Laurens, I'm using v2.02, and am getting a strange behaviour on certain feeds. These are taking a long time to update, and then if I watch, they actually start counting backwards ... e.g. 79%, 78%, 77% ... until I stop them.

Could you see if you can reproduce my problems on ...

http://www.elsua.net/feed/
http://feeds.feedburner.com/ross (and http://ross.typepad.com/blog/rss.xml )
http://feeds.feedburner.com/Torontoist

I have the setup for "Output only new feed items", if that makes a difference. Thanks.

This is normal. The percentage will go "backwards" as new links and images are discovered while spidering the pages. This behaviour occurs particularly for pages with lots of links.

Laurens, I think that there may be something else going on.

As an example, I've been updating http://www.elsua.net/feed/ for about 90 minutes, and it's stuck at updating 54%. The CPU is running 99%, and the Windows Task Manager says that the process is running 1,226,696K. At this point, I'm shutting it down.

Yes, the three feeds I suggested tend to have a lot of links, but all of the others complete within 5 minutes, whereas these ones can run for hours.

Sorry, I tried this feed and it, indeed, takes a long time to download. After 15 mins it was still at 10%, but no crash. There is, however, a limit to how much time I'm able and willing to spend on supporting Sunrise and this is well over that. So for now, don't use these feeds. Hope you understand.

Laurens, thanks, I just wanted to make sure that it wasn't just me having the problem. These are bloggers that link to a lot of other places, and it's not clear that I really want all of the links anyway ... so I've set the link depth to "restrict to domain".

For http://www.elsua.net/feed/ and http://ross.typepad.com/blog/rss.xml , this sped things up tremendously, i.e. updates in seconds.

Do you have any suggestions on how to set http://feeds.feedburner.com/Torontoist ? The blog per se is at http://www.torontoist.com/ , but while the metadata indicates a feed at http://www.torontoist.com/index.rdf , it seems to be redirected to feedburner. My solution of "restrict to domain" then reads off Feedburner.

I think that it's okay to read off Feedburner ... although it's not going to the original source.

Specify an include filter for <a> anchor links using "http://www.torontist.com/*"