perfsonar development work

[Jason Zurawski <>]

Szymon Trocha wrote:
> Jeff,
>
> Jeff W. Boote pisze:
> > It is not clear if this is a discovery issue, or an advertising 
> > problem at this point.
> >
> > One aspect of this is that the pS-PS hLS summarization algorithm is 
> > advertising too-wide of ranges, and we are working on fixing that. 
> > But, we also need to understand better the summarization algorithm of 
> > the MDM hLS. It looks suspect that it is advertising /32's.
>
> I think /32 can reflect correctly what is in PIONIER RRD MA. It has only 
> single interface in each of these networks in metadata config file. So 
> MDM hLS advertises /32s for these two IPs. Or you would expect something 
> different here?


One reason I would like to see what the API is sending is to verify 
this.  I don't believe we every had a use case where the hLS would 
advertise a range as small as 1.  Advertising a /32 isn't wrong, just 
unexpected when services would normally be deployed for an entire domain 
of a much larger size.

We have planned for the use case that people would search for a very 
specific IP, but as shown by Nina's tests this can be expensive.  I will 
look into both issues after getting more info.

-jason