[ntp:questions] no drift-file on 2008 R2 vps and the time diff. is getting bigger and bigger?

Harlan Stenn stenn at ntp.org
Sun Sep 14 05:48:23 UTC 2014


gooly writes:
> Am 13.09.2014 23:47, schrieb David Woolley:
> > On 13/09/14 15:51, gooly wrote:
> >> And this would not explain why ntp.drift is not created?

On some systems (as I recall), a non-existent drift file will not be
created - you might try creating an empty one and seeing what happens.

If you run at debug level 1, you should also see messages about the
drift file update process.

> > It's not created because synchronisation has never been acquired.  The
> > lack of ntp.drift is a secondary effect.
>
> ok - but why the "synchronisation has never been acquired"?

run 'ntpq -p' and see what happens.  You'll want to see a non-zero
"reach" and values other than 0 for various other things.

> And what do I have to do so that not only at (re-) start ntp sets the 
> time of the vps?

ntpd should do this for you, especially if you start it with the -g
option.

> I don't see any error message and no notice that e.g. a try to change 
> the time failed - the event reader (eventvwr) shows only that ntp has 
> been started correctly?
> 
> The time difference increases ~7 sec. every 15 minutes.

If yuor system is *slowing* by 7 sec every 15 minutes, I suspect you are
losing clock interrupts.

At that rate ntpd will never be able to keep your clock sync'd.

> PS: may be this helps?
> > C:\Windows\System32>ntpq -p
> >      remote           refid      st t when poll reach   delay   offset  jit
> ter
> > ===========================================================================
> ===
> >  janetzki.eu     5.9.80.113       3 u   66   64  377    9.893  251511. 2370
> .61
> >  net2.uni-paderb .DCF.            1 u   12   64  377   18.546  251914. 2338
> .44
> >  2a01:4f8:110:51 178.63.61.67     3 u   34   64  377   10.463  251768. 2344
> .13
> >  16-164-ftth.ons 193.67.79.202    2 u   60   64  377   21.001  251558. 2356
> .75
> >  y.ns.gin.ntt.ne 145.238.203.14   2 u   62   64  377   20.690  251544. 2374
> .56
> >  ts2.aco.net     .PPS.            1 u   57   64  377   16.820  251589. 2379
> .13
> >  ts1.univie.ac.a .PPS.            1 u   10   64  377   17.828  251940. 2333
> .10
> >  mail.somenet.or 217.19.37.27     2 u   46   64  377   16.373  251679. 2382
> .17
> >  palmers.nobody. 193.171.23.163   2 u   29   64  377   17.824  251802. 2323
> .55
> >  cl-2008.mbx-01. 195.58.160.5     3 u    5   64  377   37.461  251981. 2372
> .91
> >  trust.nobody.at 131.130.251.107  2 u   61   64  377   19.276  251548. 2361
> .68
> >  2001:a60::123:1 212.18.1.106     2 u   24   64  377   12.096  251845. 2337
> .77
> >  ptbtime1.ptb.de .PTB.            1 u   16   64  377   17.064  251890. 2327
> .26
> >  ptbtime2.ptb.de .PTB.            1 u    8   64  377   16.556  251957. 2364
> .28
> >  ptbtime3.ptb.de .PTB.            1 u   38   64  377   65.845  251757. 2411
> .84
> >  foxtrot.zq1.de  235.106.237.243  3 u   61   64  377    9.426  251546. 2364
> .88
> >  ntps1-0.eecsit. .PPS.            1 u   12   64  377   17.605  251919. 2347
> .01
> >  time.fu-berlin. .PPS.            1 u   63   64  377   17.566  251534. 2357
> .45
> >  ns2.probe-netwo 131.188.3.221    2 u   13   64  377    5.397  251915. 2360
> .15
> >  zeit.fu-berlin. .PPS.            1 u   27   64  377   18.060  251823. 2351
> .31
> 
> and
> > C:\Windows\System32>ntpq -crv
> > associd=0 status=c018 leap_alarm, sync_unspec, 1 event, no_sys_peer,
> > version="ntpd 4.2.6p5 at 1.2349-o Jul 30 11:55:08 (UTC+02:00) 2012  (2)",
> > processor="x86", system="Windows", leap=11, stratum=4, precision=-21,
> > rootdelay=30.116, rootdisp=2726.363, refid=5.9.67.110,
> > reftime=d7bf26ea.de20f8a0  Sat, Sep 13 2014 22:26:18.867,
> > clock=d7bf9c38.52662d5e  Sun, Sep 14 2014  6:46:48.321, peer=0, tc=6,
> > mintc=3, offset=0.000, frequency=500.000, sys_jitter=0.000,
> > clk_jitter=0.000, clk_wander=0.143

The frequency of 500ppm indicates ntpd has "hit its limit" of how much
adjusting it can do.  leap=11 means ntpd is not synchronized.

The first thing to fix is the problem with clock frequency.

H


More information about the questions mailing list