[gin-data] SRM interop plan

[Alex Sim]

Hello Ian,
Yes, Jarek was willing to help on the issue, and we're trying to come up
with the correct questions and code samples now. thanks
--Alex
 



  _____  

From: Ian Foster [mailto:foster at mcs.anl.gov] 
Sent: Wednesday, April 19, 2006 9:34 AM
To: asim at lbl.gov; 'Jensen, J (Jens)'; 'Aleksandr Konstantinov'
Cc: Erwin.Laure at cern.ch; gin-data at ggf.org; 'Srm-Devel (E-mail)'
Subject: RE: [gin-data] SRM interop plan


Alex:

Don't hesitate to consult with the Globus team with such questions.

Ian.


At 09:30 AM 4/19/2006 -0700, Alex Sim wrote:


Recently we encountered a problem in communication between GT3.2 and GT4
client and server. GT4 client doesn't reach the GT3.2 server any more, nor
vice versa, through the same interface, whereas they do communicate within
the same GT versions. I'm not sure if it's related to this GSS module issue,
and there seem few issues around framework and versions. 
These are important issues that we as a collaboration have to resolve for
our interoperation and compatibility. Let's try to compile the list and
discuss in one of the conf calls and/or meetings.
--Alex

| -----Original Message-----
| From: Jensen, J (Jens) [mailto:J.Jensen at rl.ac.uk] 
| Sent: Wednesday, April 19, 2006 12:51 AM
| To: asim at lbl.gov; Aleksandr Konstantinov
| Cc: Erwin.Laure at cern.ch; gin-data at ggf.org
| Subject: RE: [gin-data] SRM interop plan
| 
| Hi Alex,
| 
| This is because in Globus 2 at least, it turns out that if 
| you build your server/client with GLOBUS_IO_MODULE then it is 
| apparently incompatible with servers/clients built on top of 
| the lower level module GLOBUS_GSSAPI_MODULE [module names 
| from memory so may be slightly wrong].
| 
| Now Aleksandr discovered this independently and dug a bit 
| deeper than we had done, but it still appears that these are 
| incompatible.
| 
| This is mainly important for the SOAP stuff because as you 
| say it goes via HTTPG.  The de-facto standard in LCG is the 
| GSSAPI module approach.
| 
| Even if you don't care about HTTP as a data transport 
| protocol, Aleksandr's point is still valid.  It's a question 
| of testing different implementations of GSI sockets.
| 
| Cheers,
|                       --jens
| 

_______________________________________________________________
   Ian Foster, Director, Computation Institute
Argonne National Laboratory & University of Chicago
Argonne: MCS/221, 9700 S. Cass Ave, Argonne, IL 60439
Chicago: Rm 405, 5640 S. Ellis Ave, Chicago, IL 60637
Tel: +1 630 252 4619.  Web: www.ci.uchicago.edu
<http://www.ci.uchicago.edu/> .
      Globus Alliance: www.globus.org <http://www.globus.org/> .
        


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.ogf.org/pipermail/gin-data/attachments/20060419/347e3293/attachment.htm