Bug 4932 - timer problems with kernels => 2.6.12
Summary: timer problems with kernels => 2.6.12
Status: REJECTED UNREPRODUCIBLE
Alias: None
Product: Timers
Classification: Unclassified
Component: Realtime Clock (show other bugs)
Hardware: i386 Linux
: P2 normal
Assignee: john stultz
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-07-23 09:19 UTC by Georg Armbruster
Modified: 2005-08-08 18:20 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.13.rc*
Subsystem:
Regression: ---
Bisected commit-id:


Attachments

Description Georg Armbruster 2005-07-23 09:19:28 UTC
Distribution: Gentoo
Hardware Environment: various: P3-600 MHz, AMD64 3000Mhz, Celeron 800MHz
Software Environment: various, with/without X

Since updating to the 2.6.12-gentoo-r4 kernel, my machines clocks
go crazy. They go wrong for a couple of minutes per hours.
I tried the 2.6.13-rc3 kernel too, and 2.6.11 was running painlessly.

Unfortunately, I don't have a clue where to touch, so here a
dmesg of 2.6.11 (working good) and 2.6.12 (with this failure).

If you need any further information, I'll provide these happily.

Peace,
Georg

kernel 2.6.11-gentoo-r11: http://www.argeo.de/dmesg-2.6.11-gentoo-r11
kernel 2.6.12-gentoo-r5: http://www.argeo.de/dmesg-2.6.12-gentoo-r5


Reproducible: Always
Steps to Reproduce:
guess it was installing the current kernel
Actual Results:  
computers clock goes wrong

Expected Results:  
guess the time should be right :)

enina ~ # emerge info
Portage 2.0.51.22-r1 (default-linux/x86/2005.0, gcc-3.3.5, glibc-2.3.5-r0, 2.6.1
2-gentoo-r4 i686)
=================================================================
System uname: 2.6.12-gentoo-r4 i686 Intel(R) Pentium(R) 4 CPU 3.20GHz
Gentoo Base System version 1.6.12
dev-lang/python:     2.3.4-r1, 2.4.1-r1
sys-apps/sandbox:    1.2.10
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.5
sys-devel/binutils:  2.16.1
sys-devel/libtool:   1.5.18-r1
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86 ~x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-O3 -march=pentium4 -funroll-loops -fprefetch-loop-arrays -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3.4/env /usr/kde/3.4/share
/config /usr/kde/3.4/shutdown /usr/kde/3/share/config /usr/lib/X11/xkb /usr/lib/
mozilla/defaults/pref /usr/share/config /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O3 -march=pentium4 -funroll-loops -fprefetch-loop-arrays -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/Linux/
distributions/gentoo"
LANG="de_DE.utf8"
LC_ALL="de_DE.utf8"
LDFLAGS="-Wl,-O1"
MAKEOPTS="-j3"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="x86 X a52 aac acpi alsa apache2 apm arts avi berkdb bitmap-fonts cdr cjk cr
ypt cups curl dvd dvdr eds emboss encode esd evo fam foomaticdb fortran gdbm gif
 glx gnome gphoto2 gpm gstreamer gtk gtk2 imlib ipv6 java jpeg libg++ libwww mad
 mikmod mmx motif mozilla mp3 mpeg mysql ncurses nls ogg oggvorbis opengl oss pa
m pdflib perl png python quicktime readline samba sdl slang spell sse ssl tcpd t
iff truetype truetype-fonts type1-fonts unicode usb vorbis wmf xine xml2 xmms xv
 zlib userland_GNU kernel_linux elibc_glibc"
Unset:  ASFLAGS, CTARGET, LINGUAS, PORTDIR_OVERLAY
Comment 1 Andrew Morton 2005-07-23 17:33:21 UTC
Are you running the NTP daemon?   If so, could you try disabling
it, see what happens?
Comment 2 Georg Armbruster 2005-07-25 04:33:40 UTC
Hi Andrew,
you are right, it must have to do with the ntpd daemon;
disabling chrony and checking the time every hour using
ntpdate shows no differences in time over the last 24
hours.

Since the problem began nevertheless with the kernel
update, do you have any clue where to start searching?

Thanks for your help!

Peace,
Georg
Comment 3 john stultz 2005-07-25 10:52:28 UTC
Interesting, I've not heard of chrony (for reference - 
http://chrony.sunsite.dk/index.php ) before. Does it provide anything like the 
peerstats logging that the ntp.org ntpd provides? 
Comment 4 Georg Armbruster 2005-07-26 01:05:56 UTC
Hi John,
chrony is a lightweight ntp-daemon and worked quite reliable - 
that is, until recently :)

I'm not exactly sure how to read them, but I put the chrony
logs on the web for your reference:
http://www.relaxing-solutions.net/measurements.log
http://www.relaxing-solutions.net/statistics.log
http://www.relaxing-solutions.net/tracking.log

Regards,
Georg
Comment 5 john stultz 2005-07-27 16:16:20 UTC
Hmm. Odd. I'm not seeing anything in the logs that point to time drifting away
from the time server.

Just to clarify the problem, when you say the clocks go crazy and are wrong for
a couple of minutes per hour, do you mean the clock drifts (slows down or speeds
up) few minutes every hour, or that for a discrete amount of time each hour,
gettimeofday returns incorrect time values?
Comment 6 Georg Armbruster 2005-07-28 03:37:48 UTC
Hi John,
it's like you said, during one hour, the clock drifts away for
several minutes, it mostly sped up. Interestingly, during midday
more than in the morning and the evenings.

The really strange thing is: for the last couple of days,
the problem doesn't exist anymore.
I had this phenomenon on four pc's, and I switched back only
one of these to the old kernel.
The other three, I swear, were not touched at all concerning
configuration.

Yet for the last couple of days, they work normally again.

Which is, basically, something I cannot understand. (Since it
was me who quite often told folks that computers don't do
anything on their own. Boy was I wrong :=)

Well, basically, I cannot understand this. If you haven't any 
additional ideas, I propose closing
this bug until the problem comes up again. What do you think?
Comment 7 john stultz 2005-08-08 18:00:56 UTC
Georg: Hmm. Ok, I'll go ahead and close this. Please do reopen if you've had any
continuing issues.

Note You need to log in before you can comment on or make changes to this bug.