sipdht-devel

[Sipdht-devel] Sipdht2 - gsim problem

[<ser...@gm...>]

Hi all,
I've been trying to install sipdht2 on my ubuntu but I am experiencing a
problem and just think that maybe you could give me a hand with it.

I have compiled and installed libxpp, libsipdht2, sofia and all required
libs and also gsim-sipdht2 (finally) without errors, but when trying to
execute gsim it didn't find libxpp.so (i have it in usr/local/lib).
The thing is that I'm sure the path is correctly set inside configure file
of gsim....

Did anyone have the same problem? DO you know how can i fix this?

Thank you in advance.

--=20
Sergi Puigvent=F3s

Re: [Sipdht-devel] Sipdht2 - gsim problem

[<vic...@gm...>]
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Re: [Sipdht-devel] Sipdht2 - gsim problem

[Enrico M. <enr...@te...>]

Sergi Puigvent=F3s wrote:
> I have compiled and installed libxpp, libsipdht2, sofia and all require=
d
> libs and also gsim-sipdht2 (finally) without errors, but when trying to=

> execute gsim it didn't find libxpp.so (i have it in usr/local/lib).
> The thing is that I'm sure the path is correctly set inside configure
> file of gsim....
>=20
> Did anyone have the same problem? DO you know how can i fix this?

Probably you can make it work just telling the linker where to find
libxpp.so, as /usr/local/lib is not a standard path.  There are many
ways to do that; a simple one is to set the LD_LIBRARY_PATH environment
variable with something like:
$ export LD_LIBRARY_PATH=3D/usr/local/lib

--=20
Ciao,
Enrico

Re: [Sipdht-devel] Sipdht2 - gsim problem

[<ser...@gm...>]

Thanks to both of you! It works fine now!



On Jan 4, 2008 12:36 PM, Enrico Marocco <enr...@te...>
wrote:

> Sergi Puigvent=F3s wrote:
> > I have compiled and installed libxpp, libsipdht2, sofia and all require=
d
> > libs and also gsim-sipdht2 (finally) without errors, but when trying to
> > execute gsim it didn't find libxpp.so (i have it in usr/local/lib).
> > The thing is that I'm sure the path is correctly set inside configure
> > file of gsim....
> >
> > Did anyone have the same problem? DO you know how can i fix this?
>
> Probably you can make it work just telling the linker where to find
> libxpp.so, as /usr/local/lib is not a standard path.  There are many
> ways to do that; a simple one is to set the LD_LIBRARY_PATH environment
> variable with something like:
> $ export LD_LIBRARY_PATH=3D/usr/local/lib
>
> --
> Ciao,
> Enrico
>
>


--=20
Sergi

[Sipdht-devel] GSIM-DHT die error

[Adetola O. <ao...@es...>]

Hello,

I am getting a common error when i try to kill nodes in the simulator.
Please find below the errors in 2 different scenarios 

died: sip:peer43@p2p
died: sip:peer99@p2p
[sip:peer58@p2p] I'm authoritative for neighbor: sip:peer43@p2p
[sip:peer36@p2p] I'm authoritative for neighbor: sip:peer99@p2p

died: sip:peer52@p2p
died: sip:peer51@p2p
[sip:peer21@p2p] I'm authoritative for neighbor: sip:peer52@p2p
Segmentation fault

what do you think can be causing this?

Thanks and Regards,

Tola

Re: [Sipdht-devel] GSIM-DHT die error

[stefano <st....@gm...>]

Hi tola,

we need a gdb backtrace for finding a segfault.

Moreover these two conditions help us a lot:
- max 5/10 peers;
- repeatable procedure for causing the segfault.

Ciao,
Stefano

2008/2/11, Adetola Oredope <ao...@es...>:
>
> Hello,
>
> I am getting a common error when i try to kill nodes in the simulator.
> Please find below the errors in 2 different scenarios
>
> died: sip:peer43@p2p
> died: sip:peer99@p2p
> [sip:peer58@p2p] I'm authoritative for neighbor: sip:peer43@p2p
> [sip:peer36@p2p] I'm authoritative for neighbor: sip:peer99@p2p
>
> died: sip:peer52@p2p
> died: sip:peer51@p2p
> [sip:peer21@p2p] I'm authoritative for neighbor: sip:peer52@p2p
> Segmentation fault
>
> what do you think can be causing this?
>
> Thanks and Regards,
>
> Tola
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Sipdht-devel mailing list
> Sip...@li...
> https://lists.sourceforge.net/lists/listinfo/sipdht-devel
>

[Sipdht-devel] SIPDHT Experiments

[Oredope, A. <ao...@es...>]

Hello All,

 

Thanks a lot for all your help so far, I really do appreciate it.

 

I am actually trying to run some experiments to measure the amount of
packets that are exchanged during churn using sipdht.

