Out-of-retention message for newly posted file

[sinnerg]

I posted a test to alt.binaries.test, created the nzb and loaded in sabnzbd. It's just a test pdf.

sabnzbd downloads the pdf, but then reports "out of retention" error. Yet the pdf is fully intact in the incomplete folder. There are par2 files, but it doesn't bother to fetch those.

getnzb and newsleecher fetch the file just fine.
There are no missing articles and the articles are (as of now) only a few hours old.

The nzb for testing is here => https://docs.google.com/file/d/0B0MoqHv ... sp=sharing

Is it possible to get some more info on why this is reported as "out of retention" when it isn't? I have a feeling my CRC calculations for the articles might be at fault. The error "out of retention" doesn't give me a clue w.r.t that but it has indeed managed to fetch the complete file.

PS: the pdf posted is corrupted, I know, but that's not the issue here as it is still exactly the file I posted.

[sander]

What a strange post. http://binsearch.info/?q=%3B.net-May.20 ... 00&server= gives confusion information.

[sinnerg]

That because that one I posted twice for testing. It's not the one in the nzb anyway. The nzb has the title "001".
That's just binsearch grouping the duplicate post.

I had to get the nzb from nzbindex.nl and I just noticed it doesn't have the initial par2 file. My generated nzb (typically sitting at home right now) has it. I'll upload it later

[sander]

PS: you know "out of retention" actually means "not available (anymore) on the newsserver"? So, a possible cause could be: NZB has article that do not exist.

(The message itself will be changed in SABnzbd 0.8.0)

[sinnerg]

Yes, I am fully aware of what that means. :-)

But if you looked at what I mentioned, the pdf in there downloads fine and is sitting in the incomplete folder.

I suspect now that something went wrong with the initial par2 file post. Will check the script tonight to see if that's the case. Bit odd since the usenet server reported posting ok.

Then again, that doesn't make sense either as using just the pdf articles (all of which exist) still give the same message. I'll do a script to verify them all this evening.

PS: I'm probably got a silly error somewhere and for the life of can't spot it.

[sinnerg]

Figured it out. Nothing wrong with my application. sabnzbd just doesn't like the "." at the start of the filename. Renamed post to start with "dot" and all is good.

[shypike]

Figured it out. Nothing wrong with my application. sabnzbd just doesn't like the "." at the start of the filename. Renamed post to start with "dot" and all is good.

You mean, SABnzbd doesn't like filenames beginning with a "."?
I assume you mean this: Renamed post NOT to start with "dot" and all is good.

[sinnerg]

I replaced the character "." with the word "dot". As in ".XYZ" => "dotXYZ"

[shypike]

I cannot reproduce a problem with files starting with a "." character.