[patricxu]

Dear all,

How does a new node join the KAD? Whether was it initialized by sending a KADEMLIA_BOOTSTRAP_REQUEST to a remote node specified by the user or connecting the nodes prestored in local file NODE.DAT? Thanks in advance.

Sincerely
Patric

[Nissenice]

patricxu, on Jul 19 2009, 05:55 AM, said:

Dear all,

How does a new node join the KAD? Whether was it initialized by sending a KADEMLIA_BOOTSTRAP_REQUEST to a remote node specified by the user or connecting the nodes prestored in local file NODE.DAT? Thanks in advance.

Sincerely
Patric

May I ask why you are asking these questions? What are you up to? Working on a new Chinese client?

patricxu, on Feb 28 2009, 08:09 AM, said:

Quote

I am having trouble compile Emule. I am using Visual Studio 2008. ...

I am facing the same problem,and have not resolved.

patricxu, on Jun 24 2009, 04:46 PM, said:

As discribed by the topic ("How Does A Node Generate Its Id In Kad?"), I am confused by the mechanism that generates the node ID for a single machine in KAD network.

patricxu, on Jul 16 2009, 04:23 AM, said:

I made a experiment of publishing resources via emule 0.49c at my lab. I pulished a file (with an unpopular name to ensure there are no more hosts in KAD publishing another file using the same name) at machine A. Then I try to download the file at machine B. What puzzled me most is that after a few hours I still can not download the file even though B can detect its existance in KAD, i.e. it has been published. Can you show me why and how to get it fixed?

[fox88]

Nissenice, on Jul 19 2009, 12:23 PM, said:

May I ask why you are asking these questions? What are you up to? Working on a new Chinese client?

Good question. Especially after reading this:

patricxu, on Jul 16 2009, 11:17 AM, said:

I can't tell whether my ID is high or low.

[patricxu]

fox88, on Jul 19 2009, 11:12 AM, said:

Nissenice, on Jul 19 2009, 12:23 PM, said:

May I ask why you are asking these questions? What are you up to? Working on a new Chinese client?

Good question. Especially after reading this:

patricxu, on Jul 16 2009, 11:17 AM, said:

I can't tell whether my ID is high or low.

I am working on the experiment for my paper which is going to bring up a new p2p mechanism similar with KAD.

[Nissenice]

patricxu, on Jul 20 2009, 08:30 AM, said:

I am working on the experiment for my paper which is going to bring up a new p2p mechanism similar with KAD.

Interesting! Sounds like a lot of hard work involved there though.

You know since yesterday I found out about your blog site, where I can find things like "random graphs" (origin), "kernel32.dll" (google translation), "DDOS" (origin) and a lot of references to "A performance evaluation of the Kad-protocol by René Brunner from Sinsheim". I would find it quite interesting if you could elaborate a bit more about how you intend to put this together for use in the experiment. Especially the ddos part...

fox88, on Jul 19 2009, 11:12 AM, said:

Nissenice, on Jul 19 2009, 12:23 PM, said:

May I ask why you are asking these questions? What are you up to? Working on a new Chinese client?

Good question. Especially after reading this:

patricxu, on Jul 16 2009, 11:17 AM, said:

I can't tell whether my ID is high or low.

Perhaps not so strange. It's many ID's involved; user hash, clientID - KadID - Hash(Kad), ID's calculated from your IP or recieved from server, Low and High. I bet I missed some.

This post has been edited by Nissenice: 20 July 2009 - 08:21 PM

[patricxu]

Nissenice, on Jul 20 2009, 07:55 PM, said:

patricxu, on Jul 20 2009, 08:30 AM, said:

I am working on the experiment for my paper which is going to bring up a new p2p mechanism similar with KAD.

Interesting! Sounds like a lot of hard work involved there though.

You know since yesterday I found out about your blog site, where I can find things like "random graphs" (origin), "kernel32.dll" (google translation), "DDOS" (origin) and a lot of references to "A performance evaluation of the Kad-protocol by René Brunner from Sinsheim". I would find it quite interesting if you could elaborate a bit more about how you intend to put this together for use in the experiment. Especially the ddos part...

Welcom to my blog. I am going to unveil the implication between user's ID which was generated by IP in KAD and user's Hash. I suspect user's hash=MD4(user's ID). Is this correct?

[Stulle]

huh? do you mean they are somehow related? that would be fairly senseless. user hash is totally random and iirc the same applies for KAD IDs...

This post has been edited by Stulle: 22 July 2009 - 09:08 AM

[patricxu]

Stulle, on Jul 22 2009, 10:08 AM, said:

huh? do you mean they are somehow related? that would be fairly senseless. user hash is totally random and iirc the same applies for KAD IDs...

Thanks. I found it just as what you said. Bust could you explain more broadly what is iirc. Does the following code generate the so-called user hash?
void CPrefs::Init(LPCTSTR szFilename)
{
m_uClientID.SetValueRandom();
...
}
CUInt128& CUInt128::SetValueRandom()
{
AutoSeededRandomPool rng;
byte byRandomBytes[16];
rng.GenerateBlock(byRandomBytes, 16);
SetValueBE( byRandomBytes );
return *this;
}

[Stulle]

iirc - If I Remember Correctly

anyway, the functions are pretty self-explaining. they create random values.

i am not firm with any of the KAD code and conception because i just never bothered much about it. anyway, if you truly want to write something indepth about the code you should probably improve your c++ skill so you can figure out what is done reading the code. i doubt anybody will give you a full documentation if there exists none because writing will take plenty of time.

[patricxu]

Stulle, on Jul 22 2009, 04:23 PM, said:

iirc - If I Remember Correctly

anyway, the functions are pretty self-explaining. they create random values.

i am not firm with any of the KAD code and conception because i just never bothered much about it. anyway, if you truly want to write something indepth about the code you should probably improve your c++ skill so you can figure out what is done reading the code. i doubt anybody will give you a full documentation if there exists none because writing will take plenty of time.

Yes, you are right. Thank you anyway.