From: Logical on
Hi

Been fiddling with ntpd on my slackware 10.0 box for a while now but it
constantly seems to be freezing and allowing the time to drift
significantly.

root[at]mythtv:/home/mythtv# uname -a
Linux mythtv 2.6.12.3-mythtv #3 Tue Jul 26 23:03:30 EST 2005 i686 unknown
unknown GNU/Linux

---
root[at]mythtv:/home/mythtv# cat /etc/ntp.conf
# Sample /etc/ntp.conf: Configuration file for ntpd.
#

logfile /var/log/ntpd.log
server ntp.iinet.net.au
server ntp.nsw.westnet.com.au
server time.deakin.edu.au
server ntp0.cs.mu.OZ.AU
server ntp1.cs.mu.OZ.AU
server au.pool.ntp.org

fudge 127.127.1.0 stratum 10

driftfile /etc/ntp/drift
multicastclient # listen on default 224.0.1.1
broadcastdelay 0.008

#
# Keys file. If you want to diddle your server at run time, make a
# keys file (mode 600 for sure) and define the key number to be
# used for making requests.
# PLEASE DO NOT USE THE DEFAULT VALUES HERE. Pick your own, or remote
# systems might be able to reset your clock at will.
#
#keys /etc/ntp/keys
#trustedkey 65535
#requestkey 65535
#controlkey 65535

# Don't serve time or stats or trust anyone else by default (more secure)
restrict default noquery notrust nomodify
# Trust ourselves. :-)
restrict 127.0.0.1

---

I am starting ntpd up from /etc/rc.d/rc.local with:

# start up ntpd
ntpdate dns.iinet.net.au
/usr/sbin/ntpd

When ntpd first starts up, the log file registers:
6 Jan 01:10:53 ntpd[2597]: frequency initialized 0.000 PPM from
/etc/ntp/drift

but there are no log entries after that date.

ntpd seems to be running still:

root[at]mythtv:/etc/rc.d# ps aux | grep ntpd
root 2620 0.0 0.7 3640 3636 ? SLs Jan06 1:23
/usr/sbin/ntpd
root 10970 0.0 0.1 1772 624 pts/1 S+ 19:29 0:00 grep ntpd


is SLs it's normal running state? I'm not sure...
From: Grant on
On Sat, 14 Jan 2006 15:50:29 +1100, Logical <me(a)privacy.net> wrote:

>Been fiddling with ntpd on my slackware 10.0 box for a while now but it
>constantly seems to be freezing and allowing the time to drift
>significantly.
>
>root[at]mythtv:/home/mythtv# uname -a
>Linux mythtv 2.6.12.3-mythtv #3 Tue Jul 26 23:03:30 EST 2005 i686 unknown
>unknown GNU/Linux

That's a very old kernel :(

I show my .conf as it has some references:

# cat /etc/ntp.conf
# /etc/ntp.conf for slackware on deltree 2004-11-18
#
# Configuration, see
# http://twiki.ntp.org/bin/view/Servers/NTPPoolServers
# http://ntp.isc.org/bin/view/Support/ConfRestrict

# configure local clock as dummy refclock, making sure any clients
# know that it is not a high quality clock with stratum = 10
# Note that setting a dummy refclock seems a bad idea (2005-01-08)
#server 127.127.1.0
#fudge 127.127.1.0 stratum 10

# drift and log files
driftfile /etc/ntp/drift
logfile /var/log/ntp_log

# Access control and server selection:

# set default restrictions
restrict default kod nomodify notrap nopeer noquery

# select time servers
server au.pool.ntp.org
server us.pool.ntp.org
server ca.pool.ntp.org
server uk.pool.ntp.org

# restrict local server users less
restrict 127.0.0.1
restrict 192.168.1.0 mask 255.255.255.0 nomodify notrap nopeer
restrict 192.168.2.0 mask 255.255.255.0 nomodify notrap nopeer
# end

>ntpd seems to be running still:
>
>root[at]mythtv:/etc/rc.d# ps aux | grep ntpd
>root 2620 0.0 0.7 3640 3636 ? SLs Jan06 1:23
>/usr/sbin/ntpd
>root 10970 0.0 0.1 1772 624 pts/1 S+ 19:29 0:00 grep ntpd

# ps aux | grep ntpd
root 315 0.0 2.7 3564 3564 ? SLs 16:25 0:00 /usr/sbin/ntpd

SLs looks normal.

# ntpq -pn
remote refid st t when poll reach delay offset jitter
==============================================================================
203.12.160.2 128.250.37.2 2 u 35 64 3 67.584 0.004 0.174
220.233.126.241 203.12.160.2 3 u 34 64 3 106.858 11.430 5.163
192.83.249.31 204.123.2.72 2 u 33 64 3 243.986 9.281 4.893
64.235.218.180 199.212.17.15 3 u 34 64 3 312.490 12.533 10.187
212.13.207.101 194.81.227.227 2 u 32 64 3 407.372 6.152 5.093

Erk! I been testing e100 drivers, box only 12 minutes from last boot,
ntpd not yet locked in.

What CPU, SMP? Some x86-64 still having timer problems due to drift
greater than delta rate ntpd can correct by. Also check your timing
sources, many .au servers been turned off to public access due to abuse.

Grant.
--
Cats are smarter than dogs. You can't make eight cats pull
a sled through the snow.
From: Mark Post on
On Sat, 14 Jan 2006 15:50:29 +1100, Logical <me(a)privacy.net> wrote:

> Hi
>
>Been fiddling with ntpd on my slackware 10.0 box for a while now but it
>constantly seems to be freezing and allowing the time to drift
>significantly.

