View previous topic :: View next topic |
Author |
Message |
salmonix Guru
Joined: 16 Jul 2006 Posts: 410
|
Posted: Thu Oct 12, 2006 4:07 pm Post subject: |
|
|
Well, i can make a try with a debian netboot installer.
But if i add the line in yaboot.conf as suggested for netrooting into the squashfs image file, the kernel still panics because it can not load it.
Can it be because the server can not mount squashfs image? (That is its kernel is not patched to do that?) _________________ Quis custodiet ipsos, custodes? |
|
Back to top |
|
|
salmonix Guru
Joined: 16 Jul 2006 Posts: 410
|
Posted: Sun Oct 15, 2006 11:41 am Post subject: |
|
|
Now, i have decided to use nfs method.
To do that i have debian kernel and initrd.
The yaboot conf. has the followings:
Quote: |
image=vmlinux-2.4.33.3
initrd=vmlinuz-ppcbug.initrd
label=linux
initrd-size=8192
APPEND="ip=dhcp root=/dev/nfs nfsroot=192.168.0.52:/iso/image.squashfs init=/linuxrc" |
Now, i have the following message on the ppc:
Quote: |
<5>RAMDISK: Couldn't find valid RAM disk image starting at 0.
Looking up port of RCP 100005/2 on 192.168.0.52 (<the server's IP)
Root-NFS: Unable to get nfsd port number from server, using default
Looking up port of RCP 100005/1 on 192.168.0.52
Root-NFS: Server returned error -13 while mounting /iso/image.squashfs
VFS: Unable to mount root fs via NFS, trying floppy. |
This is the end. The same if i try to mount an extracted stage3 folder. Each is allowed in "exports".
Firewall is down, nfsd is running, just for the case, accidentally netatalk is running.
I have tried several append conposition - they all end up with this error message.
Without append the debian installer starts.
Slowly I am getting to accept the fact that i have to use Debian... _________________ Quis custodiet ipsos, custodes? |
|
Back to top |
|
|
salmonix Guru
Joined: 16 Jul 2006 Posts: 410
|
Posted: Sun Oct 15, 2006 6:25 pm Post subject: |
|
|
OK. nfsd was somehow mysteriously down. Corrected.
The file system is found now - an uncompressed stage3 -, but using apple.igz the following error comes:
Quote: | hfs_fs: unsipported device block size: 1024
Kernel panic: VFS: Unable to mount root fs on 00:ff |
Well, if i use the debian initrd, it stops with
Quote: | INIT: version 2.86 booting
INIT: Id "c1" respawning too fast: disabled for 5 minutes
(the same runs with "c2" etc. "c6")
INIT: no more processes left in this runlevel |
Then really 5 minutes later the same runs.
Any idea or should i turn to a Debian forum?
(It seems to me some nfsd issue but not sure.) _________________ Quis custodiet ipsos, custodes? |
|
Back to top |
|
|
salmonix Guru
Joined: 16 Jul 2006 Posts: 410
|
Posted: Sun Oct 22, 2006 1:13 pm Post subject: |
|
|
Now, I have a 2.4.33 kernel compiled on a PPC (funny, cause i had to net install a DEBIAN first!), and this kernel has everything needed to run the system. (No modules are allowed, has reiser and ext support and all.)
Now, it boots fine, recognises everything etc.
The NFS root is a path with and untarrecd stage3 tarball. (the latest i could find) but (@chrooting?) i have the message:
Kernel too old
and kernel panic and reboot in 180 seconds.
What is the problem? The kernel i use is a recent one.
I have checked a thread here with a long discussion about a boot disk (that is useless for me cause no floppy around).
What is now?
