http://bugzilla.novell.com/show_bug.cgi?id=584484
http://bugzilla.novell.com/show_bug.cgi?id=584484#c13
Jiří Suchomel
Do you have meant comment #5 ... nevertheless my guess was that NTP destroys the setup as it does not know about localtime im CMOS.
Read comment #5 and the correction in comment #8 as simple cook book to get it right.
The problem is, I do not see comment #5 (+8) as "simple cook book". I believe we had a working solution in previous versions and now it's getting too complicated and I just do not see what current YaST is doing wrong. So, I'll try to summarize once again what YaST is doing in case user says CMOS is in localtime: 1. Set the time zone (/usr/sbin/zic -l <timezone>) 2. /sbin/hwclock --systz --localtime --noadjfile && touch /dev/shm/warpclock 3. This is done before the ntp call, so ntpd should know the time zone. 4. Now, ntpd is called 5. And the (NTP) changed system time is saved back to HW clock by "/sbin/hwclock --systohc --localtime" So, does NTP destroy the setup? It changes the system time, but that one should be correctly written to bios using the hwclock call in 5, and the hwclock has the knowledge about the time format in BIOS... I'm realizing that /etc/sysconfig/clock is not saved before that NTP call, so maybe this is the problem? But do I really have to call mkinitrd when hwclock should know about the time format, because of --localtime argument? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.