Download Failed Out of your Server's Retention

[LordDusk]

http://pastebin.org/139389 ( post was over 20k chars )

Entered the nzb in question and pushed pause after 10 secs.

[shypike]

Sorry, it contains no useful data.
You should have zipped all the 5 logfiles and emailed it to bugs at sabnzbd.org
(and I have should have told you so)

[LordDusk]

sent

[shypike]

The logs are full of error messages about files that cannot be saved.
Which is no surprise given that SABnzbd wants to write in a folder who's name is forbidden.
Alas the part where the NZB  enters the queue is not in the logs.
What I don't understand is how this illegal name can used as the job name.
In fact it can only happen through the API, because there is a bug where the nzbname parameter is not checked for illegal characters.
Again: when the NZB file enters the queue from a disk file, it simply cannot have an illegal name.

Do you get it directly from the search engine's URL?
How does the job name look when it's in the queue?
Does it contain double quote marks?

[LordDusk]

let me try to make it clear :

step 1 - Enter the filename in nzbindex.nl


step 2 - Click download


step 3 - Double click it in the downloads folder


step 4 - open sabnzbd

[shypike]

This specific NZB goes wrong?
It's downloading now without any problems(not finished yet) finished correctly.
I'm quite convinced that it has nothing to with the NZB itself,
but everything with the way it's entered.
Maybe because of the way you use it, maybe because of a bug in SABnzbnd.
For the time being: do not use double-clicking of the NZB, but use the watched folder or "add nzb" button.

PS: I'm removing your screenshots, they're against the forum policy.

[ilyalozovyy]

This specific NZB goes wrong?
It's downloading now without any problems (not finished yet).
I'm quite convinced that it has nothing to with the NZB itself,
but everything with the way it's entered.
Maybe because of the way you use it, maybe because of a bug in SABnzbnd.
For the time being: do not use double-clicking of the NZB, but use the watched folder or "add nzb" button.

PS: I'm removing your screenshots, they're against the forum policy.

After running into the same issues I have done a bit of research and ended up finding out that the issue is with the Chrome plugin that I was using
called "SABconnect++".
In the past year or so I've been using Giganews service and have never ran into any retention issues with them and just a couple of days ago I started having problems and the only thing that has changed since was me starting to use Chromium vs Firefox. Once I reverted to using Firefox again with the trusty old nzbdStatus plugin, everything started working once again.

Hopefully this helps someone.

[shypike]

and ended up finding out that the issue is with the Chrome plugin that I was using called "SABconnect++".

This does confirm my suspicion that it's the name of the job that causes this problem,
not the content of the NZB.
The error message that SABnzbd gives is obviously confusing.
The coming release 0.5.5 is very likely to solve this problem because it will clean up
incorrect names in the queue.

[LordDusk]

and ended up finding out that the issue is with the Chrome plugin that I was using called "SABconnect++".

This does confirm my suspicion that it's the name of the job that causes this problem,
not the content of the NZB.
The error message that SABnzbd gives is obviously confusing.
The coming release 0.5.5 is very likely to solve this problem because it will clean up
incorrect names in the queue.

Hmm indeed, when i turn off that addon, all goes back to normal ( i can now download the nzb without the error ), but the weird thing, it worked on 5.3 :O

[shypike]

The cause is also inside SABnzbd, due to lack of filtering of names.
This will be fixed in the next (minor) release.