Hi everyone,
We have a T5240 and noticed we were having issues booting the system using our jumpstart server (it seems as though we don't have the correct drivers for the nxge NIC)
We thought we needed to update our jumpstart server image 9/10 Solaris SPARC image.
Now I'm still learning Jumpstart and I've spent all week with a co-worker on this item so I was helping someone could point us in the right direction.
We downloaded the ISO, mounted it then ran the setup_install_server (is the correct procedure to update the image?) <-- or are we missing something? We tried the following steps but received errors when trying to boot the client.
lofiadm -a /var/tmp/solaris_image /dev/lofi/1
mount -F hsfs -o ro /dev/lofi/1 /mnt
./setup_install_server /export/install/sparc_10
I was using this as reference:
Part*I Using Custom JumpStart (Solaris 10 10/09 Installation Guide: Custom JumpStart and Advanced Installations)
So far we still get errors
On Monday I will get the full details on our errors
but the error we receive when booting a client is along the lines of:
...can not open kernel/sparcv9/unix
Which then has an interactive prompt asking us for the location
(are there any daemons we need to reboot after changing the image? tftpd, bootparamd, nfsd?)
When we copy back our previous image our jumpstart works properly (but not with our T5240)
Any ideas would be greatly appreciated!
This started out as a favour, but now that I can't figure this out
I've been obsessing over it the last week 
./setup-install-server is the correct method to add the iso into the jumpstart tree.
there are however a number of caveats:
did you overwrite the existing iso image (location) that was on the jumpstart server? or did you install to another directory.
the error you are getting means the system cannot boot the kernel - depending on what you have done - it may be necessary to update /etc/bootparams (add_install_client) - make sure you have the right architecture in the command line & iso (root) location.
has this ever worked with the jumpstart server in question? do the sysidcfg / profile configurations exist?
So what we did was we moved the original iso image into another directory then ran the setup_install_server script into the original iso image directory
we then ran our usual wrapper script that would call the add_client_server when adding a jumpstart client to our jumpstart server
the boot, config and config_install locations are all the same in our /etc/bootparams
we made sure to keep our config location the same as the /etc/bootparams
and the sysidcfg and profiles do exist;
If I re-install a jumpstart image do we also have to re-install the boot server as well? (we have the boot server and install server on the same machine)
When I've looked at official documentation all I see is
# ./setup_install_server -b /export/install
But on the below link I saw the CPU architecture supplied with the script; is this just incorrect?
http://adminschoice.com/solaris-jumpstart-howto-guide
# ./setup_install_server -b /export/install sun4u
no need to issue -b if the boot server is the same machine as the jumpstart server. -b flag is required if you're jumpstarting across subnets or you want the boot server to be a different host etc. setup-install-server copies the boot stuff anyway.
okay, so when you boot the client in verbose logging mode what does it say?
boot net -v - install
Thanks for the quick reply Celtic_Monkey!
I'll have to wait until Monday to pass you the OBP information but I can give you the information we recieved from our WORKING jumpstart setup (which let us to believe we needed to update the jumpstart image)
This is what we got from the T5240 server:
SunOS Release 5.10 Version Generic_120011-14 64-bit
Copyright 1983-2007 Sun Microsystems, Inc. All rights reserved.
Use is subject to license terms.
strplumb: unable to get device info for /pci@500/pci@0/pci@8/network@0
strplumb: resolve_netdrv: 6
NOTICE: init_config: unable to get device info for root device /pci@500/pci@0/pci@8/network@0
WARNING: Unable to mount NFS root filesystem: error 6
Cannot mount root on /pci@500/pci@0/pci@8/network@0 fstype nfsdyn
panic[cpu0]/thread=180e000: vfs_mountroot: cannot mount root