Downloading missing Blocks...

Get help with all aspects of SABnzbd
Forum rules
Help us help you:
  • Are you using the latest stable version of SABnzbd? Downloads page.
  • Tell us what system you run SABnzbd on.
  • Adhere to the forum rules.
  • Do you experience problems during downloading?
    Check your connection in Status and Interface settings window.
    Use Test Server in Config > Servers.
    We will probably ask you to do a test using only basic settings.
  • Do you experience problems during repair or unpacking?
    Enable +Debug logging in the Status and Interface settings window and share the relevant parts of the log here using [ code ] sections.
Post Reply
ronvegas
Newbie
Newbie
Posts: 1
Joined: August 1st, 2011, 1:10 pm

Downloading missing Blocks...

Post by ronvegas »

Hi,

When sab downloads more blocks I understand it is downloading
more par2 files to repair the file.
If the NZB i download already includes all the available pars, is it
a case of sab not downloading them until the end and it decides how
many it needs OR has it already used them and decides it needs more,
in which case where does it get more from if they are already included?

I'm wondering if I can feed it more pars in the begining to save time
on the repeated verifying it does when a repair is needed.

On a seperate note:
On very old post that couldn't be repaired I was finding it necesary to download
the bulk of a NZB again with my second server (using Usenetserver & ngroups)
and then let Quickpar make one good file, but sometimes found still missing blocks.
I'm now testing Giganews to replace both the others and so far the completion is better
with only one download.

Thanks for your help.

Ron.
User avatar
shypike
Administrator
Administrator
Posts: 19773
Joined: January 18th, 2008, 12:49 pm

Re: Downloading missing Blocks...

Post by shypike »

At first SABnzbd will download only the smallest par2 file of the ones present in the NZB.
If the first verify/repair attempt fails, more par2 files will be downloaded.
So, SABnzbd is not capable of finding more par2 files that there are in the NZB.
However, you can look for more par2 files using search engines and
feed SABnzbd an additional NZB file when you use the retry button in the History.
firepath
Newbie
Newbie
Posts: 3
Joined: September 20th, 2011, 7:12 am

Re: Downloading missing Blocks...

Post by firepath »

Hi, I've just installed SABnzbd 0.6.9 on Windows XP, and I'm not witnessing the behaviour described by shypike.

SABnzbd downloads all PAR2 files for a release. It even grabs the PAR2 files first if they are at the top of the NZB file, taking the smallest PAR2 file last if it is at the bottom of the NZB file (together with the SFV and NFO files, as is often the case). I would have expected the application logic to skip the "other" PAR2 files (that contain recovery blocks), to start with the smallest PAR2 file, and then to continue grabbing the remainder of the release's archive files, checking them against the smallest PAR2 as they are downloaded and decoded.

On the "Switches" config screen I have only checked the following boxes:
[X] Enable Quick Check
[X] Post-Process Only Verified Jobs
[X] Enable MultiCore Par2

On the "Categories" config screen, I have selected as default Processing operation "+Repair", and I use the default profile for each download.

What am I doing wrong, and how can I correct this situation? The goal is to download and repair all archive files, but not to perform any extraction or any further processing.


Thanks for any help.
Chris
User avatar
shypike
Administrator
Administrator
Posts: 19773
Joined: January 18th, 2008, 12:49 pm

Re: Downloading missing Blocks...

Post by shypike »

I think you misinterpret what happens.
Initially an NZB is sorted with all the par2 files at the top.
As soon as the first (smallest bla.vol,xx.yyy.par2) is in, the rest of the pars goes to the bottom.
One par2 is needed, otherwise verification is not possible.
Very small extra par2 files can be downloaded, due to the parallel downloading.

+Repair does just that: repair and no unpacking.
Which attribute do the files have when they enter the queue?
If you enter them through some third-party utility, then I wouldn't surprised if
that set another value than +Repair.
firepath
Newbie
Newbie
Posts: 3
Joined: September 20th, 2011, 7:12 am

Re: Downloading missing Blocks...

Post by firepath »

Hi,

Thanks for the quick reply. After reading your description I did notice that not all .par2 files are downloaded; many even seem to disappear from the queue and not appear in the download-in-progress folder (or at the bottom of the queue).

However, SABnzbd does not limit itself to downloading a single .par file. Last night it downloaded the small .par2 file, and in addition the .vol000+01.par2 file (which does not provide any added value compared to the .par2 file when it comes to verifying downloaded files). I have noticed occasions where SABnzbd downloads 3 or maybe 4 .par2 files.

I can live with 1 or 2 redundant .par2 files, but as I am on a fairly slow connection it would be a shame to waste time downloading hundreds of MBs for no good reason...


All files were added using SABnzbd's web interface. On that topic: I also noticed that if I change the post-processing option of a download-in-progress from +Repair to +Unpack it has no effect: archives are not extracted. Should I always pause and resume the download in order for the updated parameter to take effect? Or does SABnzbd require the presence of WinRAR for it to be able to extract archives? I only have 7-Zip installed. I haven't had the occasion to test with a new download, due to the slow connection speed.


Chris.
User avatar
shypike
Administrator
Administrator
Posts: 19773
Joined: January 18th, 2008, 12:49 pm

Re: Downloading missing Blocks...

Post by shypike »

It's not many megabytes.
The determining factor is the amount of parallel connections.
Of some par2 files only a few articles are read.
One reason that we postpone the decision about files is that there are many
posts that don't use the standard rules for embedding the file name in the subject.
We only know for certain what a files is really called, when the first article is in.
It's a compromise.
Post Reply