Planet Peer - The anonymous networking community
September 03, 2010, 03:48:56 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
News: Planet Peer - The anonymous networking community
...because anonymity is better ;-)
 
   Home   Help Search Calendar Login Register  
Pages: [1] 2   Go Down
  Print  
Author Topic: Rodi  (Read 9690 times)
0 Members and 1 Guest are viewing this topic.
bitz
Advanced
***

Karma: +2/-0
Offline Offline

Posts: 160

Planet Peer Community


View Profile WWW
« on: March 25, 2005, 07:26:05 AM »

Rodi or Rodia (Ρόδι or Ροδιά) means pomegranate in Greek. The Rodi program is a tiny P2P client/host (under 300K of binary code) implemented in pure Java. It's network use is similar to the bitTorrent concept. The program will serve the filesharing community with fast data delivery and serve the Open Source community by facilitating faster software deployment

http://larytet.sourceforge.net/btRat.shtml

Hmm looks to be another p2p that isn't well known. Didn't even know about it until I noticed a post here on planetpeer.  Grin

Anyways upon basic inspection, I'm not impressed. I was able to easily get and run the core. However the UI named timple alludes me completely. They have not provided a binary download of it yet (only one for the core). And upon attempting to compile it, well let's just say the default build script they provide doesn't work.

Though even if I were able to get it compiled and working, I somehow think it won't be easy/simple to configure and use...

Also I am very unclear as to the purpose of this p2p application. I did read somewhere on it's website that it doesn't provide good anonymity. I dunno if it's even really intended for filesharing or what.

I really suspect it'd take a group effort to figure this application out and put together a nice distribution (one that includes timple and a better userguide). Dunno if all that effort would be worthwhile or not.
Logged

crypton
Elite
*****

Karma: +9/-0
Offline Offline

Posts: 1677



View Profile
« Reply #1 on: March 25, 2005, 10:37:28 AM »

Download page:

http://sourceforge.net/projects/larytet/

http://sourceforge.net/project/showfiles.php?group_id=97969&package_id=127912

and the latest release:

http://prdownloads.sourceforge.net/larytet/rodi-0_1_0C.zip?download
Logged


tm
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 20

Planet Peer Community


View Profile
« Reply #2 on: March 28, 2005, 11:43:37 PM »

Rodi is a new developmental P2P network that is currently in testing. What makes Rodi unique? Many features, such as IP-spoofing for anonymity and packet-mimicking, so the P2P traffic can appear as one of many different internet traffic patterns - such as HTTP, FTP, etc - that are less likely to get blocked or throttled by an ISP's packet shaping. Unlike traditional proxied (very slow) anonymous networks (Freenet, Mute, Ants, Winny, etc) the use of IP spoofing can allow high-speed full-bandwidth downloads while keeping the uploader's true IP address hidden from the downloader.

 Rodi is open source and an ongoing discussion thread with the developer can be found on P2PForums.com 

http://www.p2pforums.com/viewtopic.php?t=11246
Logged
larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #3 on: March 30, 2005, 02:45:09 AM »

i am a developer of Rodi project.
createTimple.bat should work for Timple (i needed some name for proof of concept GUI and this is transfomed name from original Simple GUI). It is not going to compete with Azureus or eMule in graphics. Rodi Core provides flexible open interface and everybody can create own GUI in any language.

 official release on SF.net does not contain latest version, but only some reasonably stable Rodi Core.
i update CVS frequently. WEB interface of the CVS is not updated immediately after file commit and you can find source code in inconsistent state. to get latest source use cvs command on Linux machine (see SF.net documentation)

One of the ways to get binarys is to download them directly from the web site http://larytet.sourceforge.net/rodi_shtml.jar and http://larytet.sourceforge.net/rodiTimple_shtml.jar
lesson 6 and lesson 5 in the user manual will help you to get on http://larytet.sourceforge.net/userManual.shtml#Lesson%206.0


thank you
Logged
larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #4 on: March 30, 2005, 03:02:11 AM »

P.S. if you have any problems with compilation just let me know.
Most typical problem that batch files creating JARs try to sign applets using certificates. If you are going to run application locally you do not need signed JARs. i will modify batch files accordingly tomorrow. Typically it takes between 5 to 12 hourse before WEB interface of the CVS is updated.

From p2pforums - this is my response to the question which sounded like "Why do i need another P2P application
Quote
actually this is a mix of DC and BT to some extend

you can create hub and restrict access or use it exactly as BT.

you can run index server and tracker and you can live without them. you can spoof your IP address/IP port (if your ISP permits) preventing DDoS against your server.

