headers
FYP 2005 | 3 Oct 2005 08:21
Picon

about peer_id

After I sent the http get request to the tracker, it only gives me the IP and port. However, I expect that the tracker can give me the peer id. May be I show u my http get request:
If I set the no_peer_id from 1 to 0 , it gives me an error message: " d14:failure reason56:unsupported tracker protocol, please upgrade your cliente "
What can I do in order to get the peer id from the tracker.
Is there anythings wrong with it ?
 
 
Thx !!!
_______________________________________________
BitTorrent mailing list
BitTorrent <at> lists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/bittorrent
Permalink | Reply | 4 comments |
headers
Olaf van der Spek | 3 Oct 2005 09:24
Picon

Re: about peer_id

On 10/3/05, FYP 2005 <hkucsfyp2005 <at> gmail.com> wrote:
> After I sent the http get request to the tracker, it only gives me the IP
> and port. However, I expect that the tracker can give me the peer id. May be

Why do you expect that?

> I show u my http get request:
> http://tk.greedland.net/announce?info_hash=8R%D0%05%F4%01%C9%C3%F71%D6%17%0D%DA%BD9V%D1%B87&peer_id=exbc%009%5C%E8%5C%22%09%B3%B5%FC%EC%05%9ACDj&port=10015&natmapped=1&localip=192.168.40.105&uploaded=0&downloaded=0&left=25165824&numwant=200&compact=0&no_peer_id=1&key=26881&event=started

Just wondering, why compact=0?

> If I set the no_peer_id from 1 to 0 , it gives me an error message: "
> d14:failure reason56:unsupported tracker protocol, please upgrade your
> cliente "

Looks like XBT Tracker. :)

> What can I do in order to get the peer id from the tracker.
> Is there anythings wrong with it ?

Yes. Your expectations. :)
Don't expect to get peer_ids wrong the tracker.
Permalink | Reply | 0 comments |
headers
FYP 2005 | 4 Oct 2005 19:31
Picon

Re: about peer_id

> After I sent the http get request to the tracker, it only gives me the IP
> and port. However, I expect that the tracker can give me the peer id. May be

>Why do you expect that?
  Because I read the BT Specification, it tells me that the tracker can respond me the peer id.
 
Compact = 0 because I don't want to accept a compact response from the tracker. I tried to use compact =1 , but it gives me some "Strange" words. However, if i use compact =0 , it can show me port and IP without giving me some "Strange" words.
 
> If I set the no_peer_id from 1 to 0 , it gives me an error message: "
> d14:failure reason56:unsupported tracker protocol, please upgrade your > cliente "

>Looks like XBT Tracker. :)
> What can I do in order to get the peer id from the tracker.
> Is there anythings wrong with it ?

>Yes. Your expectations. :)
>Don't expect to get peer_ids wrong the tracker.
 
Sorry, I don't understand what you mean . Can you tell me more ?
 
 
In fact, I am implementing a small programme to communicate with other peers...
I just want to get the information of the peers instead of download or upload anything to them...
First of all, I need to get the peer id from the tracker in order to communicate with them.
However, I only can get the IP and port of the peers from the tracker.
Can you give me some ideas on getting the peer id of the peers from the tracker ? 
 
 
Thank you for your reply . I look forward to hearing from you.
 
 
Regards,
Robbie


 

_______________________________________________
BitTorrent mailing list
BitTorrent <at> lists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/bittorrent
Permalink | Reply | 2 comments |
headers
Olaf van der Spek | 4 Oct 2005 20:03
Picon

Re: about peer_id

On 10/4/05, FYP 2005 <hkucsfyp2005 <at> gmail.com> wrote:
> > After I sent the http get request to the tracker, it only gives me the IP
> > and port. However, I expect that the tracker can give me the peer id. May
> be
>
> >Why do you expect that?
>   Because I read the BT Specification, it tells me that the tracker can
> respond me the peer id.

But the protocol has been improved and peer_id isn't returned anymore
when compact or no_peer_id are enabled.

> Compact = 0 because I don't want to accept a compact response from the
> tracker. I tried to use compact =1 , but it gives me some "Strange" words.
> However, if i use compact =0 , it can show me port and IP without giving me
> some "Strange" words.
>
> >Yes. Your expectations. :)
> >Don't expect to get peer_ids wrong the tracker.
>
> Sorry, I don't understand what you mean . Can you tell me more ?
>
>
> In fact, I am implementing a small programme to communicate with other
> peers...
> I just want to get the information of the peers instead of download or
> upload anything to them...
> First of all, I need to get the peer id from the tracker in order to
> communicate with them.

