 Metric: 1
RX packets: 13 errors: 170 dropped: 0 overruns: 0 frame: 0
TX packets: 0 errors: 10 dropped: 0 overruns: 0 carrier: 10
collisions: 0 txqueuelen: 1000
RX bytes: 2597 (2.5 KiB) TX bytes: 0 (0.0 b)
Interrupt: 201 Base address: 0xec00
I noticed that it says UP but not RUNNING.
The Ubuntu LiveCD access the internet just fine so I know it's possible
to MAKE it work. I just like to see it work before making room for a
I've used Linux before but it's been ages ago and before I had a NIC so
I'm pretty new to this part.
Any help would be appreciated!
I am familiar with installation CD customization for FC3(genhdlist's times).
And now, I am just starting with FC6.
Now, I am just trying to remove some package and add another one
under Fedora/RPMS folder.
I did that already playing also with ks.cfg and "createrepo -g
../../base/comps.xml ." into
the RPMs directory.
1) How do I check all the rpm packages dependencies of what I added
and removed in the RPMS folder ?
2) Everytime I change the RPMs folder's content, should I rerun
"createrepo -g" ?
3) Based on regular installation from the original FC6 DVD I changed
the bottom of my ks.cfg to:
Now, during the installation, anaconda is complaining that cannot
found "comps-extras". Also /mnt/source directory is not being
Thanks in advance for any help.
When installaling a kde livecd and selecting a different timezone (eg. one
from europe) the monitor will suspend
When I select a different timezone (eg. one from europe) in Anaconda during
installation from a kde livecd the monitor will suspend when Anaconda changes
the timezone (AFAIK after formatting the partitions). It will come back when
moving the mouse or hitting any key. But I think this is a little bit
confusing for non-experienced users.
In gnome livecd this would not happen. I'm not sure if this is related to
anaconda, metacity/gnome or kwin/kde (powersaving is disabled in kde). I've
also tried this with installing and using openbox on the gnome cd. Then the
monitor is also suspended.
Any ideas how to solve this?
Thanks to the switch to the Fedora 7 base I've been able to get much
farther in livecd creation. However I'm not having a problem where after
creating my initial image, I run a -base-on instance with a custom
package that assigns local settings and a root user password. At the end
of each run I see a flag go by that says "Removing password for user
root". Sure enough that is happening. I can't find anywhere in my custom
package that would be doing it, so I have to assume it's the
livecd-creator application at the end of its run. Is there a way to
disable that? Or a method other than a script within an RPM that would
be preferred for this?
12 Commerce Ave.
W Lebanon, NH 03784
Ph: 603-298-5509 x365