check_ntp_peer unreliable on macs

Andrew Davis nccomp at gmail.com
Wed Mar 25 19:06:22 CET 2009


I encountered this too. My solution was a script in /usr/local/scripts 
called by cron that stops the NTP daemon, runs "ntpdate <local_server>" 
twice, then restarts the NTP daemon. This runs in cron every 2 hours and 
seems to keep things in sync...

  A. Davis
  Email:     nccomp at gmail.com

  "There is no limit to what a man can accomplish
   if he doesn't care who gets the credit." - Ronald Reagan



Peter Doherty wrote:
> On Mar 13, 2009, at 5:25 PM, Keith Erekson wrote:
>
>   
>> I found this in my mailing list archives, while looking for  
>> information about check_ntp_peer. As far as I can tell, nobody ever  
>> answered you...
>>
>> I was just looking into this exact problem. If you check the verbose  
>> output, you will probably see something like this:
>>
>> 0 candiate peers available
>> warning: no synchronization source found
>> warning: LI_ALARM bit is set
>>
>> I do get valid output from "ntpq -p hostname", however.
>>
>> Apparently, the problems with OS X's NTP are well-known and  
>> documented. For example,
>>
>> http://knol.google.com/k/dirk-h-schulz/time-synchronization-ntp-on-mac-os-x/2bcee0ik2900p/18#
>> http://support.ntp.org/bin/view/Support/KnownOsIssues#Section_9.2.5
>>
>> As a way around this, I thought I would just use check_ntp_time, to  
>> compare the xserve's clock against that of the nagios box. However,  
>> no luck there either:
>>
>> sending request to peer 0
>> response from peer 0: offset -0.9300264975
>> sending request to peer 0
>> response from peer 0: offset -0.9299369976
>> sending request to peer 0
>> response from peer 0: offset -0.9299154976
>> sending request to peer 0
>> response from peer 0: offset -0.9298709977
>> discarding peer 0: stratum=0
>> overall average offset: 0
>> NTP CRITICAL: Offset unknown|
>>
>>
>> It seems that OS X is responding as a stratum 0 server, which is a  
>> no-no.
>>
>> Also, while fiddling with check_ntp_peer, I noticed that it doesn't  
>> seem to accept a port (-p or --port), as the help output suggests it  
>> should be able to. Am I crazy?
>>
>> -Keith
>>     
>
>
> Yeah, after a little more diagnostic work I eventually concluded that  
> it was just OS X's implementation of NTP that is just broke.  It seems  
> to be in sync for a while, then it just forgets it for a while, and  
> eventually, maybe it'll sync up again.
>
> Maybe they'll fix that for 10.6 this year.
>
> --Peter
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Nagios-users mailing list
> Nagios-users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/nagios-users
> ::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
> ::: Messages without supporting info will risk being sent to /dev/null
>   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.monitoring-lists.org/archive/users/attachments/20090325/4e540c74/attachment.html>
-------------- next part --------------
------------------------------------------------------------------------------
-------------- next part --------------
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null


More information about the Users mailing list