I've just purchased an old used Blade 2000 (or 1000, whatever) and I'm
setting about installing Solaris 10 08/07 onto it. The machine's DVDROM
is cranky about burned DVDs, but I sort of expected that. So, I planned
on jumpstarting it off my OpenSolaris (b72) x86 machine.

I've got the install server set up, DHCP configured and whatnot. But
when it comes to booting it, it seems to stop soon after loading the
sun4u/inetboot, just after mounting the net root. I never actually make it
to the Jumpstart phase.

snoop shows NFS activity still happening, though I can't tell what the
machine is actually doing (lots of GETATTR3s). When I do a "boot net:dhcp
-v - install", this is what I see:

Rebooting with command: boot net:dhcp -v - install
Boot device: /pci@8,700000/network@5,1:dhcp File and args: -v - install
Timeout waiting for BOOTP/DHCP reply. Retrying ...
4000 Using BOOTP/DHCP...
BOUND: IP address is: aaa.bbb.ccc.ddd
Found aaa.bbb.ccc.1 @ 0:7:84:bc:a0:c0
root server: jumpstarter (aaa.bbb.ccc.eee)
root directory: /jumpstarter_d0/jumpstart/Solaris10SPARC/Solaris_10/Tools/Boot
Found aaa.bbb.ccc.eee @ 0:c:76:94:92:40
Size: 0x7f8c8+0x1c829+0x3049f Bytes
module /platform/sun4u/kernel/sparcv9/unix: text at [0x1000000, 0x107f8c7] data at 0x1800000
module misc/sparcv9/krtld: text at [0x107f8c8, 0x1099c97] data at 0x184ccc8
module /platform/sun4u/kernel/sparcv9/genunix: text at [0x1099c98, 0x12066a7] data at 0x1852a00
module /platform/SUNW,Sun-Blade-1000/kernel/misc/sparcv9/platmod: text at [0x12066a8, 0x1206b37] data at 0x18aa5e0
module /platform/sun4u/kernel/cpu/sparcv9/SUNW,UltraSPARC-III+: text at [0x1206b80, 0x12196b7] data at 0x18aaf00
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.
Ethernet address = 0:3:ba:3a:9:8d
Using default device instance data
mem = 2097152K (0x80000000)
avail mem = 2080169984
root nexus = Sun Blade 2000/1000 (2 X UltraSPARC-III+)
pseudo0 at root
pseudo0 is /pseudo
scsi_vhci0 at root
scsi_vhci0 is /scsi_vhci
schppm0 at root: SAFARI 0x8 0x410050 ...
pcisch0 at root: SAFARI 0x8 0x700000
pcisch0 is /pci@8,700000
PCI-device: ebus@5, ebus0
ebus0 is /pci@8,700000/ebus@5
xcalppm0 at ebus0: offset 1,e
xcalppm0 is /pci@8,700000/ebus@5/ppm@1,e
pseudo-device: dld0
dld0 is /pseudo/dld@0
PCI-device: network@5,1, eri0
eri0 is /pci@8,700000/network@5,1
SUNW,eri0 : 100 Mbps full duplex link up
root on jumpstarter:/jumpstarter_d0/jumpstart/Solaris10SPARC/Solaris_10/Tools/Boot fstype nfsdyn
todds12870 at ebus0: offset 1,300070
todds12870 is /pci@8,700000/ebus@5/rtc@1,300070
mc-us30 at root: SAFARI 0x0 0x400000 ...
mc-us30 is /memory-controller@0,400000
mc-us31 at root: SAFARI 0x1 0x400000 ...
mc-us31 is /memory-controller@1,400000
bbc_beep0 at ebus0: offset 1,32
bbc_beep0 is /pci@8,700000/ebus@5/beep@1,32
upa64s0 at root: SAFARI 0x8 0x480000
UPA64S-device: SUNW,afb@0,0, afb #0
afb0 is /upa@8,480000/SUNW,afb@0,0
se0 at ebus0: offset 1,400000
jPCI-device: usb@5,3, ohci0/serial@1,400000
ohci0 is /pci@8,700000/usb@5,3
cpu0: UltraSPARC-III+ (portid 0 impl 0x15 ver 0x23 clock 900 MHz)
cpu1: UltraSPARC-III+ (portid 1 impl 0x15 ver 0x23 clock 900 MHz)
pcisch1 at root: SAFARI 0x8 0x600000
pcisch1 is /pci@8,600000
PCI-device: SUNW,qlc@4, qlc0
qlc0 is /pci@8,600000/SUNW,qlc@4fp1 is /pci@8,600000/SUNW,qlc@4/fp@0,0

.... and then we stay here forever.

