View previous topic :: View next topic |
Author |
Message |
wyrdvans n00b
Joined: 15 May 2002 Posts: 3 Location: Philadelphia
|
Posted: Wed May 15, 2002 4:41 am Post subject: Installation: bootstrap |
|
|
I'm trying to install gentoo on my 8500 powermac. When I get to the scripts/bootstrap.sh step it starts to compile but eventually gives a kernel panic after it prints out a bunch of EXT2-FS Errors. I'm using an ext2 partition but I also tried an xfs and an ext3 partition with similar errors relating to the file system ending in a kernel panic. Anybody have any ideas? |
|
Back to top |
|
|
Gerk Retired Dev
Joined: 07 May 2002 Posts: 435
|
Posted: Wed May 15, 2002 3:02 pm Post subject: |
|
|
Is this with latest ISO image? More details would be helpful. |
|
Back to top |
|
|
wyrdvans n00b
Joined: 15 May 2002 Posts: 3 Location: Philadelphia
|
Posted: Thu May 16, 2002 3:53 am Post subject: |
|
|
I'm using the latest iso(version 1.2). I installed BootX version 1.2.2 and copied the vmlinux file from the boot directory on the cd to the Linux Kernels directory and I copied the initrd.img to the System Folder. When it reboots I have to click options and specify a boot image and make the ram disk size 32000 otherwise after it tries to free the kernel memory I get the error attempt to access beyond the end of device. So I fix that by changing the ramdisk size.
Then I get a bunch of busybox Read-only file system messages and the message Code: | /etc/init.d/rcS: /sbin/ifconfig: No such file or directory | . Before the PCI autodection completes I get this error Code: | /etc/init.d/pnp: cut: command not found | . I assume this is because of the busybox problem.
I remount the ramdisk to be rw(like it says in the Quickstart Guide) Then I proceed with the rest of the installation as it is written in the Guide up until the booststrap part where it fails. I will post the error later because I have to run it again to get the message but if there is any problems so far please enlighten me.
Thanks. |
|
Back to top |
|
|
Gerk Retired Dev
Joined: 07 May 2002 Posts: 435
|
Posted: Thu May 16, 2002 3:10 pm Post subject: |
|
|
to avoid having to remount rw, you can also add the following to the kernel args in bootx: rw
That will mount the ramdisk rw from the get do (and also allow busybox to set al the symlinks).
There have been other reports of problems with people bootstrapping on pre G3 machines, but there are no easy answers yet. I have been told that there is a stage3 tarball (i.e. already bootstrapped and emerge system) going online very soon for ppc users. This may not be a solution, but it will definately be a workaround for people that don't want to spend days bootstrapping on older machines.
This info (about the rw and more bootx info in general) should make it to the quickstart guide online very very soon (is in CVS waiting to get posted right now). |
|
Back to top |
|
|
wyrdvans n00b
Joined: 15 May 2002 Posts: 3 Location: Philadelphia
|
Posted: Thu May 16, 2002 8:49 pm Post subject: |
|
|
I used your suggestion and that solved the problems with busybox but I'm still having problems with the bootstrap. It bombs when it decompresses the glibc package. I get a bunch of these with different blocks and counts:
Code: |
EXT2-fs error (device sd(8,18)): ext2_new_block: Allocating block in system zone - block = 65586
EXT2-fs error (device sd(8,18)): ext2_free_blocks: Freeing blocks in system zones - Block = 65586, count = 7
|
Then it ends with this:
Code: |
kernel BUG at buffer.c:579!
Oops: Exception in kernel mode, sig: 4
|
Then it drops to the shell prompt for a few minutes then throws another:
Code: |
kernel BUG at buffer.c:579!
Oops: Exception in kernel mode, sig: 4
|
any ideas? |
|
Back to top |
|
|
karasu n00b
Joined: 23 Jun 2002 Posts: 21 Location: Eugene, Oregon
|
Posted: Sun Jun 23, 2002 6:50 am Post subject: |
|
|
This is the same problem I encountered while using the stage1 ppc tbz2 ( I can't be sure about version, think is was 1.2 .. maybe it was a 1.1? ) off the CD. Mistakenly posted my problem here: https://forums.gentoo.org/viewtopic.php?t=5852
I found some additional forum messages which possible refer to this same problem
https://forums.gentoo.org/viewtopic.php?t=5363. In that thread, rt_clik mentioned:
So what I did was do a `mke2fs /dev/hda13` since that "error" fubared my whole filesystem, and mounted it up `mount /dev/hda13 /mnt/hikkosu` and then made the tmp dir myself and grabbed the stage2 archive with wget `wget http://eclipse.darkstar.net/~gerk/PowerPC/stage2-ppc-1.2.tbz2 ` and used that instead of the CD's stage1 tarball.
Seems to have done the trick, but I think it'll be another hour before I can be certain.
karasu |
|
Back to top |
|
|
karasu n00b
Joined: 23 Jun 2002 Posts: 21 Location: Eugene, Oregon
|
Posted: Mon Jul 01, 2002 1:06 am Post subject: |
|
|
I'm pretty sure its a kernel issue. Try burning the newest ISO, or get your system up using a different kernel and do the bootstrap/emerge. |
|
Back to top |
|
|
fsck_ms n00b
Joined: 02 Jul 2002 Posts: 25 Location: just outside reality
|
Posted: Wed Jul 03, 2002 6:02 pm Post subject: |
|
|
I've had good luck on my 7300 booting from a 1.1 iso, and using the 1.2 stage1 tarball saved on an hfs partition. Took forever, but it worked. _________________ fortune: The best defense against logic is ignorance. |
|
Back to top |
|
|
pakkypnut n00b
Joined: 10 Jul 2002 Posts: 26 Location: Canberra
|
Posted: Wed Jul 10, 2002 1:02 am Post subject: kernel panic |
|
|
Had exactly the same problem, kernal panics and ext3 errors, I had discovered however, that once I had partitioned the drive with MACOS and formatted as hfs under MACOS, (even the swap space) these problems went away never to be seen again. I think my initial mistake was using what macos (and mac-fdisk) called -free space- as my swap drive - wrong, it must be partitioned and formatted! _________________ And some said that even the trees had been a bad move, and that no one should ever have left the oceans.
Douglas Adams - Hitchhikers Guide to the Galaxy |
|
Back to top |
|
|
|