perfSONAR User Q&A and Other Discussion

[Jason Zurawski <>]

Hey Peter;

I would nuke the ESnet server too - NTP works best when the servers are 
within the same timezone/continent, it gives the algorithms a level playing 
field to choose from.  

A related note is that we are always looking to add new servers into our list 
around the world - if you know of 'open' clocks in the region that we can 
front-load into the list, that can be arranged.  

Thanks;

-jason

On May 23, 2013, at 11:31 AM, Peter van Heusden wrote:

> Thanks for the hint!
> 
> The default perfSonar setup uses a set of NTP servers on internet2.edu
> that are not accessible from my perfSonar server(s). I replaced them
> with accessible time servers and the output of ntpq -p is now:
> 
>     remote           refid      st t when poll reach   delay   offset 
> jitter
> ==============================================================================
> +zibbi.meraka.cs 238.72.153.243   2 u   48   64    7   72.510   -5.882  
> 8.591
> -firewall.sanbi. 41.216.193.18    3 u   47   64    7    0.094   -2.696  
> 7.960
> -ntp.mtnbusiness 41.73.40.11      3 u   47   64    7    5.543    4.661  
> 7.965
> *chronos.es.net  .PPS.            1 u   46   64    7  232.200   -6.152  
> 8.662
> +ntp2.is.co.za   146.64.58.41     2 u   44   64    7    5.475   -3.928  
> 6.901
> 
> 
> The restarting is still continuing though. Can I expect this to
> stabilise in the next while?
> 
> Peter
> 
> On 23/05/2013 20:27, Alan Whinery wrote:
>> Peter -
>> 
>> It might be useful to look at your NTP status. I note that you used the
>> work "internal network". Does that mean either significant firewalling,
>> or no routing to the outside?
>> 
>> You need to have several NTP servers peered with your box, and bwctl
>> will not start until it sees NTP as "synced".
>> 
>> What are the servers defined, and what is the output of "ntpq -p" ?
>> 
>> 
>> 
>> On 5/23/2013 7:31 AM, Brian Tierney wrote:
>>> Hi Peter.
>>> 
>>> That is normal until the clock gets synced, which can take several hours.
>>> It should have stopped by now.
>>> 
>>> It really should back off before it tries again, but that is not the case 
>>> at the moment.
>>> 
>>> 
>>> On May 23, 2013, at 7:05 AM, Peter van Heusden 
>>> <>
>>>  wrote:
>>> 
>>>> Hi there
>>>> 
>>>> I'm trying to use perfSonar for testing on our internal network.
>>>> Unfortunately it is stuck in a loop where bwctl is being continuously
>>>> restarted. This is perfSonar 3.3rc4 installed on CentOS 6.4. The message
>>>> I am seeing in /var/log/messages is:
>>>> 
>>>> May 23 16:02:26 ps bwmaster.pl[2849]: Restart
>>>> bwctl->KNOPPIX->172_16_203_14:! at
>>>> /opt/perfsonar_ps/perfsonarbuoy_ma/bin/bwmaster.pl line 688.#012
>>>> May 23 16:02:26 ps bwmaster.pl[2849]: Restart
>>>> bwctl->KNOPPIX->PS_SANBI_AC_ZA:! at
>>>> /opt/perfsonar_ps/perfsonarbuoy_ma/bin/bwmaster.pl line 688.#012
>>>> May 23 16:02:26 ps bwmaster.pl[2849]: Restart
>>>> bwctl->PS_SANBI_AC_ZA->KNOPPIX:! at
>>>> /opt/perfsonar_ps/perfsonarbuoy_ma/bin/bwmaster.pl line 688.#012
>>>> May 23 16:02:26 ps bwmaster.pl[2849]: Restart
>>>> bwctl->172_16_203_14->KNOPPIX:! at
>>>> /opt/perfsonar_ps/perfsonarbuoy_ma/bin/bwmaster.pl line 688.#012
>>>> May 23 16:02:26 ps bwmaster.pl[2849]: Restart
>>>> bwctl->KNOPPIX->172_16_203_14:! at
>>>> /opt/perfsonar_ps/perfsonarbuoy_ma/bin/bwmaster.pl line 688.#012
>>>> 
>>>> This perfSonar was installed from RPM.
>>>> 
>>>> Does anyone know how to debug / solve this error?
>>>> 
>>>> Thanks!
>>>> Peter
>>>> 
>