I am having problems achieving this using the simulator; because I don't
think I can kill and add peers at the same time. The simulator actually
breaks when trying to do this. What am I doing wrong?

Also as an alternative, I am thinking of using real sipdht peers, i.e.
running like 10 instances of a sipdht peers on a single node and then
maybe across 10 to 20 nodes.

What do you think about this? Also if I follow this approach is there a
way a seeing an overall map of the overlay? Is there a way of getting
exact overall measurements of time, packets exchanged and active
sessions?

I also thinking doing this with the next weeks, do you think this is a
feasible time or do I need leave out more time.

 

I look forward from hearing from you all and all suggestions will be
highly appreciated.

 

Thanks a lot in advance.

 

Regards,

 

Tola 

 

Re: [Sipdht-devel] SIPDHT Experiments

[Oredope, A. <ao...@es...>]

Hi,

 

I sent an email earlier on (please see below)

Please is there anyone out there that can help?

I look forward to hearing from you.

 

Regards,

 

Tola

 

________________________________

From: sip...@li...
[mailto:sip...@li...] On Behalf Of
Oredope, Adetola
Sent: 12 February 2008 22:14
To: sip...@li...
Subject: [Sipdht-devel] SIPDHT Experiments

 

Hello All,

 

Thanks a lot for all your help so far, I really do appreciate it.

 

I am actually trying to run some experiments to measure the amount of
packets that are exchanged during churn using sipdht.

I am having problems achieving this using the simulator; because I don't
think I can kill and add peers at the same time. The simulator actually
breaks when trying to do this. What am I doing wrong?

Also as an alternative, I am thinking of using real sipdht peers, i.e.
running like 10 instances of a sipdht peers on a single node and then
maybe across 10 to 20 nodes.

What do you think about this? Also if I follow this approach is there a
way a seeing an overall map of the overlay? Is there a way of getting
exact overall measurements of time, packets exchanged and active
sessions?

I also thinking doing this with the next weeks, do you think this is a
feasible time or do I need leave out more time.

 

I look forward from hearing from you all and all suggestions will be
highly appreciated.

 

Thanks a lot in advance.

 

Regards,

 

Tola 

 

Re: [Sipdht-devel] SIPDHT Experiments

[Enrico M. <enr...@te...>]

Oredope, Adetola wrote:
> I am actually trying to run some experiments to measure the amount of
> packets that are exchanged during churn using sipdht.
> 
> I am having problems achieving this using the simulator; because I don’t
> think I can kill and add peers at the same time. The simulator actually
> breaks when trying to do this. What am I doing wrong?

Probably nothing, the code is still full of bugs, plus the simulator
tends to hang when running many nodes because of too many refresh
operations happening in the GUI.

> Also as an alternative, I am thinking of using real sipdht peers, i.e.
> running like 10 instances of a sipdht peers on a single node and then
> maybe across 10 to 20 nodes.

That may be a viable option.  I'd suggest to use csipdht2 which is the
lightest tool in the suite.  If you are interested in churn handling
only, I'd also suggest to try and disable outbound and gruu (don't
remember the command line option), which haven't been tested extensively.

> What do you think about this? Also if I follow this approach is there a
> way a seeing an overall map of the overlay? Is there a way of getting
> exact overall measurements of time, packets exchanged and active sessions?

No, running the peers as standalone processes there's no way to have an
overall picture of the CAN geometry.

> I also thinking doing this with the next weeks, do you think this is a
> feasible time or do I need leave out more time.

I really cannot answer this question.  What I just can say is that in
such a distributed and stressed scenario you'll certainly find many
bugs.  We'll help to fix them, but I really can say anything about timings.

> I look forward from hearing from you all and all suggestions will be
> highly appreciated.

I'm sorry for answering this late: it's been a busy busy week.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments

[Oredope, A. <ao...@es...>]

Hello Enrinco,

Thanks a lot for your help.
We are considering going back to the first version of SIPDHT (based on
chord) but I will keep you informed on how it goes.

Kind Regards,

Tola

-----Original Message-----
From: Enrico Marocco [mailto:enr...@te...] 
Sent: 18 February 2008 07:58
To: Oredope, Adetola
Cc: sip...@li...
Subject: Re: [Sipdht-devel] SIPDHT Experiments

Oredope, Adetola wrote:
> I am actually trying to run some experiments to measure the amount of
> packets that are exchanged during churn using sipdht.
> 
> I am having problems achieving this using the simulator; because I
don't
> think I can kill and add peers at the same time. The simulator
actually
> breaks when trying to do this. What am I doing wrong?

Probably nothing, the code is still full of bugs, plus the simulator
tends to hang when running many nodes because of too many refresh
operations happening in the GUI.

> Also as an alternative, I am thinking of using real sipdht peers, i.e.
> running like 10 instances of a sipdht peers on a single node and then
> maybe across 10 to 20 nodes.

