perfSONAR User Q&A and Other Discussion

[Andrew Lake <>]

Hi,

The correct MA URL is http://nautilus.sr.unh.edu/esmond/perfsonar/archive/. 
Do NOT add ?format=json to your maddash configuration, that's just something 
needed in your browser, not by clients like maddash. You will need to wait a 
few hours until your checks run again before seeing any change to the 
dashboard. An ACL seems to be blocking me from looking at your archive, but 
assuming it has data you may just need to wait for maddash to refresh.

In general, if you manage your maddash,yaml file by hand, then change the 
format of the ma_url to http://<hostname>/esmond/perfsonar/archive/ where 
<hostname> is the name of the host running the MA. Depending on how your 
dashboard is setup you may be able to use %row or %col variables to limit the 
number of ma_urls you need to specify. For example, 
http://%row/esmond/perfsonar/archive/. The use of %row and %col is no 
different than before, so really it should just be a matter of replacing the 
old port and path in the URL with http://%row/esmond/perfsonar/archive/.

For the graphURL, you can use the new graphs with something like 
http://maddash-server-name/serviceTest/graphWidget.cgi?url=%maUrl&dest=%col&source=%row.
 You may want to swap %row and %col depending on what you want labelled as 
the forward and reverse direction. You do not need to specify length. 


http://stats.es.net/serviceTest/graphWidget.cgi?url=http://albq-pt1.es.net:8085/esmond/perfsonar/archive&source=198.129.254.6&dest=198.124.252.141&url=http://albq-pt1.es.net:8085/esmond/perfsonar/archive&source=198.129.254.6&dest=198.124.252.154&url=http://albq-owamp.es.net:8085/esmond/perfsonar/archive&source=198.129.254.218&dest=198.124.252.141&url=http://albq-owamp.es.net:8085/esmond/perfsonar/archive&source=198.129.254.218&dest=198.124.252.154

When specifying multiple hosts for the graphs, it can get cumbersome to 
maintain. You may use the MeshConfig software to automatically generate URLs 
like the above. MeshConfig configuration references are in the midst of being 
created that should help with this. 

Thanks,
Andy





On Oct 15, 2014, at 3:36 AM, Mark Maciolek 
<>
 wrote:

> hi,
> 
> My Dashboard is also all "unable to retrieve data"
> 
> https://nautilus.sr.unh.edu/maddash-webui/
> 
> using:
> 
> http://nautilus.sr.unh.edu/esmond/perfsonar/archive/
> 
> Sorry, not implemented yet. Please append "?format=json" to your URL.
> 
> http://nautilus.sr.unh.edu/esmond/perfsonar/archive/?format=json
> 
> this gives me web page filled with text.
> 
> Another script I need to run?
> 
> Mark
> 
> 
> 
> On 10/14/2014 4:52 PM, Andrew Lake wrote:
>> Hi Jeremy,
>> 
>> You need to change the url to 
>> http://%row/esmond/perfsonar/archive/
>> 
>> 
>> That should bring everything back, sorry for the inconvenience.
>> 
>> Thanks,
>> Andy
>> 
>> 
>> On Oct 14, 2014, at 4:27 PM, Jeremy Palmer 
>> <>
>>  wrote:
>> 
>> 
>>> Our dashboard completely stopped working after upgrading to 3.4. All 
>>> boxes in the grid are now showing the orange "Unable to retrieve data" 
>>> color. I'm getting a ton of errors like the following in the 
>>> maddash-server.netlogger.log file:
>>> level=INFO ts=2014-10-14T08:20:06.313974Z 
>>> event=maddash.RunCheckJob.execute.runCheck.end 
>>> guid=0e7c7ee5-840b-487f-8e26-c662648f45ad resultMsg=" Unable to contact 
>>> MA. Please check that the MA is running and the URL is correct."
>>> 
>>> The maURL in my maddash.yaml file is:
>>> maUrl: default: 
>>> "http://%row:8085/perfSONAR_PS/services/pSB";
>>> 
>>> 
>>> However, none of my hosts seem to be listening on tcp/8085 anymore since 
>>> the upgrade.
>>> 
>>> Was working fine before the upgrade. Is there something that needs to be 
>>> done to convert the dashboard config to work with the new version?
>>> 
>>> -- 
>>> Jeremy Palmer
>>> Senior Network Engineer
>>> ViaWest, Inc.
>>> Office: 720.891.1045
>>> Fax: 303-874-5236
>>> 
>>> http://www.viawest.com
>>> 
>>> 
>>> 
>> 
>