Rainer Zocholl schrieb: > #date --date "+10 min" +"%Y-%m-%d %H:%M:%S" > or less transparent: > # date --date "+10 min" +"%F %T" >/proc/acpi/alarm > # cat /proc/acpi/alarm > 2005-05-29 21:47:36 I can not use `date`-command because the busybox version has limited options and I don't want use the util-linux-`date`. [...] > - Is the hardware clock touched from the shutdown scripts? (with hwclock?) > -> comment out this lines > > Why? Touching the rtc disables the wakeup function on some motherboards. Therefore I'm storing the actually time at first and then I'm setting the wakeup time. > Hm, i saw that somewhere... > msi:/etc/init.d# mv /etc/rc0.d/K25hwclock.sh /etc/rc0.d/_K25hwclock.sh > msi:/etc/init.d# mv /etc/rc6.d/K25hwclock.sh /etc/rc6.d/_K25hwclock.sh > > too i disabled the alarm time in BIOS now > > And now: > it wakes up! (but the BIOS still shows a wrong time) > > Wow! > > > > Is that maybe the reason why nvram-wakeup sometimes need a reboot to work? Nvram-wakeup is messing arround with all the bits. I would recommend to do a CMOS-Reset. I testet ACPI-wakeup on four differend Motherboard, neither of them need a reboot. Alfred