I seem to be able to boot a 11/06 CD without any issue, and if worse comes
to worse I'll burn 08/07 onto CDs and install that way. But I'm interested
in what might be going on here, and if I'm missing something, or just being
dumb somehow. This is the first Blade I've worked with.

--
Brandon Hume - hume -> BOFH.Ca, http://WWW.BOFH.Ca/

hume.spamfilter@bofh.ca wrote:
> I've got the install server set up, DHCP configured and whatnot. But
> when it comes to booting it, it seems to stop soon after loading the
> sun4u/inetboot, just after mounting the net root. I never actually make it
> to the Jumpstart phase.
>
> snoop shows NFS activity still happening, though I can't tell what the
> machine is actually doing (lots of GETATTR3s).

All for the same FH or different ones? Any attempts to actually read or
write? Any filenames you can see in the traffic just before the hang?

> When I do a "boot net:dhcp
> -v - install", this is what I see:

> Rebooting with command: boot net:dhcp -v - install
> Boot device: /pci@8,700000/network@5,1:dhcp File and args: -v - install
[snip lots of good boot stuff]
> PCI-device: SUNW,qlc@4, qlc0
> qlc0 is /pci@8,600000/SUNW,qlc@4fp1 is /pci@8,600000/SUNW,qlc@4/fp@0,0
>
> ... and then we stay here forever.

So it's definitely got a good root mount and loaded the kernel. It's
like there's a bad module or the image is corrupt.

You might try reloading the disk image from the .iso.

Another thing is to double-check that you *always* had the image shared
with 'ro' and 'anon=0'. I've seen jumpstarting clients mount the image
read/write, then during boot they'll scribble some stuff on it that
corrupts it.

--
Darren Dunham ddunham@taos.com
Senior Technical Consultant TAOS http://www.taos.com/
Got some Dr Pepper? San Francisco, CA bay area
< This line left intentionally blank to confuse you. >

Darren Dunham <ddunham@taos.com> wrote:
> So it's definitely got a good root mount and loaded the kernel. It's
> like there's a bad module or the image is corrupt.
> You might try reloading the disk image from the .iso.

That seems to have been exactly the problem. Instead of reloading the
image, I just shared out the DVD (mounted ro), and the machine booted
fine. So either the copied image got wrecked (not sure how, it was shared
ro as well) or there was some kind of weirdness with sharing it off ZFS.

Regardless, I'm on my way. Thanks muchly.

--
Brandon Hume - hume -> BOFH.Ca, http://WWW.BOFH.Ca/

On Oct 9, 3:53 am, hume.spamfil...@bofh.ca wrote:
> Darren Dunham <ddun...@taos.com> wrote:
> > So it's definitely got a good root mount and loaded the kernel. It's
> > like there's a bad module or the image is corrupt.
> > You might try reloading the disk image from the .iso.
>
> That seems to have been exactly the problem. Instead of reloading the
> image, I just shared out the DVD (mounted ro), and the machine booted
> fine. So either the copied image got wrecked (not sure how, it was shared
> ro as well) or there was some kind of weirdness with sharing it off ZFS.

FWIW I had to perform some zfs set jiggery-pokery to my 2nd JS server
(the one I use
to upgrade JSS #1) . I too used a zfs partition as the NFS mount
point.
shareall just didnt cut it : > I can look up what I did if you want...

gerryt <lepsysinc@gmail.com> wrote:
> shareall just didnt cut it : > I can look up what I did if you want...

Nah, the machine is built and being configured now. I thought about trying
again using "zfs share" instead, but lost interest.

--
Brandon Hume - hume -> BOFH.Ca, http://WWW.BOFH.Ca/