rutorrent loads or resets RSS entries marked as "Already Loaded"

[GoogleCodeExporter]

What steps will reproduce the problem?
I won't know how exactly to reproduce this, but it had happened multiple times, 
so it is reproducible somehow
1. Add RSS feeds and mark them as Already Loaded
2. Wait
3. rutorrent will either reload some or all of them or reset their status to 
Unloaded (not sure which one, I've never witnessed it as it was happening)

What is the expected output? 
rutorrent will skip any RSS entry marked "Already Loaded"

What do you see instead?
rutorrent loads or resets the status of any/some RSS entry marked "Already 
Loaded"

What environment are you using?
1. My ruTorrent version is...3.6
2. My rTorrent version is...0.9.3
3. I use web-server...nginx it version is...1.2.1 on OS...Ubuntu 12.10
4. I use browser...Firefox it version is...27.0.1 on OS...Windows 7

Are some errors present in the web-server log?
Yes, but nothing related to rutorrent

Are some errors present in the browser error console?
No

Please provide any additional information below.
This first happened to me about a week ago, rutorrent loaded every entry in one 
RSS feed, even though they were all already loaded, I thought it wasn't a big 
deal because it was only a few files.
Today, rutorrent loaded every RSS entry in every RSS feed I have, using 100+GB 
of bandwidth/HDD space, so I decided it's probably worth reporting.

This has only started happening within the last week or so, neither my 
rutorrent or server configurations have changed at all, so I can't imagine what 
is causing it.

I imagine an easy fix for this would be to implement some kind of "Load within 
<timeframe>" ie: 1 day, 1 week. Like with Match interval, but it would match 
against the timestamp of the RSS entry and only load it if it's new enough. Of 
course, that wouldn't fix the underlying issue.

Original issue reported on code.google.com by lulzw...@gmail.com on 18 Feb 2014 at 10:05

[GoogleCodeExporter]

Check this:

http://forums.rutorrent.org/index.php?topic=183.msg13180#msg13180

May be, this is your case?

Original comment by novik65 on 23 Feb 2014 at 11:21

[GoogleCodeExporter]

It's possible that it's the same issue, but my problem wasn't with that 
particular tracker.

If I understand the link correctly, the RSS entries get reloaded because the 
pubDate on the RSS entries changes?

This seemed to happen randomly, it could just be errors on the servers end I 
guess.

Is there any way to prevent this on my end? In that forum thread you said you 
made a workaround, but I'm using 3.6 release version (SVN $2404), which I 
believe was released after you made that post, so I should have that commit.

Original comment by lulzw...@gmail.com on 23 Feb 2014 at 12:43

[GoogleCodeExporter]

>If I understand the link correctly, the RSS entries get reloaded because the 
pubDate on the RSS entries changes?

This may be one of possible reasons. For saying something, i must look on your 
situation.

>In that forum thread you said you made a workaround

This is a workaround for concrete tracker.

Original comment by novik65 on 23 Feb 2014 at 5:28

[GoogleCodeExporter]

Original comment by novik65 on 4 May 2014 at 9:05

• Changed state: Incomplete

[GoogleCodeExporter]

[deleted comment]

[GoogleCodeExporter]

I am having this exact same problem, not the one in the forum but the one OP is 
having.  I am certain the created on dates are not changing.

Using rTorrent 0.9.3
ruTorrent 3.6 Rev 2404
Apache (not sure the version, it is hosted.)
Debian  Debian 3.2.57-3 x86_64

Original comment by rhomat2...@gmail.com on 28 Sep 2014 at 3:34

[GoogleCodeExporter]

Forgot to mention, I've used both Chrome and Safari browsers

Original comment by rhomat2...@gmail.com on 28 Sep 2014 at 3:35

[GoogleCodeExporter]

I actually ended up "fixing" this myself, my problem turned out to be that 
occasionally the RSS feed I was using was returning a blank, but valid, RSS 
feed, which I guess messed up how rutorrent identifies already loaded entries. 
The feed I was using had a pubDate for the "channel" entry, but the individual 
items only had guids, so that might have also been part of the problem, I 
haven't bothered looking into exactly how the rss plugin parses the entries.
In the end I just ended up writing my own PHP script that parses the search 
page and turns it into RSS instead of using the sites RSS feed, since it ended 
up being more reliable (and including more data).

Original comment by lulzw...@gmail.com on 1 Oct 2014 at 6:56