initscripts save clock to hardware upon halt but do not load it on reboot.
This is fine when UTC is true in /etc/sysconfig/clock, because then the kernel
boots with the correct time already.
This is not fine when UTC is false, since then the clock is off after reboot by
whatever your timezone offset is.
Clock needs to be adjusted by init scripts on boot.
It *should* be happening already. See /etc/udev/rules.d/88-clock.rules.
Does "echo add > /sys/class/rtc/<device>/uevent" cause the clock to be set for you?
I don't have /sys/class/rtc. I am using 184.108.40.206-49.fc8. I have avoided using
the kernel Rawhide kernel for a long time because it regularly hangs on my
hardware (see ticket 283161) and because vmware server 1.0.4 wouldn't run on it
(at least not when I last tried).
I just tried to use kernel-2.6.25-0.121.rc5.git4.fc9.i686 to see if the hang
and/or vmware issue has been fixed, and it stopped in the middle of booting.
Didn't hang, exactly, since the keyboard was still active and ctrl-alt-delete a
few times got it to reboot, but it never got to the point of running init
scripts. Is that a known issue when upgrading from a pre-FC9 system? Any
Hm, I wonder if we need to require a new enough kernel. Still, there should be
*some* rtc device event that triggers this.
Changing version to '9' as part of upcoming Fedora 9 GA.
More information and reason for this action is here:
Whoops, found the error. Setting as a duplicate of the (later) bug with a patch.
*** This bug has been marked as a duplicate of 447019 ***