[dundee] Smaller file size than expected

[Edmund Huggett]

Hi,

This was meant to be a quick tip, however, its now a short tutorial!

I have had great success in the past using bittorent to repair not only
incomplete but also corrupted downloads (which you can't do with
curl/wget/http range requests). Curl/wget are quicker to start running and
more bandwidth efficient for incomplete downloads, but you can use the
following method if the file still does not match its checksum after doing
this!

This works best when you can find a torrent that only contains the file you
tried to download but you can obviously mark any extra files from the
torrent you are not interested in as do not download if you have to.

1. Find a torrent for the _exact same file_ you wish to repair,
http://burnbit.com/ can be very useful here as a last resort!
2. add the torrent to your bittorent client so that it creates the (empty)
files it will download the data into, but do not start the download (in
some clients you might need to start it downloading and then stop it a few
seconds later to force it to do this).
3. Make sure the file is not downloading and then copy and paste the
damaged file over the empty file you bittorent client created.
4. Tell your bittorent client to re-check/verify the data it has
downloaded, when you do this the download percentage should go anywhere
from zero up to 99% (this depends on how much data is missing or corrupted,
if it stays near zero something went wrong).
5. Start the torrent downloading and Wait, your bittorent client will now
download the missing/corrupted parts to repair the file and will skip over
the parts that verified successfully.

This method takes advantage of how bittorent works. You can get a better
understanding from wikipedia but essentially bittorent divides files up
into lots of smaller 'chunks' of data and a torrent file contains a list of
each chunk and a _checksum for each chunk_ which is used to verify that
each chunk has been downloaded correctly. Its not 100% efficient as any
corruption will cause the entire chunk to be discarded and downloaded again
but its still a better alternative to downloading a DVD image again (chunks
are normally a few megabytes or less).

Cheers

-Ed
On 16 Oct 2013 10:00, "azmodie" <azmodie at gmail.com> wrote:

> This may prove helpful in saving bandwidth
> .
> http://www.commandlinefu.com/commands/view/2876/use-curl-to-resume-a-failed-download
>
> azmodie
>
> --
> http://gplus.to/azmodie
> "Since light travels faster than sound, people appear bright until you
> hear them speak."  -- some bright spark
>
>
> On 16 October 2013 09:27, malcolm <malcolm at wxlr.co.uk> wrote:
>
>> Thanks all.  The checksum was different so I'll have another go at
>> downloading it.  It that fails I'll wait until I can get to a good
>> connection on the mainland.
>>
>> With my lack of knowledge I'd prefer not to risk a stopped and restarted
>> file.
>>
>>
>> ______________________________**_________________
>> dundee GNU/Linux Users Group mailing list
>> dundee at mailman.lug.org.uk  http://dundeelug.org.uk
>> https://mailman.lug.org.uk/**mailman/listinfo/dundee<https://mailman.lug.org.uk/mailman/listinfo/dundee>
>> Chat on IRC, #tlug on irc.lug.org.uk
>>
>
>
> _______________________________________________
> dundee GNU/Linux Users Group mailing list
> dundee at mailman.lug.org.uk  http://dundeelug.org.uk
> https://mailman.lug.org.uk/mailman/listinfo/dundee
> Chat on IRC, #tlug on irc.lug.org.uk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.lug.org.uk/pipermail/dundee/attachments/20131018/5f1cc0ec/attachment.html>