SABnzbd Forums

Just got the 0.6.0Beta2 update on ubuntu 10.04, tried fetching and starting a new nzb after the update was complete, and the download fails with warnings:

2011-02-25 11:49:35,211 WARNING: None -> Unknown encoding

It pretty much just zips through the entire nzb saying every article is not found.

Forgot to mention, nzb is from nzbmatrix, and hosting is newshosting.

This usually means that the material is outside the retention time of your Usenet server.

shypike wrote: This usually means that the material is outside the retention time of your Usenet server.

I would agree, but the nzbs that were downloading were RSS feeds, only hours old, and another that was ~80 days old (newshosting retention is 860 days)

Please send one of these NZBs to bugs at sabnzbd.org with a reference to this message.
Is it possible to test the same NZB with a different Usenet client?

shypike wrote: Please send one of these NZBs to bugs at sabnzbd.org with a reference to this message.
Is it possible to test the same NZB with a different Usenet client?

I can try them with a previous version of sabnzbd on my laptop.  If it is successful on there I will send the nzb in.

It definitively is a problem - I have the same issue.
Usually download fails completely - and then I can retry and it works. have tried to set the categories to other options.

Ditto.  Have gone back to 0.5.6.  Which works fine.

The problem is that you need to make sure to set the retention under  Config -> Servers.

It defaults to 0, set it to something at or above (if giganews) the retention. I had this error on Supernews until I set that to 800. After I set that value it worked just fine.

premiso wrote: The problem is that you need to make sure to set the retention under  Config -> Servers.

It defaults to 0, set it to something at or above (if giganews) the retention. I had this error on Supernews until I set that to 800. After I set that value it worked just fine.

This solved my problem (plus a restart of sabnzbd daemon)!  Thanks!

Just wanted to make a note on this issue. After I set the retention I had a bunch of downloads work just fine. Just added a few using nzbdstatus 2.0f, and all of them failed with this error. I hit the retry and they worked fine on the retry.

Something might be buggy about the retention check, or retention check when receiving NZBs from the api.

EDIT:
I attached my log from when the error occurred, hopefully this helps diagnose the issue.

premiso wrote: I attached my log from when the error occurred, hopefully this helps diagnose the issue.

The log file shows unrelated problems.
The "ImportError:" suggests an incorrect installation of SABnzbd.
The "None -> Unknown encoding" is about an article that contains an unsupported encoding.
(It could be a side-effect of other problems).

I repeat my call to send in evidence, in the form of NZB files (and a description of circumstances).
Retention 0 means the retention setting is ignored.
Of course it is possible that an error occurs due to some unexpected situation,
or just plain incorrect program code,  but I need more information to reproduce such an error.

Sent the information via email.

To completely fix the issue, I just generate a brand new .ini file and then rebuild it from the ground up. That seems to have worked, I am thinking it could be a category issue, but not 100% sure, but that has fixed my issues completely.

So in the end, when you upgrade just regenerate the sabnzbd.ini file and that should work fix the issue.

The only slight clue I have for this is the report from
other people who claim that they had to set the server retention to a large value.
Implying that retention 0 is not "switch off" but actually 0.
I cannot confirm this, but will investigate it further anyway.
There is a known bug in the default category, but that results in other problems.

I had this issue and the resolution for me seemed to be to disabling the "Send Group" option under switches. before figuring that out I also created a new virgin configuration ini so that might have had something to do with it as well, but not until i unticked Send Group did it stop the constant errors.

"Send group" is off by default, because only few (old fashioned) servers require it.
"On" should work with any server.
My guess is that it's the new INI file and/or a queue without 0.5.6 jobs in it, that fixes things.
As you can see in the "known issues" message at the top of this board,
there are some potential upgrade issues.