Skip to Content.
Sympa Menu

perfsonar-dev - Re: [pS-dev] Re: Self-test triggering/response messages pointer

Subject: perfsonar development work

List archive

Re: [pS-dev] Re: Self-test triggering/response messages pointer


Chronological Thread 
  • From: Verena Venus <>
  • To: Aaron Brown <>
  • Cc: Michael Bischoff <>, , "Luchesar V. ILIEV" <>, Roman Lapacz <>, Szymon Trocha <>, Maciej Glowiak <>, Stijn Melis <>, Cándido Rodríguez Montes <>, Nicolas Simar <>, Domenico Vicinanza <>, perfSONAR developers <>, , "Jeff W. Boote" <>
  • Subject: Re: [pS-dev] Re: Self-test triggering/response messages pointer
  • Date: Sun, 22 Jun 2008 18:02:54 +0200
  • Organization: Rechenzentrum Erlangen

Aaron,

Am Montag, 23. Juni 2008 20:45 schrieb Aaron Brown:
> Michael Bischoff wrote:
> > Because perfsonar is suppose to be implementation ambiguous you can't use
> > specific tests but always have to rely on more abstact test/questions.
> > (is well configured?, is operational? etc) The answer should (offcourse)
> > be as specific as possible eg ('something' failed) isn't going to make
> > anything happy.
> >
> > This is also the reason why my proposal doesn't have to return one value,
> > but could return more then one. (errors that is as success(es) need(s)
> > not to be reported back. I'll add an example that also illustrates
> > that(which I probably should have added in the first place, sorry))
>
> There's no reason why this new proposed schema only has to provide 1
> return value. You could ask for the generic test and it could spit back
> the result of a series of tests.
>
> The request in verena's example would be something like:
>
> <nmwg:message id="msg1234" type="EchoRequest"
> xmlns:nmwg="http://ggf.org/ns/nmwg/base/2.0/";>
> <nmwg:metadata id="meta">
>
> <nmwg:eventType>http://schemas.perfsonar.net/tools/admin/selftest/1.0</nmwg
>:eventType> </nmwg:metadata>
> <nmwg:data id="data" metadataIdRef="meta"/>
> </nmwg:message>
>

That is exactly the message I sent to my service to get back the respose I
pasted on the wiki site. Maybe I should have added this.

Reporting back the results with complete metdata blocks containing event Type
has the advantage, that it is possible for the service desk to rerun single
tests which reported an errror. Just parse the response for failure return
values and then send a request containing the corresponding eventtype.

So in the example:

Repsonse from the service to generic request:
...
<nmwg:metadata id="bwctld_running_test">
<result:subject id="subjreturn" metadataIdRef="meta"/>

<nmwg:eventType>http://schemas.perfsonar.net/tools/admin/selftest/MP/BWCTL/bwctld_running_test/error/1.0</nmwg:eventType>
</nmwg:metadata>
<nmwg:data id="data_bwctld_running_test"
metadataIdRef="bwctld_running_test">
<nmwg:datum>bwctld not running! </nmwg:datum>
</nmwg:data>
...

And then the next request from service desk could be:
<nmwg:message id="msg1234" type="EchoRequest"
xmlns:nmwg="http://ggf.org/ns/nmwg/base/2.0/";>
<nmwg:metadata id="meta">

<nmwg:eventType>http://schemas.perfsonar.net/tools/admin/selftest/MP/BWCTL/bwctld_running_test/1.0</nmwg:eventType>

</nmwg:metadata>
<nmwg:data id="data" metadataIdRef="meta"/>
</nmwg:message>

I m not sure if the service desk is going to use this, but it would be
possible.

This is not implemented at the moment, but it may also solve Stijns problem.
Moreover I think this is easy to achieve, if you have already implemented the
first step (generic case like stated above).

To summarize:
I like to keep the idea of reporting back fully qualified metadata/data
blocks
for each specific selftest which was performed, while asking in a generic
way. It gives the requester a lot more possibilities to handle the results.

Regards,
Verena



Archive powered by MHonArc 2.6.16.

Top of Page