Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

Close

Circular redirect

Help
csaba
2011-11-13
2012-10-31
  • rssowluser4896
    rssowluser4896
    2012-02-04

    I've also encountered this. The feed url is
    http://clientsfromhell.net:80/rss, and
    it also validates as correct. I've found that it sometimes works (once only)
    if I remove the :80 from the address, but then it reappears the next time I
    load RSSOwl, and the error returns.

     
  • genodeftest
    genodeftest
    2012-02-04

    Ok, I was wrong. You are right, this can be reproduced.

    The strange thing is: Even after restarting RSSOwl multiple times after adding
    the feed I had no problems. But after hibernating my Operating System it
    didn't work any more.

     
  • genodeftest
    genodeftest
    2012-02-04

    1. RSSOwl.org seems to use a old (unsupported/obsolete?) namespace (ironic?)

    2. confirmed. Don't know why, I'll keep investigating ;)

    3. Your RSS URL is incorrect. The correct one is http://stackoverflow.com/feeds/tag/java ("tag" instead of "tagged"). Seems like stackoverflow changed their URLs.

    4. Works fine for me.

    5. This page does not provide a RSS, Atom or RDF feed.

     
  • rssowluser4896
    rssowluser4896
    2012-02-04

    Yeah, most of the feeds haven't been checked since I imported them, and are
    quite old. The stackoverflow feed is wrong, but the right link (which has a
    pile of tags appended to it) 404's.

     
  • I am also still having the “Circular redirect” problem in RSSOwl 2.1.4.
    If I manually check what happens when a problematic feed is retrieved (like http://13thwitness.com/rss), they first reply with a “HTTP/1.1 302 Moved Temporarily” that points to for instance /rss?ef7ddf30.
    Then, if I retrieve http://13thwitness.com/rss?ef7ddf30 I get another 302 redirect back to the original /rss. Retrieving this will now provide the actual content.

    No idea why they're doing this, I suppose to thwart crawlers or something. Anyhow, it seems the test for the circular redirect decides too early that there is a loop. It should try a little longer.

     
    Last edit: Alexander Thomas 2012-10-14