The "ntpq -p" command will tell you if you're actually getting connected and
using the time servers you've specified in /etc/ntp.conf. If not, you
probably need to review the "restrict" parameters for each of them. For my
time server, I specify "mask 255.255.255.255 nomodify notrap noquery" for
them.


Mark Post

From: Logical on
Mark Post wrote:
> The "ntpq -p" command will tell you if you're actually getting connected and
> using the time servers you've specified in /etc/ntp.conf. If not, you
> probably need to review the "restrict" parameters for each of them. For my
> time server, I specify "mask 255.255.255.255 nomodify notrap noquery" for
> them.

root(a)mythtv:/home/mythtv# ntpq -p
remote refid st t when poll reach delay offset
jitter
==============================================================================
dns.iinet.net.a .RSTR. 16 u - 64 0 0.000 0.000
4000.00
fa0-0-1.syd-cor .RSTR. 16 u - 128 0 0.000 0.000
4000.00
sol.ccs.deakin. .RSTR. 16 u - 64 0 0.000 0.000
4000.00
murgon.cs.mu.OZ .RSTR. 16 u - 64 0 0.000 0.000
4000.00
ntp1.cs.mu.OZ.A .RSTR. 16 u - 64 0 0.000 0.000
4000.00


That's my ntpq -p output, the jitter value seems significantly high?

I tested each of those servers with ntpdate previously and that allowed
me to sync the clock correctly.

I will look at upgrading the kernel also.

It's a AMD 2100XP, on a Abit NFS2-7 motherboard (nforce2 chipset) with
512 ram.

no SMP, not a 64bit chip.
From: Thomas Ronayne on
Logical wrote:
> Hi
>
> Been fiddling with ntpd on my slackware 10.0 box for a while now but it
> constantly seems to be freezing and allowing the time to drift
> significantly.
>
> is SLs it's normal running state? I'm not sure...
Yes, it is.

Couple of hints:

* /etc/ntp/drift should contain a floating point number -- if it is
not, put 0.0 in it to start with.
* You should have these first
server 127.127.1.0 # local clock
fudge 127.127.1.0 stratum 10
(so, when you lose your internet connection, ntpd can sync to the
local clock)
* You really only need three outside servers (doesn't matter, but
three's usually good); use servers with the lowest ping time
* Try making these entries the default values
multicastclient # listen on default 224.0.1.1
broadcastdelay 0.008

Something else you can do is keep statistics (which, frankly, aren't too
useful until you get synchronized) -- stick these lines in your ntp.conf
file

#
# Keep statistics
#
statsdir /var/log/ntpstats/
filegen peerstats file peerstats type day enable
filegen loopstats file loopstats type day enable
filegen clockstats file clockstats type day enable

and

mkdir /var/log/ntpstats

Copy the files in /usr/doc/ntp-4.2.0/scripts/stats to /var/log/nptstats,
you'll need to edit summary.sh, though -- a working version is listed below.

For some reason that I've never been able to discern, NTP didn't work
too well in Slackware 10.0 (you might want to think about updating to
10.2), but, once you get it started it'll go forever. Try running

/usr/sbin/ntpdate -v ntp-1.mcs.anl.gov ntp-2.mcs.anl.gov

(using two servers close to you!) in your rc.ntpd program (or on the
command line) and see what is reported -- don't forget to stop any
running ntpd before you run it, though.

The final thing is making sure your clock is set (with ntpdate) before
you start ntpd -- NTP will not sync if the clock is too far off. May
seem silly but check your timezone and make sure you're not off an hour
or so? My rc.ntpd looks like this (I specify the log file in it rather
than in ntp.conf, six of one, half dozen of the other):

#!/bin/sh
#
# start the Network Time Protocol
#
if [ ${1} = "start" ]
then
# set the clock from a public server
echo "Setting clock from ntp-1.mcs.anl.gov..."
/usr/sbin/ntpdate -v ntp-1.mcs.anl.gov ntp-2.mcs.anl.gov
sleep 2
echo "Starting Network Time Protocol daemon..."
>/tmp/ntp.log
/usr/sbin/ntpd -l /tmp/ntp.log
elif [ ${1} = "stop" ]
then
echo "Stopping Network Time Protocol daemon..."
pid=`/bin/ps -e | /usr/bin/grep ntp |\
/usr/bin/sed -e 's/^ *//' -e 's/ .*//'`
if [ "${pid}" != '' ]
then
kill ${pid}
fi
fi

When it's running (after 15 minutes or so) you should see something like
this (LOCAL is there for when the internet goes away for whatever reason):

ntpq -p
remote refid st t when poll reach delay
offset jitter
==============================================================================
LOCAL(0) LOCAL(0) 10 l 46 64 377 0.000
0.000 0.001
*adsl.remco.org .PG0A. 1 u 757 1024 377 131.632
-13.555 0.586
+salukes.opensou 80.127.4.179 2 u 865 1024 377 125.241
-14.664 0.693
+ensb.cpsc.ucalg 192.12.19.20 2 u 871 1024 377 60.771
-16.972 0.636

Hope some of this helps a little.

--
Everything works -- if you let it.

 |  Next  |  Last
Pages: 1 2 3 4 5
Prev: Driver for Buffalo LPC PCM CLX
Next: Apple iPod