[ntp:questions] ntpd wedged again

Danny Mayer mayer at ntp.org
Sun Feb 26 03:37:29 UTC 2012


On 2/12/2012 4:55 PM, A C wrote:
> I'm not exactly sure what happened but I may have caught the system in
> the act of trying to run away.  Below is the last two lines from the
> peers log for one of the peers and below that is the output of ntpq
> showing what happened to that peer immediately after.  Note the offset.
>  This somewhat matches what I had been seeing previously where my system
> would be reporting sane numbers and then very suddenly go haywire.
> 
> 55969 77832.471 67.23.181.241 9324 0.006359155 0.081262776 0.001634487
> 0.008837966
> 55969 77897.557 67.23.181.241 9324 399326.864663492 0.000122070
> 0.001346708 399326.861631493
> 
> 
>  remote           refid      st t when poll reach   delay   offset jitter
> 
> ==============================================================================
>67.23.181.241   128.4.1.1        2 u   53   64  377    0.122  3993268 3993268

This shows that this server is contacting rackety. Out of curiosity I
looked at 67.23.181.241 and it turns out that it's running ntpd 4.2.2p1
which is rather ancient:
> version="ntpd 4.2.2p1 at 1.1570-o Fri Nov 18 13:21:21 UTC 2011 (1)"?,
> processor="x86_64", system="Linux/2.6.18-274.17.1.el5.centos.plus",

It could well be that rackety is sending KOD packets and this server is
not recognizing them as such. rackety has a heavy load under normal
circumstances so it may well do so. When it does so the ntp timestamps
are going to be totally wrong, deliberately, so they should have been
discarded immediately. I don't remember whether or not KOD had been
introduced back when 4.2.2 was shipped so it may not be ignoring it when
it should.

I suggest you pick a different server with a more modern version of ntpd
and see if that stabilizes the situation.

Danny


More information about the questions mailing list