Hello,
Recently our jumpstart server (running solaris 9 on an ultra 5) failed
us.
It stopped booting but had recently been used to successfully install
solaris 9 on a variety of newer suns in our lab. Instead of trying to
trouble shoot the old sun hardware, I simply put the disk in another
ultra 5.

It comes up with the correct hostname and doesn't appear to have any
problem
on our net. When I tried to jumpstart an sunblade 1000 (same results
with a
2000) I get the following error.

----Begin output -----

Rebooting with command: boot net:dhcp - install
Boot device: /pci@8,700000/network@5,1:dhcp File and args: - install
Timeout waiting for BOOTP/DHCP reply. Retrying ...
---- end block

This is all the same as before, so far so good.
Here's where the error comes up

---- output
29200 krtld: error during initial load/link phase
panic - boot: exitto64 returned from client program
Program terminated
{0} ok
----End output ----

The two machines (the 1000 and the 2000) yield the same
output and have both been jumpstarted before. While
this same disk was in the old server. So all the on-disk
configuration stuff should be the same. (unless something
has been corrupted)

Any ideas as to where the problem could be? Am I going to
have to rebuild my Jumpstart server?

Thanks

--
Ben VanderSluis
sysadmin
Unniversity of Minnesota
Department of Mechanical Engineering
slice [at] me.umn [dot] edu

As a courtesy to anyone else trying to solve this problem...
Its my guess that there was corruption somewhere in the Boot
environment that gets mounted, under OS in the jumpstart tree.
Or possibly the tftp boot images had somehow been comprimised.
I rebuilt the machine, used the setup_install_server script to
rebuild the OS directory. The rest I just copied from a backup
and the system seems to be working now.

Best of luck.