SUMMARY: Jumpstart fails to mount root - netmask problem

From: Luke A. Kanies <luke_at_madstop.com>
Date: Tue Sep 18 2001 - 10:53:22 EDT
I haven't been able to find an actual solution to this problem, although I
got emails from two other people experiencing the same problem.  If I had
a contract I would report it to Sun, but it'll have to wait until then.

In the meantime, I am jumpstarting my workstations within the same virtual
Class C as the server.

Original question is below.

------------------------

Hi Sun Managers, I am trying to jumpstart a Blade 100 using the 07-01
release of Solaris 8, and am having a problem with the netmask.  I am
using 192.168.240.0 as my network address, with a netmask of
255.255.248.0.  Obviously, this isn't actually my choice, so I can't
change that to suit my needs.

My jumpstart server's IP address is 192.168.240.71.  If I try to jumpstart
a machine with an IP address of 192.168.240.73, it works fine, but if I
try it with an IP address of 192.168.246.243, which is within the network
space, then I get to the point where Root should be mounted and it bombs
out.

Here is the output from a boot net - install:

Sun Blade 100 (UltraSPARC-IIe), No Keyboard
OpenBoot 4.0, 640 MB memory installed, Serial #51177550.
Ethernet address 0:3:ba:c:e8:4e, Host ID: 830ce84e.



Rebooting with command: boot net - install
Boot device: /pci@1f,0/network@c,1  File and args: - install
Timeout waiting for ARP/RARP packet
SunOS Release 5.8 Version Generic_108528-09 64-bit
Copyright 1983-2001 Sun Microsystems, Inc.  All rights reserved.
rtioctl: kstr_ioctl failed: error 128
whoami: couldn't add route: error 128.
WARNING: nfsdyn_mountroot: NFS3 mount_root failed: error 128
Cannot mount root on /pci@1f,0/network@c,1 fstype nfsdyn

panic[cpu0]/thread=10408000: vfs_mountroot: cannot mount root

0000000010407970 genunix:vfs_mountroot+70 (10435800, 0, 0, 10410910, 10,
14)
  %l0-3: 0000000010435800 0000000010438fa8 000000004e000000
0000000010435be0
  %l4-7: 0000000000000000 0000000010413608 00000000000b6476
0000000000000476
0000000010407a20 genunix:main+94 (10410158, 2000, 10407ec0, 10408030,
fff2, 1004
e5e0)
  %l0-3: 0000000000000001 0000000000000001 0000000000000015
0000000000000e7f
  %l4-7: 0000000010428ab8 0000000010462210 00000000000cbe50
0000000000000540

skipping system dump - no dump device configured
rebooting...
Resetting ...

-------------------

Here is a snoop off of the same hub as the booting machine (snipped for
brevity, and only snooping for traffic to this host):

192.168.246.243 -> 192.168.240.71 NFS C READ2 FH=9FE6 at 6144 for 1024
192.168.240.71 -> 192.168.246.243 NFS R READ2 OK (1024 bytes)
192.168.246.243 -> 192.168.240.71 NFS C READ2 FH=9FE6 at 7168 for 1024
192.168.240.71 -> 192.168.246.243 NFS R READ2 OK (1024 bytes)
192.168.246.243 -> BROADCAST    BPARAM C WHOAMI? 192.168.246.243
192.168.240.71 -> 192.168.246.243 BPARAM R WHOAMI? bengel in xxx.xxx.com

-----------

At this point, basically as soon as the machine receives the WHOAMI
response, it dumps.

I am not using a name service, and I seem to be having some problems with
name resolution; I had to mess with the nsswitch.conf and hosts settings
to get the machine to work at the lower address.  I plan on moving to an
LDAP NS at some point, but I don't believe that will suffice anyway.

Regardless, I can surmount the NS problems; how do I fix the problem with
the netmask?  I have run rpc.bootparamd in debug mode, and it appears to
be sending all of the correct information, including netmask, I've set the
correct netmask in the sysidcfg file, and I can't find anywhere else to do
so.  There is a netmasks file in the Tools/Boot dir of the image, but that
is the directory I am trying to mount (and it's only a soft link to
something in /tmp anyway), so I don't expect that changing that will fix
anything.

Does anyone have any ideas?  I could continue to use a temporary address
to build these machines, or I could try to get addresses reallocated
within the same class C as the jumpstart server (or move the jumpstart
server), but neither one of those actually solves the problem, and this is
just going to get more difficult, because these are the first solaris
boxen on this network and we're going to continue adding more.  I need to
be able to jumpstart any boxes within 192.168.240.0.

Any help would be greatly appreciated, and I will, of course, summarize.

-- 
"The optimist proclaims that we live in the best of all possible worlds,
and the pessimist fears that this is true."
                -- James Branch Cabell 1879-1958



_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
Received on Tue Sep 18 15:53:22 2001

This archive was generated by hypermail 2.1.8 : Wed Mar 23 2016 - 16:32:31 EDT