[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Sunblade 100 IDPROM not valid



Rebooting my Sunblade 100 stopped at the ok prompt, the startup information included something like this:
Ethernet address ff:ff:ff:ff:ff:ff, Host ID ffffffff
And there was a message about the IDPROM being not valid.
ok>


The Debian installation guide mentions this briefly:
-------
5.1.4. IDPROM Messages

If you cannot boot because you get messages about a problem with “IDPROM”, then it's possible that your NVRAM battery, which holds configuration information for you firmware, has run out. See the Sun NVRAM FAQ for more information.
--------

The suggested FAQ link is very detailed and helpful but involves replacing the dead battery chip:
http://www.squirrel.com/sun-nvram-hostid.faq.html

The machine is about 5 years old and has run almost continuous all that time. But it was powered off for about a week recently. When it failed I wanted a quick fix that did not involve hardware and found the following which may be helpful to someone else.

I have only tried it with a reboot command, not power-off or net boot yet.

Step 1
boot failed.
Ethernet address ff:ff:ff:ff:ff:ff, Host ID ffffffff
Machine fails to boot, does not load SILO, stops at ok>
Now it appears that most commands do not work, but fortunately "boot" works
ok> boot

Step 2
SILO loads, select required kernel from the menu
Boot sequence stops with message "CLOCK: Clock was stopped. Kick start "
Then it appears to hang.
But wait a minute or less and the boot continues...
The machine boots up but ifconfig shows the ethernet address is FF:FF:FF:FF:FF:FF
and network cannot be reached.

Step 3
Gentoo gives a clue here in Code List 3.2/3.4
http://www.gentoo.org/doc/en/gentoo-sparc-faq.xml
So i did this as root:
# eeprom local-mac-address?=true
# reboot

If I understand correctly, this forces the MAC address from the network card and the machine reboots with a valid MAC address.
Maybe I was just lucky, or the battery is not completely dead yet?
But i am reporting this as it might be a temporary fix that will help someone in a hurry.
The mentioned NVRAM FAQ explains the proper solution.

thanks

Jim


Reply to: