How does the IDK work? (an addendum)

| | Comments (0) | TrackBacks (0)

As with most things I post here, you probably don't need to know the details to get by, but it sure helps to understand them. Case in point...

As part of some behind the scenes "figuring stuff out so we can tell the client how it works", I happened to ping the Function1 guys (some ex-BID consulting folks) about the IDK. What I was trying to figure out was how the IDK PRC actually talked to Collaboration/Publisher. It didn't take long for them to come up with an answer. Here's a blog post on what they found.

Interestingly enough, the PRC goes directly to the Collaboration and Publisher hosts, completely bypassing the API server (after an initial call).

Update:

As per the always informative Gerald Kanapahty comments on my blog: the base URL for the Publisher and Collaboration web services actually come from the portal itself. The IDK looks up the UUID of the Publisher/Collaboration remote server objects in portal administration and uses those base URL's to build the full web service URI's.

Definitely a great analysis, guys (and thanks for plugging my blog).

0 TrackBacks

Listed below are links to blogs that reference this entry: How does the IDK work? (an addendum).

TrackBack URL for this entry: http://hross.net/mt/mt-tb.cgi/19

Leave a comment

About this Entry

This page contains a single entry by hross published on October 17, 2007 2:32 PM.

Automation Server isn't the only thing that runs jobs. was the previous entry in this blog.

Where are the logs? is the next entry in this blog.

Blogroll


Integryst

Function1

Fabien Sanglier

Bill Benac

Jordan Rose

Chris Bucchere

Robert Herrera

Nanek Blog Aggregator

Spartan Java




if you'd like to be listed here.




I don't blog about non-tech issues here, but you can check my Google Reader Shared Items if you want to know what I'm currently interested in.

Categories