The network is going to be searchable both by filename and by context sensitive content, for example, you can search PDF files containing fiction stories, or video clips containing specified metadata text

Rodi can encrypt payload and fake popular protocols like RTP, DNS, etc. and helps to prevent intelligent firewalls on your ISP from droping your packets

Some cases which are answered by the functional requirements and design

- you are behind NAT you do not control
- you are behind intelligent firewall or your ISP uses traffic shapers to drop P2P packets
- you want to hide your IP address and your ISP permits source IP address spoofing
- you want to stream data block by block
- you want application with CLI interface and script support
- you want to distribute huge files using some kind of small Java applet on your WEB site
- you prefer open source freeware free as in freedome

and one link more - http://www.p2pforums.com/viewtopic.php?p=85828#85828 (skip first 4 or 5 paragraphs)
« Last Edit: March 30, 2005, 03:04:22 AM by larytet » Logged
bitz
Advanced
***

Karma: +2/-0
Offline Offline

Posts: 160

Planet Peer Community


View Profile WWW
« Reply #5 on: March 30, 2005, 07:51:49 PM »

I guess perhaps it's far too early for any regular users... I'm not really sure about this project, seems like it has some questionable elements to it. Though there are some interesting aspects, I think I'll probably stay away from this one at least until it's been further developed. Good luck with it.
Logged

larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #6 on: March 30, 2005, 08:53:51 PM »

Quote
seems like it has some questionable elements to it

for example ?
Logged
tm
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 20

Planet Peer Community


View Profile
« Reply #7 on: April 02, 2005, 04:00:43 AM »

I guess perhaps it's far too early for any regular users... Though there are some interesting aspects, I think I'll probably stay away from this one at least until it's been further developed.
Rodi is not yet to the point of being user-friendly to the average P2P user - who has become accustomed to expecting point-and-click simplicity. Although testers are always welcome, potential users should keep in mind that at this stage Rodi still requires a high degree of command-line configuration (until the GUI gets completed). I would consider it to be still in an early stage of development - at least compared to the potential of what Rodi could eventually become.

Rodi is based on some of the most advanced and futuristic ideas and  concepts I have yet seen in any P2P, but to be honest, the average non-technical user might soon grow frustrated trying to operate what is still primarily a command-line client. But hopefully there are a few people out there who are comfortable operating in a command-line environment and would be interested in helping with its development.

As this website appears to be dedicated to secure, anonymous P2P networks, I think that Rodi is definitely one to watch, as it is in a class all by itself.
Logged
larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #8 on: April 03, 2005, 12:22:57 AM »

i am a strong beleiver in early testing. to make reliable product you have to start testing as early.

« Last Edit: April 03, 2005, 12:33:23 AM by larytet » Logged
ndurner
Advanced
***

Karma: +5/-1
Offline Offline

Posts: 176

Planet Peer Community


View Profile WWW
« Reply #9 on: April 03, 2005, 10:16:57 AM »

Quote
Many features, such as IP-spoofing for anonymity and packet-mimicking
If it relies entirely on that, Rodi is likely to fail.
In an ideal world, every router (other than backbones and carriers) would just drop packets that don't belong to the source net. That would help to stop DDoS attacks that also rely on packet spoofing.
So Rodi has a problem, if
 - the ISP drop packets to stop DDoS
 - the ISP/government/... just doesn't like Rodi, so it's vulnerable to blocking and censorship
Logged
tm
Newbie
*

Karma: +0/-0
Offline Offline

Posts: 20

Planet Peer Community


View Profile
« Reply #10 on: April 03, 2005, 12:50:54 PM »

It seems that most ISPs already prevent IP spoofing, or at least make it difficult, and in the future the proportion will likely increase. However, in a network in which some people can IP spoof and some people cannot - either because of technical reasons or improper configuration - there is still a degree of protection afforded to everyone because it makes it harder to prove that an IP address belonged to a user; there may always be some degree of doubt.

Those people capable of IP spoofing could serve as the original sources of files. Anti-P2P forces such as BayTSP have declared that they will be targeting original seeders, and that they have even developed software that will specifically do this. IP spoofing would be ideal in this situation, since it would allow any spoofing-capable original seeders to remain anonymous yet release files at full bandwidth.

The only other IP spoofing client (that works reliably) that I am aware of is SUMI, but it is not in active development although opensource, and is limited to serving files using an IRC server as the ACK proxy. SUMI serve is command line based.

I agree that the ability to IP-spoof will probably eventually disappear - especially if IP-spoofing P2P networks become popular -  but for today, it is an ability still available to many people, and is the only method of providing both anonymity and full-bandwidth file transfer at the same time.

It is an ongoing arms race. Early P2P networks could be easily defeated simply by blocking or throttling ports. 2nd-generation P2Ps allowed variable ports, or even port-roaming to defeat port blocks. Then blocking/throttling by packet-signature was developed. This will be defeated by Rodi (and other future P2P clients) that will use variable packet signatures.

