[K12OSN] Problems loading Kernel on new install
Steve
skerr at rx30.com
Fri Oct 21 16:14:33 UTC 2011
Its there, in fact in my last revision, I accidentally have it twice.
once as a global and specified under the
workstation parameters. Ill include it in case there is anything helpful.
I also just captured a packet dump. I have a couple udp packets covering
the dhcp negotiation, then its all tftp
beginning with a request from the client to read the specified kernel on
the server and ending with an ack from
the client to a kernel data send from the server, and thats it.
rpcinfo seems to indicate nfs is ok, exports is configured wide open as
per doc. firewall stopped. selinux disabled.
even set /opt 777
One thing Im curious about. What determines where the client sees
lts.conf ? I notice its now under /var/lib/tftpboot,
the lts.conf in the chroot is empty, whereas in 4.2 it read the version
in the chroot.
Heres the dhcpd.conf file. I've made a number of revisions, from going
with the default to using the workstation defs
as specified on the wiki when using a fedora 11 chroot.
Again this is an attempt at a mixed network setup. TC's have been
factory flashed to work over 1067. I have ltsp-dhcp disabled
dhcp args set to 1067 under sysconfig and bridge removed. per netstat ,
and log files were listening ok on the server.
# Sample configuration file for LTSP dhcpd
# The goal of this provided dhcpd.conf is to boot as many different
clients as possible without manual configuration.
# Some clients cannot be booted with this, and you will need to create
your own custom dhcpd.conf.
# Some more complex network topologies will also need a custom dhcpd.conf.
#
default-lease-time 21600;
max-lease-time 21600;
ddns-update-style none;
allow booting;
allow bootp;
option subnet-mask 255.255.255.0;
option broadcast-address 10.10.0.255;
option routers 10.10.0.58;
option domain-name-servers 10.10.0.58;
next-server 10.10.0.58;
option domain-name "ltsp";
option option-128 code 128 = string;
option option-129 code 129 = text;
option root-path "10.10.0.58:/opt/ltsp/fed11-i386";
option option-221 code 221 = text;
option option-128 e4:45:74:68:00:00;
option option-129 "DPORT=1067";
get-lease-hostnames true;
shared-network WORKSTATIONS {
subnet 10.10.0.0 netmask 255.255.255.0 {
range dynamic-bootp 10.10.0.62 10.10.0.65;
use-host-decl-names on;
option log-servers 10.10.0.58;
}
}
## example configurations for specifying specific kernels to specific
clients
## These examples below are VERY OLD, uncertain if they work
group {
use-host-decl-names on;
option log-servers 10.10.0.58;
host lt01 {
hardware ethernet 00:13:90:02:7f:11;
fixed-address
lt01; ### ip of lt01
specified in hosts
filename "/ltsp/fed11-i386/wraplinux-nbi.ltsp";
option root-path "10.10.0.58:/opt/ltsp/fed11-i386";
}
# host ws002 {
# hardware ethernet 00:D0:09:30:6A:1C;
# fixed-address 172.31.100.2;
# filename "/ltsp/vmlinuz.ltsp";
# option option-128 e4:45:74:68:00:00;
# option option-129 "NIC=ne";
# }
# host ws003 {
# hardware ethernet 00:D0:09:30:28:B2;
# fixed-address 172.31.100.3;
# # kernels are specified in /tftpboot/ltsp/i386/pxelinux.cfg/
# filename "/ltsp/i386/pxelinux.0";
# }
# Apple Specific Settings
# host ws007 {
# hardware ethernet 00:30:65:69:23:60;
# fixed-address 172.31.100.4;
# option root-path "172.31.100.254:/opt/ltsp/ppc";
## filename "yaboot";
# option vendor-class-identifier "AAPLBSDPC";
# }
}
On 10/21/2011 10:54 AM, Dean Jones wrote:
> Looks like your dhcpd.conf is not specifying a root path to me, so the
> clients do not know where to mount root.
