perfSONAR User Q&A and Other Discussion

[Andrew Lake <>]

On Apr 16, 2014, at 2:50 PM, Michael Tiernan 
<>
 wrote:

> My question(s):
> 1) For entries in the "Inactive Tests" list. Does perfsonar continue to 
> check to see if these T/F pairs work even if they're not able to collect 
> data or is that list of "Inactives" mean, for lack of a better term, we 
> gave up and won't try again?
> 
> 2) If the answer above is "we give up", how can I (or can I) trigger a 
> retry on these sites?
> 
> 3) Shouldn't the reloading of a mesh config cause these targets to at least 
> get tried again? If so, wouldn't it be prudent to put in that table "last 
> successful contact" and/or "last attempt made"?

The answer to all three is it never "gives-up". Inactive means they have not 
seen any data for the specified period of time. For owamp its 30 minutes and 
bwctl its 7 days. If you look under the "Active Tests" and "Inactive Tests" 
it tries its best to explain this. If it hasn't collected any data, what that 
means is that it has tried and the test does not complete for some reason. 
There is a whole bunch of reasons a test could not complete. A good starting 
point is often /var/log/perfsonar/owamp_bwctl.log.

> 
> 4) How is the "Bidirectional" flag chosen and maybe why? (I believe this 
> last question to be something I should've been able to learn from the docs 
> but haven't found yet so please feel free to point me at the FM and tell me 
> where to RT(FM).

Bidirectional means it has been able to collect data in both directions (i.e. 
from source to destination and destination to source). If it is no, that 
means the test is failing in one direction for some reason. 

> Lastly, a general question that I got asked but had no clue on. Is there a 
> map (i.e. Google maps) where all the sites are listed so one can see where 
> the relative pathways are that traffic is being tested across? (Yes, I know 
> it's not a straight line but it seemed a reasonable question for people to 
> know.)

The page http://stats.es.net/ServicesDirectory/ has a map of all the 
services. It does not map where measurements are being run, but that's an 
interesting idea.

Hope that helps,
Andy


> 
> Thanks in advance for any advice/answers.
> 
> -- 
>  <<  MCT>>    Michael C Tiernan 
> xmpp:
>  +1 (617) 324-9173
>  MIT - Laboratory for Nuclear Science - http://www.lns.mit.edu
>  High Perf Research Computing Facility at The Bates Linear Accelerator
>    Please avoid sending me MS-Word or MS-PowerPoint attachments.
>    See http://www.gnu.org/philosophy/no-word-attachments.html
> 
>