I see that Rodi has its own forum section now. One other anonymizing technique worth mentioning is the idea of random distributed files. Scatterbrain Backup and Peerio Data use this technique, in which files are push-uploaded to other peers involuntarily, distributed across the network and stored in encrypted cache files. Although there are no peers serving as proxies, but because everyone is sharing files that they never personally downloaded - or are even aware exist on their hard drives - it would be hard to prove any personal responsibility/liability for sharing files. As with Rodi, Scatterbrain and Peerio offer less anonymity than traditional 'anonymous' networks, but because of the lack of proxies, much higher speed is possible.
Logged
larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #11 on: April 03, 2005, 07:00:01 PM »

To tm: very clear explanation of IP spoofing. thanks.

Rodi bouncer does not care what kind of traffic goes through it and in what direction. Technically speaking you can bounce data too. But in this case you should not expect high transfer rates. it's going to work probably slightly better than TCP based systems using proxys, because Rodi completely ignores latency, jitter and packet loss, but your upstream will be limited by upstream of the worst bouncer in the ring of bouncers1


Another thing to consider is WiFi networks. in the WiFi networks you can spoof IP and MAC. And one example more. When you are connected to LAN, for example, in University campus, nothing prevents you from spoofing IPs. Slowly IPSec will gain popularity and universities will drop unsigned packets2, but until then IP spoofing is a viable option.

Quote
Scatterbrain and Peerio offer less anonymity than traditional 'anonymous' networks, but because of the lack of proxies, much higher speed is possible.
Such networks at least double bandwidth usage, because they push data to people who does not need it and then make them to upload. Because some files tend to be extremely popular and most of files are not popular at all, such system wastes both disk space and bandwidth.

-------------------------------------------------
1one can use multiple bouncers in round roubin fashion. Because it does not matter for the leacher where packets come from seed (or publisher) can use multiple bouncers to proxy upstream.

2in IPv6 IPSec header is mandatory
« Last Edit: April 04, 2005, 04:02:18 PM by larytet » Logged
larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #12 on: April 03, 2005, 07:09:44 PM »

P.S. To ndurner. tm put it correctly. this is ongoing process of improving on both sides. i tend to think that eventually Internet will fork.

One free of everything "interesting" and other underground, where you can find "stuff".

The underground Internet will use LANs in communities, WiFi and satelite connections. Today you can install unportected WiFi modem and be hacked and in some circumstances it can be not that bad, if you catch my drift.
« Last Edit: April 12, 2005, 01:25:55 AM by larytet » Logged
crypton
Elite
*****

Karma: +9/-0
Offline Offline

Posts: 1677



View Profile
« Reply #13 on: April 04, 2005, 05:20:16 PM »

can i use rodi timple without cli version.?

i have test i cant run rodi timple without rodi CLIi . must need the CLI version run to can use rodi Timple.

can you make a version where ist both in one?

Or must need in future both to use rodi?

Can use rodi multisourcing Downloads?
« Last Edit: April 04, 2005, 05:26:15 PM by defnax » Logged


larytet
Advanced
***

Karma: +0/-0
Offline Offline

Posts: 143

Planet Peer Community


View Profile
« Reply #14 on: April 04, 2005, 05:50:10 PM »

Quote
can i use rodi timple without cli version.?
No. Timple is just a graphical (i do my best here) front end for the Rodi engine.

Quote
can you make a version where ist both in one?
It's possible and actually it's rather simple, but i would like to remain with two applications. Rodi Core requires about 20M of RAM and consumes no CPU. The moment you started download you can close Timple and save two thirds of your system resources. The other interesting result of having two separate applications that can you can run Rodi Core on one PC and Timple (GUI) on another - you can control Rodi remotely. Imagine that you runf Rodi Core on high performance server and control it from your home using Timple.

Quote
Or must need in future both to use rodi?
I am not sure i understand the question. I hope that in the future somebody will come and write another graphical front end for Rodi dedicated for Windows XP or Linux, etc. I would glad to see many different Rodi clients. Keeping Rodi Core and GUI separated can help future developers. For example, one can start new GUI project wihtout rewriting RodiCore.


Quote
Can use rodi multisourcing Downloads?
Frankly, i did not try it. Existing code should support multiple downloads - up to 5 or something like that
P.S. i am trying multiple downloads. there is a bug in the latest release, give me a couple of hours or use previous 0.2.0A
Workarounds
- ignore error, when after download start Timple reports Failed to start download for "filename"  without specifying reason
- start download using CLI command dld start followed by hash of the file to download
I will relase fix in two hours
« Last Edit: April 04, 2005, 06:24:56 PM by larytet » Logged
Planet Peer - The anonymous networking community
   

 Logged
Pages: [1] 2   Go Up
  Print  
 
Jump to:  


Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!