>
>
> On Fri, Oct 21, 2011 at 6:48 AM, Steve<skerr at rx30.com> wrote:
>> After poking around some, I discovered gdm was not configured to allow
>> remote
>> X connections, so I edited custom.conf apropriately, in hope I'll sometime
>> get far enough that
>> it matters.
>>
>> ok , so I set dhcpd.conf back up to point to the fed11 chroot.
>> specifying elf.ltsp kernel:
>> After .... (ELF) .... done I get
>> dracut: Fatal no or empty root = argument
>> dracut: Refusing to continue
>> kernel panic - not syncing : Attempted to kill init
>>
>> I then pointed filename to the wraplinux kernel:
>> It looks like it goes further, get a scrolling screen like the kernel is
>> loading
>> for instance I see:
>> registered taskstats version 1
>> magic number: 11:139:136
>> freeing unused kernel memory 404k
>> then basically the same dracut error messages and:
>> Pid: 1 init tainted:G
>> then a call trace
>>
>> vmlinuz.ltsp - does not fit in memory
>>
>> pxelinux.0
>> shows fedora 11 scroll bar moving left to right at the bottom of
>> the screen
>> then displays similiar dracut errors
>> then kernel panic - not syncing. Attempted to kill init
>> then another call trace
>>
>>
>>
>>
>> On 10/20/2011 11:08 PM, Burke Almquist wrote:
>>
>> It almost looks to me as if it's having trouble mounting the NFS root for
>> the client.
>> It appears to be getting the kernel over tftpd and then hanging?
>>
>>
>> On Oct 20, 2011, at 12:43 PM, Steve wrote:
>>
>> Hi All,
>> Im trying to setup LTSP using the InstallGuide and running into
>> difficulties.
>> I installed a fresh copy of CentOS 6 on a Dell precision T3400 in a mixed
>> network environment
>> with my dhcp server using port 1067.
>> I cant get into the TC console at the moment for exact specs, but its a "red
>> faced" jammin from disklessworkstations
>> , manufactured june 2009, flashed to work over 1067. It works with my older
>> versions of ltsp on centos 5.4.
>>
>> I , think, followed the directions on the wiki carefully, didnt run into any
>> problems or receive any error messages.
>> When the TC is booting it appears to be detected as etherboot-5.4 and
>> attempts to load the elf.ltsp kernel.
>> In messages, on the server, i see
>> DHCPDISCOVER
>> DHCPOFFER
>> DHCPREQUEST
>> DHCPACK
>>
>> all with sane values, dhcp looks ok
>> the last line is
>> xinetd : START tftpd pid="pidnumber" from ="ipaddy"
>>
>> on the TC , it shows the dhcp negotion and the last line is:
>> Loading 10.10.0.58:/ltsp/i386/elf.ltsp (ELF) ... done
>>
>> there it stops. No further output on TC or client.
>>
>> I also tried forcing the wraplinux and pxe.0 kernels, which seem to get
>> further. I get the scrolling bar at the bottom that
>> says scientific linux 6.1 on the bottom right, but ultimately results in
>> dracut error messages and a kernel panic.
>>
>> I hoping I missed something simple. Any pointers are appreciated.
>>
>> Thanks
>> Steve
>>
>>
>>
>>
>> _______________________________________________
>> K12OSN mailing list
>> K12OSN at redhat.com
>> https://www.redhat.com/mailman/listinfo/k12osn
>> For more info see<http://www.k12os.org>
>>
>> _______________________________________________
>> K12OSN mailing list
>> K12OSN at redhat.com
>> https://www.redhat.com/mailman/listinfo/k12osn
>> For more info see<http://www.k12os.org>
>>
>>
> _______________________________________________
> K12OSN mailing list
> K12OSN at redhat.com
> https://www.redhat.com/mailman/listinfo/k12osn
> For more info see<http://www.k12os.org>
> .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/k12osn/attachments/20111021/e61dd9fd/attachment.htm>
More information about the K12OSN
mailing list