- From: Antoine Mensch <antoine.mensch@odonata.fr>
- Date: Fri, 09 Jan 2009 13:22:00 +0100
- To: Doug Davis <dug@us.ibm.com>
- CC: public-ws-resource-access@w3.org
Doug, I did not want to anticipate on the WG decision: if the WG decides to accept this issue as valid and resolves it by selecting WS-MC as the appropriate solution, it would be perfectly OK with me. Cheers Antoine Doug Davis a écrit : > > Antoine, > You mention using something "similar to MC" to support this - but > I'm wondering why you didn't just say "use MC" ? :-) > Reusing an existing WS-* specs would be my preferred choice and if you > look at one of the examples you mentioned (silverlight) it is already > doing this I believe. > > thanks > -Doug > ______________________________________________________ > STSM | Web Services Architect | IBM Software Group > (919) 254-6905 | IBM T/L 444-6905 | dug@us.ibm.com > > > *Antoine Mensch <antoine.mensch@odonata.fr>* > Sent by: public-ws-resource-access-request@w3.org > > 01/08/2009 01:18 PM > Please respond to > antoine.mensch@odonata.fr > > > > To > public-ws-resource-access@w3.org > cc > > Subject > [NEW ISSUE] WS-Eventing Push delivery mode does not work when the > subscriber is not addressable > > > > > > > > > > > Hi, > > My name is Antoine Mensch and I am a member of the OASIS WS-DD TC > working on the Devices Profile for Web Services (DPWS). We've > implemented WS-Eventing as part of DPWS for quite some time now, and we > have run into the following issue: > > Description: > The WS-Eventing default delivery mode (used in DPWS) is based on a push > mechanism that requires the event source to connect to the subscriber to > deliver the event notification. This creates problems in two very common > cases: > (i) when the subscriber is behind a firewall using NAT; > (ii) when the subscriber is a Web application (e.g. Ajax or Java, Flash > or Silverlight applets) which cannot accept incoming connections. > Although the first problem can sometimes be solved using dynamic port > mapping, the second one normally requires a change to the default > security policy of the Web browser, which is generally not acceptable. > > Proposed resolution: > Introduce a special URL to be used as the [address] of the notifyTo / > endTo endpoint references in the Subscribe message. This special URL > would inform the event source/subscribtion manager that event > notifications are not to be sent directly to the subscriber, but rather > that the subscriber will initiate a connection to fetch them. The > proposed mechanism could work in a way similar to the one described in > the WS-MakeConnection specification. > > Best regards, > > Antoine Mensch > Odonata > > > > > > ------------------------------------------------------------------------ > > > No virus found in this incoming message. > Checked by AVG - http://www.avg.com > Version: 8.0.176 / Virus Database: 270.10.5/1882 - Release Date: 08/01/2009 08:13 > >
Received on Friday, 9 January 2009 12:22:40 UTC