I’ve added a module to the 10.3 netboot ISO in order to support an, as yet, unsupported NIC. I’ve rebuild the initrd image and now I’m ready to turn this back into an ISO which will be bootable after being burned.
Of course, this is where I’m running into a problem. I’ve tried using mkisofs to do this but I keep getting an error which I haven’t found a resolution to. This is the command I run:
Wild attempt… throw ‘strace’ in front of your command and post the
output in here. It should show the system attempting to find your ‘bin’
and may give us a clue. Also, any chance you can post the actual
command in case it’s something lost in translation?
| I’ve added a module to the 10.3 netboot ISO in order to support an, as
| yet, unsupported NIC. I’ve rebuild the initrd image and now I’m ready
| to turn this back into an ISO which will be bootable after being
| Of course, this is where I’m running into a problem. I’ve tried using
| mkisofs to do this but I keep getting an error which I haven’t found a
| resolution to. This is the command I run:
| mkisofs -o opensuse_with_module.iso -b path_to/isolinux.bin
| The error I get states that the isolinux.bin cannot be found. I know
| it’s there. I’ve checked, double checked, pentuple checked. I just
| can’t get around the error.
| Anyone have any insight on what I should be doing to make this work?
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
I use acetoneiso on a regular basis and i have to admit that it is really usefull. About the isolinux not found, i know that it is a last chance but: is it in path? I know it should be there but it’ better to check
msnyder@msnyder:~> mkisofs -o opensuse_with_module.iso -b new_iso/boot/i386/loader/isolinux.bin new_iso/
I: -input-charset not specified, using utf-8 (detected in locale settings)
Using VMLIN000.;1 for new_iso/boot/i386/vmlinuz-xenpae (vmlinuz-xen)
Using INITR000.;1 for new_iso/boot/i386/initrd-xen (initrd-xenpae)
genisoimage: Uh oh, I cant find the boot image 'new_iso/boot/i386/loader/isolinux.bin' !
The file is located in /home/msnyder/new_iso/boot/i386/loader/. However, looking at the error output, it seems to be looking for it in boot/ instead of boot/i386/loader. I’ve even copied to boot/ and I’m still getting that error.
I figured out how to get the module into the initrd file and ensure that it is loaded at boot. However, I still couldn’t get the system to boot with network connectivity. I then ran mkinitrd -A to include everything possible and then repackaged the ISO with that (thanks to cat4926 for the info on isomaster).
When I boot the server I can see that it recognizes the module and activates the NICs but when I tell it to do an installation as opposed to booting to the hard drive, it loads the kernel from the CD and then boots the system on the hard drive.
Is there something about mkinitrd -A that makes this expected behavior?
I appear to have the initrd thing set up somewhat properly as far as the network module goes. However, I created it on a server which was already built. I think that may have introduced other problems.
When I boot do the disc I burn after replacing initrd, I see that the driver for the network loads but, instead of going to the NFS server I specified for the installation media, it tries to access the resume device. This fails and the system goes to a shell prompt.
Is there something I need to do to make this not happen?
I couldn’t get kiso to install because it doesn’t work with the latest libcdio file.
I’m curious about one thing now. I can extract the initrd which turns into a 48M directory. I then add the module I need to the modules directory and run find . -print |cpio -o > initrd to repackage it. However, that 48M directory becomes a 185M initrd. Where is all this extra size coming from?