PS: this work now is only to solve the problem, because of course i could chroot from the running DEBIAN and do whatever i want. NB: that is a strange fact that even Gentoo has not this method prepared for installation, despite of the fact that the Gentoo install is still very manual thus my humble knowledge suggests that i could not rise enormous difficulties to offer this option during installation especiually when one considers the comapartively unfrequent use of optical devices and their costs in the Mac world. _________________ Quis custodiet ipsos, custodes? |
|
Back to top |
|
|
nixnut Bodhisattva
Joined: 09 Apr 2004 Posts: 10974 Location: the dutch mountains
|
Posted: Sun Oct 22, 2006 2:41 pm Post subject: |
|
|
The system from the stage3 you extracted needs a 2.6 kernel. _________________ Please add [solved] to the initial post's subject line if you feel your problem is resolved. Help answer the unanswered
talk is cheap. supply exceeds demand |
|
Back to top |
|
|
salmonix Guru
Joined: 16 Jul 2006 Posts: 410
|
Posted: Mon Oct 23, 2006 8:16 am Post subject: |
|
|
Just one more thing (this is like Matrjoshka: in the problem there is a new, smaller, ant in that a new, smaller and so on! )
I have compiled a 2.6. kernel, checked - it recogninses everything properly.
Now, when netbooting it is still crying for proper NFS.
As I checked now, his problem is that th IP-conf sees a router - which is also a DHCP server. That would not be a problem, because that happened before, but now he sees the router as the
bootserver (not properly), and sees not the PC behind the router (unlike before.)
Of course can not open the nfs root.
Before that did not happen. (Then is was struggling with the DEBIAN installer that i could not remove from the initrd. Bahh...)
I can set the router as virtual server and put my pc behind it if it is the solution. Then:
1. the port that i have to set on the router for the PCserver should be the port of nfsd?
2. Do i have to change exports to serve the router with nfs file system? (Or better to serve the router IP and the PPC IP too with the same NFS root?)
3. The protocol is TCP, UDP or Both? (This has to be set for the router.)
thanx! Close to close this thread. _________________ Quis custodiet ipsos, custodes? |
|
Back to top |
|
|
grazie n00b
Joined: 14 Nov 2005 Posts: 61
|
Posted: Thu Mar 15, 2007 5:55 am Post subject: |
|
|
I too am trying to use an x86 server to boot ppc machines. Has anyone got a working and reliable configution? I've tried using a Debian kernel to start as that's suppposed to work out the box. However, I'm using a router that provides the DHCP service but it gives the IP address OK.
On the client all I get is
Code: | 0 > boot enet:192.168.0.2,yaboot
CLIENT 123456789abc 192.168.0.3
SERVER cba987654321 192.168.0.2
transfer FILE yaboot
TFTP RRQ Timeoutload-size=0 adler32=1
LOAD-SIZE is too small |
Any pointers to help me get a bit further would be gratefully received.
Last edited by grazie on Fri Mar 16, 2007 10:58 am; edited 2 times in total |
|
Back to top |
|
|
grazie n00b
Joined: 14 Nov 2005 Posts: 61
|
Posted: Fri Mar 16, 2007 9:11 am Post subject: |
|
|
In case anyone out there is interested my porblem was not correctly setting up the tftp server which varies slightly from distro to distro. Also I used a debian net booting kernel. It would be great if gentoo had this ability too. |
|
Back to top |
|
|
vlotho Tux's lil' helper
Joined: 06 Jun 2008 Posts: 77
|
Posted: Thu Jan 31, 2013 4:32 pm Post subject: |
|
|
grazie wrote: |
Code: | 0 > boot enet:192.168.0.2,yaboot
CLIENT 123456789abc 192.168.0.3
SERVER cba987654321 192.168.0.2
transfer FILE yaboot
TFTP RRQ Timeoutload-size=0 adler32=1
LOAD-SIZE is too small |
|
I also have this problem. could someone tell me how to solve it? _________________ Ashl3y dit: moi je suce que si le mec me baise après.
f0xy-dj1 dit: remarque s'il le fait pas il t'aura quand même bien baisé. |
|
Back to top |
|
|
Gerk Retired Dev
Joined: 07 May 2002 Posts: 435
|
Posted: Mon Mar 24, 2014 6:00 pm Post subject: |
|
|
Try checking to make sure your TFTP server is configured correctly. |
|
Back to top |
|
|
|