I realized I had probably posted this in the wrong place initially. Here goes:
Interesting problem, an http post I am executing to get an rss feed works fine in v4. But in v5 and 6 it only downloads the header section of the xml that's returned; that is, it downloads things like the feed title and language, but doesn't download any articles. But when you take that exact same endpoint -- the page and querystring that are compiled dynamically in the orchestration -- and post that in a browser's URL field, you get the full xml document including articles.
But again, in v4 you get the full document.
UPDATE: A new twist: after returning from my holidays, I ran the job again in v. 6 on my desktop, and it found a few hundred new items in the feed and ran successfully. But subsequent runs, even when there should be items in the feed as evidenced by hitting the url in a browser, again bring up the empty feed -- xml headers but no items. It's almost as if it caches the feed at a point when there is nothing new to download, and then holds on to that cache. (But it didn't cache the feed when there were hundreds of items.)
NOTICE: developerWorks Community will be offline May 29-30, 2015 while we upgrade to the latest version of IBM Connections. For more information, read our upgrade FAQ.
This topic has been locked.
1 reply Latest Post - 2013-01-03T20:31:01Z by SystemAdmin
Pinned topic http post breaks in a strange way from v4 to v5, 6
Answered question This question has been answered.
Unanswered question This question has not been answered yet.
Updated on 2013-01-03T20:31:01Z at 2013-01-03T20:31:01Z by SystemAdmin
SystemAdmin 110000D4XK1250 PostsACCEPTED ANSWER
Re: http post breaks in a strange way from v4 to v5, 62013-01-03T20:31:01Z in response to SystemAdminYou may want to try adding an HTTP header item to the HTTP request with a name of "Connection" and a value of "close." Also, setting up an HTTP trace between Studio and the HTTP endpoint can help determine when running Verify whether in fact the server is returning all content items, and not simply ones that your client hasn't already consumed.