Why do you need the peer id to communicate with them?
As far as I know you don't need it. Just connect to them and send
info_hash and *your* peer id.

> However, I only can get the IP and port of the peers from the tracker.
> Can you give me some ideas on getting the peer id of the peers from the
> tracker ?
Permalink | Reply | 0 comments |
headers
Brian Dessent | 5 Oct 2005 01:19
Favicon

Re: about peer_id

FYP 2005 wrote:

> Compact = 0 because I don't want to accept a compact response from the
> tracker. I tried to use compact =1 , but it gives me some "Strange"
> words. However, if i use compact =0 , it can show me port and IP
> without giving me some "Strange" words.

The 'compact=1' protocol significantly decreases the bandwidth that the
tracker spends.  Many trackers these days refuse clients that do not use
the compact protocol, returning an error telling the user to get a
non-brain-dead client.  If you are writing a BT client that doesn't
support compact, don't expect it to be used much, and don't expect
tracker operators to be very happy about it.

The response that you get with compact=1 is simply a list of IP
addresses (4 bytes each) and port (2 bytes) each in binary.  It's
exactly what you need.  The peer_id is superfluous and just a waste of
bandwidth.

Brian
Permalink | Reply | 0 comments |
headers
Anthony | 5 Oct 2005 07:14
Picon

about choking algorithm

Dear all,

I have got a question about the choking algorithm and optimistic
unchoke. My understanding is that each peer will normally unchoke 4
peers each time, and one more with optimistic unchoking mechanims. So
there will be at most 5 peers unchoked by a peer. Is that right? Since
choking will be held each 10 seconds, will a peer choose a peer among
the 5 unchoked peers or the 4 normal peers without the optimistic
unchoked one. Once it choke a peer, does it random choose a peer to
unchoke or choose a unchoked peer who is uploading to it now and may
be choked last iteration?

Thanks for your reading.
Permalink | Reply | 1 comment |
headers
etienne deleflie | 5 Oct 2005 09:07

seeding from same machine as tracker

Hello,

I expect this is a common question .... but I have searched and searched 
and cant find the answer.

I have a static IP on my home ADSL account.... I have a tracker on a 
server (within my home network), on port 8060.... and I have configured 
my firewall on my router to redirect all 8060 stuff to the machine that 
has the tracker. The static IP allocated to my modem is not visible to 
any computers on my network.

I want to seed files on the same machine as the tracker. So I create a 
torrent using my announce URL... then I try to download (and hence seed) 
that torrent straight away.... with this

btdownloadheadless my.torrent --ip xxx.xxx.xxx.xxx      

where xxx.xxx.xxx.xxx is the static IP allocated to my ADSL account .... 
This does not work ... my tracker does not report any seeded files and 
stdout gives lots of error 111.

Can anyone shed light on seeding from the same machine as the tracker ?

Etienne
Permalink | Reply | 8 comments |
headers
Olaf van der Spek | 5 Oct 2005 09:37
Picon

Re: seeding from same machine as tracker

On 10/5/05, etienne deleflie <et <at> lalila.net> wrote:
> where xxx.xxx.xxx.xxx is the static IP allocated to my ADSL account ....
> This does not work ... my tracker does not report any seeded files and
> stdout gives lots of error 111.

What is error 111?
Permalink | Reply | 7 comments |
headers
Arnaud Legout | 5 Oct 2005 10:21
Picon
Picon
Favicon

Re: about choking algorithm

Hi,
you can have a look at the paper here:
http://hal.inria.fr/inria-00000156/en

in section II.C we describe the choke algorithm of the mainline client.

If you want to be aware of all the small details, I urge you to have a
look at the file
choker.py that is in the source code of the mainline client. This is the
code for the choke algorithm.

Regards,
Arnaud.
Permalink | Reply | 0 comments |
headers
etienne deleflie | 5 Oct 2005 11:16

Re: seeding from same machine as tracker

Olaf van der Spek wrote:

>On 10/5/05, etienne deleflie <et <at> lalila.net> wrote:
>  
>
>>where xxx.xxx.xxx.xxx is the static IP allocated to my ADSL account ....
>>This does not work ... my tracker does not report any seeded files and
>>stdout gives lots of error 111.
>>    
>>
>
>What is error 111?
>  
>
<urlopen *error* (*111*, 'Connection refused')> *...*

>_______________________________________________
>BitTorrent mailing list
>BitTorrent <at> lists.ibiblio.org
>http://lists.ibiblio.org/mailman/listinfo/bittorrent
>
>
>  
>
Permalink | Reply | 6 comments |

Gmane