I still don't get how torrents work.
I mean I can copy and paste a link, but how the fuck does it work? What is seeding? What the fuck is a tracker? And why is it even called tracker?
I still don't get how torrents work
Kevin Adams
Hudson Parker
Good news! Google does not yet censor "how does bit torrent work".
Jonathan Gutierrez
>seeding
uploading files to other people
>peer
people you download the files from
>tracker
a server that helps you find peers
>torrent file
a file with metadata containing the names of files you want to download and a list of trackers
Ian White
oh and
>magnet link
link you can use instead of the torrent file, it serves the same purpose of the torrent file
Jace Scott
user was a nice man here
Ethan Jones
I actually didn't know what tracker was, thank you.
Jayden Hill
i have deduced all of this over the last 15 years from context alone. how hasn't is beyond me
Jayden Rivera
A shitty file sharing method created for level (all equal) local networks and co-opted by piratefags to very poorly and unreliably download garbage over the not even close to being level internet, and which is not worth the time and effort to get because all the crap they download is trash anyway, all while advertising their IP address to anyone else downloading it, including legal teams that will fuck them over.
Tyler Cooper
Cute oldfag
Connor Flores
Not my heckin' IP address.
Landon Garcia
>t. retard living in a non-free country
Andrew King
This is your brain on ISP Corporation Controlled
Jacob Reed
use tribler since you really seem to need it
Gabriel Myers
Read up on distributed hash tables
en.wikipedia.org
And read the original bittorrent spec
bittorrent.org
Levi Myers
the original idea was to allow it to work with little bandwidth
Ayden Richardson
How many seeds should I have active at once. I have over 1200 files to seed but seeding that many at once causes client instability. Even with all files "seeding" there never seems to be more than 5 that are actively uploading. Should I just leave 90% of my files in a seed que? Will the files in que automatically start seeding if there's a peer to connect with? I'm not sure how this works. How many files is best to have available to (active) seed at once?
Juan Martinez
How do torrents without trackers work?
Justin White
with the DHT, a distributed database of who has what files, shared between clients. trackerless torrents still include a bootstrapping dht node iirc, but you don't need it if you're already participating in the dht
Jaxon Nguyen
Awww, poor user, show me where torrents hurt you
Ryder Russell
Torrents started out as a caching mechanism, distributed CDN. Instead of downloading the actual file, you are downloading a file with metadata (the .torrent). The server of the site owner would seed/host the file and would give you a list of other downloaders. Instead of just downloading from the webhost, you can then try to connect to other downloaders and they would help offload the bandwidth from the webhost.
What do you need to publish a torrent download:
- Some kind of initial "seed", hosting the file that is to be distributed. That could be the webserver or a computer at home.
- A tracker, which downloaders can contact. It will add everyone asking to the list. It gives the list to the downloaders asking.
Brayden Anderson
Here is an example tracker software, which the webmaster could use on his webserver: github.com
Next, the webmaster would create the metadata for the file to download. So e.g. it is a 15 GB video file. He creates hashes for the file (multiple hashes so the incomplete file can be shared later). He also includes the tracker URL into the metadata file, which could be something like tracker.mywebsite.com
On the HTML website itself, the website owner will tell people that the file is available as torrent, and links to the .torrent. Which is a few KB in size. He tells the visitors they need a bittorrent client to download the file.
Angel Rivera
The visitor downloads the .torrent metadata file, and opens it with a bittorrent client. The client reads the filename/directory and prompts where to download to. The filenames can be instantly displayed since they are included in the .torrent.
Once you confirm, the bittorrent client would make a simple HTTP request to the "tracker" - so it downloads a page like this tracker.mywebsite.com
Included in the request is the info_hash, which identifies the torrent. There may be more than one torrent hosted on the server, so the downloaders should be separated. The bittorrent client will also report the port it is listening on and some diagnostic information.
As response, the bittorrent client gets a list of peers (other downloaders, both still downloading or with finished download) from the server. It receives IP and port and some ID for each peer.
Parker Wright
The bittorrent client now tries to connect to all the peers it received. Some are offline by now, some firewalled. But some it can get a connection. It will ask them what parts of the file they have. If a downloader has a full copy, he is referred to as a "seeder". When there is no seeder and nobody has any part of the file, nobody can download. If there are 200 downloaders, then in the ideal case, the seeder only needs to "seed" a unique part of the file once, to one other downloader. This downloader would upload it to 10 other downloaders, and so on. Everyone would have this part of the file. If there are more seeders, the sharing is accelerated by giving downloaders parts of the file, which they can then redistribute.
The webserver with the list of downloaders keeps track of them so that's why it's called a tracker.