Hi Hetz,
On Monday 23 April 2007 02:42, Hetz Ben Hamo wrote:
> Shimi,
>
> I will suggest something that you might find odd, but here goes:
>
> * Boot Windows * :)
>
> If you have a partition with windows on your machine, boot into
> windows, and write a small batch script that will show the time every
> few seconds, and see if you have the same issue as you have now. Make
> sure you use either win2k, XP or 2003, NOT 95/98/ME as they don't
> support SMPT.
This is a server box. Windows never touched it since it went out of its'
package... (nor do I see a reason for it to do that :)). I am also unsure if
Windows would "see" this problem, because I think their timekeeping is
different (live sync with the RTC if I am not wrong). The fact is that if I
used 'hwclock' WITH the SMP running, I always got the RIGHT time. So the
board's RTC is going well. It just appears that every CPU runs its own
timekeeping (and I might be saying something very silly here, but I never did
read the relevant kernel source code, so I can only guess). Ticks, Jiffies,
CPUs synchronization, alternating clock speeds (Intel SpeedStep, Frequency
Scaling) - can all cause this - and what I am seeing might just be that
"date" was assigned to a different CPU every time (perhaps I should have
forced an assignment to each CPU individually and try?)
>
> If you don't have windows, you can always create a "LiveCD" of windows
> quite easily: take any windows machime, download and run BartPE and
> use it to create a bootable LiveCD. Boot this CD and do the same
> (simple batch script to show the time)
>
> If the problem appears in Windows the same as in your Linux, then you
> have a faulty board. It could be either a BIOS bug, faulty clock, or
> something else is fucked on the board (such as bad Crystal
> oscillattor).
>
> If it only appears on Linux after your Windows testing, perhaps you
> would like to test any Linux based liveCD and see if the problem
> appears there as well. If it is - it's a kernel issue. If not, well,
> bad Gentoo kernel (highly unlikely).
>
All that, would, unfortunately, make me take another trip to the ISP that
hosts the server ;) I've wasted enough precious [down]time on this, so I am
looking for solutions that can be tried remotely at 4am (reconfiguring the
kernel is OK...) - mostly because those are the only _valid_ solutions - I
don't care if it works well in Windows, as I am not running this machine on
Windows, and same goes for a LiveCD (not going to run a server from a
LiveCD). And if it's indeed a hardware problem, there must be a way to see
that on a Linux system - and then again - I cannot fix it on that case, and
go explain the company we bought the computer from that problem, as the
machine "appears to be working fine". So I am still open for ideas on where
to look for proofs/fixes :)
Thanks for the suggestions though :)
-- Shimi
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
