[ntp:questions] Cluster NTP configuration

Steve Tryon x77ssiwdo at yahoo.com
Mon Jun 20 21:22:34 UTC 2011


Thanks to all for the responses.

I'm going to provide some background information in order to get confirmation of a root cause analysis.  Please bear with me.

The reason I ask about large corrections is a scenario that I saw with a configuration that we originally tried.  In essence, each server in the cluster was configured the same:
- clients of 0.us.pool.ntp.org and 1.us.pool.ntp.org
- peers with all nodes in the cluster
- using local clock as stratum 10 source

We saw some significant timewarps in our application and this is what we saw in our ntp log file:

May 29 04:59:20 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 05:07:57 sprint-ac-4 ntpd[12316]: synchronized to 192.43.244.18, stratum 1
May 29 05:09:58 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 05:27:05 sprint-ac-4 ntpd[12316]: synchronized to 192.43.244.18, stratum 1
May 29 06:12:23 sprint-ac-4 ntpd[12316]: time reset +28.770117 s
May 29 06:16:40 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.203, stratum 12
May 29 06:16:12 sprint-ac-4 ntpd[12316]: time reset -28.769913 s
May 29 06:20:31 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 06:21:35 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.201, stratum 13
May 29 06:23:39 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.202, stratum 11
May 29 06:33:26 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.203, stratum 12
May 29 06:33:26 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.202, stratum 11
May 29 06:35:35 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.201, stratum 12
May 29 06:35:35 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.202, stratum 11
May 29 06:46:17 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 06:54:56 sprint-ac-4 ntpd[12316]: synchronized to 192.43.244.18, stratum 1
May 29 07:20:27 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 07:29:03 sprint-ac-4 ntpd[12316]: synchronized to 192.43.244.18, stratum 1
May 29 07:35:19 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 07:52:23 sprint-ac-4 ntpd[12316]: synchronized to 192.43.244.18, stratum 1
May 29 08:22:51 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 08:39:18 sprint-ac-4 ntpd[12316]: synchronized to 192.43.244.18, stratum 1
May 29 09:14:02 sprint-ac-4 ntpd[12316]: time reset +30.134054 s
May 29 09:18:21 sprint-ac-4 ntpd[12316]: synchronized to LOCAL(0), stratum 10
May 29 09:21:36 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.201, stratum 13
May 29 09:24:48 sprint-ac-4 ntpd[12316]: synchronized to X.Y.Z.202, stratum 11
May 29 09:34:04 sprint-ac-4 ntpd[12316]: time reset -30.133806 s

where X.Y.Z represents the subnet of the peers.

I've done some reading now and it seems logical to me that the stratum 11,12,13 logs as well as the large corrections are a result of our including the local clocks as references.  Can you confirm?

Thanks,

--- On Mon, 6/20/11, Chris Albertson <albertson.chris at gmail.com> wrote:

From: Chris Albertson <albertson.chris at gmail.com>
Subject: Re: [ntp:questions] Cluster NTP configuration
To: "Chuck Swiger" <cswiger at mac.com>
Cc: "Steve Tryon" <x77ssiwdo at yahoo.com>, questions at lists.ntp.org
Date: Monday, June 20, 2011, 3:44 PM

On Mon, Jun 20, 2011 at 1:03 PM, Chuck Swiger <cswiger at mac.com> wrote:
> On Jun 20, 2011, at 6:59 AM, Steve Tryon wrote:
>> Synchronization to a tenth of a second is fine for my purposes.  All servers will be running RH 5.5.
>
> You won't need to do anything special to get better than 10ms sync accuracy, much less 100ms.

I'll second that!  Almost anything you do will work at that level.
NTP normally runs at a level of precison 10X to 1000X better than you
need.

Chris Albertson
Redondo Beach, California



More information about the questions mailing list