Sun Fire T1000 ). Only in the ServiceProcessor ( LOM - light out manager ) of the system the showhost command showed a different banner.
The system got a new job, so a colleague installed the last recommended patch cluster and after the reboot, the system hang in a reboot cycle
complaining that the Hypervisor can't stop cpu's with a bad trap (24) , there were also timeouts in cpu xcalls.
The solution was to install the last firmware for the system.
- Therefore the system must be powered down ( poweroff ).
- The firmware-patch must be downloaded to an FTP server and extracted.
( see http://www.oracle.com/technetwork/systems/patches/firmware/index.html and you need an account )
- In the Serviceprocessor the command
flashupdate -s <FTPServer> -f <path to firmare .bin>
This asks for user/password on the ftp server, downloads and installs the firmware.
- After the update a resetsc restarts the service processor and uses
the new firmware and Hypervisor software.
- poweron starts the system and a console -f showed, that the reboot
cylce is gone.