I just ran into a complete system hang during boot with my test machine, and believe it is caused by this bug. This was with full parallel booting (CONCURRENTY=makefile in /etc/default/rcS), and dependency based boot sequencing using Debian/testing. The system got stuck just after setting the system clock from the hardware clock twice in parallel. Changing the hwclock.sh dependency from mountdevsubfs to $local_fs and rerunning insserv to reorder the sequence solved the problem.
I did not get any messages from the kernel, and have no idea if there is a race condition or some other reason why running both hwclock and hwclockfirst in parallel caused the machine to hang. I am not even sure the cause was the hwclock* scripts, but modifying them solved my hang. Happy hacking, -- Petter Reinholdtsen -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
