Thread: Rtorrent skip hashing for cross seeding

There is a setting called 'check_hash = yes|no" that you can add in .rtorrent.rc but I think it only allows you to skip hashing files that are currently tied to an existing torrent which is normally found in the session directory. So if you try to load the same torrent downloaded from another tracker, it will still be hashed by rtorrent because there is no session file tied to that torrent yet.
According to rtorrent's common task documentation, if you want to bypass hash-checking of data files you need to use a script called rtorrent_fast_resume.pl. So what I did is incorporate this perl script inside a bash script so it can be loaded in .rtorrent.rc. It's not a very elegant solution but I think it will work. Note that I have never tested the script yet.

Requirements:
Perl
rtorrent_fast_resume.pl (Download here -> http://libtorrent.rakshasa.no/downlo...fast_resume.pl)
Convert::Bencode module from CPAN (enter this command to install -> $perl -MCPAN -e 'install Convert::Bencode'), you'll probably need root access to install the binaries and dependencies.
Add Below lines to your .rtorrent.rc.

# Load torrents from watch folder every 5 seconds
schedule = watch_directory,5,5,load_start=$HOME/rtorrent/watch/*.torrent

# This line will execute the script "process_torrents" every 5 seconds
schedule = exec_script,5,5,"execute=$HOME/process_torrents"

================================================== ==================================================

Restart rtorrent
Create a file in your home directory and name it process_torrents
Copy and paste contents below into process_torrents
Save
chmod 700 process_torrents

================================================== ================================================== =


#! /bin/bash
for TORRENT_FILE in $HOME/rtorrent/watch2/*; do

# Check for torrent files
if [ -f "$TORRENT_FILE" ]; then

# Create a semi-unique filename for the next torrent. Otherwise, rtorrent will untie first torrent
# and then load next torrent. We want both torrents to be loaded.
NEW_TORRENT_FILE="$HOME/rtorrent/watch2/$(basename "$TORRENT_FILE" .torrent).$RANDOM.torrent"

# Original commandline parameter: $ rtorrent_fast_resume.pl [base-directory] < original.torrent > modified.torrent where
# base-directory is the location of your torrent data directory
$HOME/rtorrent_fast_resume.pl $HOME/rtorrent/data/ < "$HOME/rtorrent/watch2/$TORRENT_FILE" > "$HOME/rtorrent/watch/$NEW_TORRENT_FILE"

rm -f "$TORRENT_FILE"
fi
done


================================================== ================================================== =

Next, drop the torrent file downloaded from tracker A to $HOME/rtorrent/watch/ folder
Wait until the data files have been fully downloaded
Drop same torrent filename downloaded from trackers B,C,D,E etc. to $HOME/rtorrent/watch2 folder