That may be a viable option.  I'd suggest to use csipdht2 which is the
lightest tool in the suite.  If you are interested in churn handling
only, I'd also suggest to try and disable outbound and gruu (don't
remember the command line option), which haven't been tested
extensively.

> What do you think about this? Also if I follow this approach is there
a
> way a seeing an overall map of the overlay? Is there a way of getting
> exact overall measurements of time, packets exchanged and active
sessions?

No, running the peers as standalone processes there's no way to have an
overall picture of the CAN geometry.

> I also thinking doing this with the next weeks, do you think this is a
> feasible time or do I need leave out more time.

I really cannot answer this question.  What I just can say is that in
such a distributed and stressed scenario you'll certainly find many
bugs.  We'll help to fix them, but I really can say anything about
timings.

> I look forward from hearing from you all and all suggestions will be
> highly appreciated.

I'm sorry for answering this late: it's been a busy busy week.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments

[Adetola O. <ao...@es...>]

Hello All,

I am very aware that you no longer support the first version of sipdht..
But, is it possible that anyone may know how increase the maximum number
of tcp connections in the sipdht version 1, cos at the moment it is
limiting me to just 20 connection after which there is a callback
failure.

Thanks and Regards,

Tola


On Tue, 2008-02-19 at 10:06 +0000, Oredope, Adetola wrote:
> Hello Enrinco,
> 
> Thanks a lot for your help.
> We are considering going back to the first version of SIPDHT (based on
> chord) but I will keep you informed on how it goes.
> 
> Kind Regards,
> 
> Tola
> 
> -----Original Message-----
> From: Enrico Marocco [mailto:enr...@te...] 
> Sent: 18 February 2008 07:58
> To: Oredope, Adetola
> Cc: sip...@li...
> Subject: Re: [Sipdht-devel] SIPDHT Experiments
> 
> Oredope, Adetola wrote:
> > I am actually trying to run some experiments to measure the amount of
> > packets that are exchanged during churn using sipdht.
> > 
> > I am having problems achieving this using the simulator; because I
> don't
> > think I can kill and add peers at the same time. The simulator
> actually
> > breaks when trying to do this. What am I doing wrong?
> 
> Probably nothing, the code is still full of bugs, plus the simulator
> tends to hang when running many nodes because of too many refresh
> operations happening in the GUI.
> 
> > Also as an alternative, I am thinking of using real sipdht peers, i.e.
> > running like 10 instances of a sipdht peers on a single node and then
> > maybe across 10 to 20 nodes.
> 
> That may be a viable option.  I'd suggest to use csipdht2 which is the
> lightest tool in the suite.  If you are interested in churn handling
> only, I'd also suggest to try and disable outbound and gruu (don't
> remember the command line option), which haven't been tested
> extensively.
> 
> > What do you think about this? Also if I follow this approach is there
> a
> > way a seeing an overall map of the overlay? Is there a way of getting
> > exact overall measurements of time, packets exchanged and active
> sessions?
> 
> No, running the peers as standalone processes there's no way to have an
> overall picture of the CAN geometry.
> 
> > I also thinking doing this with the next weeks, do you think this is a
> > feasible time or do I need leave out more time.
> 
> I really cannot answer this question.  What I just can say is that in
> such a distributed and stressed scenario you'll certainly find many
> bugs.  We'll help to fix them, but I really can say anything about
> timings.
> 
> > I look forward from hearing from you all and all suggestions will be
> > highly appreciated.
> 
> I'm sorry for answering this late: it's been a busy busy week.
> 

Re: [Sipdht-devel] SIPDHT Experiments

[Enrico M. <enr...@te...>]

Adetola Oredope wrote:
> I am very aware that you no longer support the first version of sipdht..
> But, is it possible that anyone may know how increase the maximum number
> of tcp connections in the sipdht version 1, cos at the moment it is
> limiting me to just 20 connection after which there is a callback
> failure.

As far as I remember we didn't impose a limit to the number of tcp
connections.  I suspect it is something related to sofia-sip, but
probably some debug information could help us to better understand
what's going on.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments

[Oredope, A. <ao...@es...>]

Hello Enrico,

Thanks a lot for your quick response.

I have a script that starts 25 nodes but there is always a callback
failure from the 20th node onwards
I was initially thinking maybe the amount of tcp connections are limited
to 20, but I will appreciate you advise on this.

Please find attached the log files and the prompt output. Please also
fine below a section of the code I use in generating the nodes

I look forward to hear from you

Regards,

Tola


#Define Paramenters
a=127.0.0.1
c=127.0.0.
x=1
p=5060
s=1s
t=60
n=1
tn=10

		for ((x=1; x<=25; x++ )); do
		
		if [ $x -eq 1 ]; then
			
			sipdht -a $a -p $p -s$t &
			sleep 3s 
		
		else
	
			z=`expr $x - 1`	
			
			rm -f /tmp/sipdhtlog/$x.log; SIPDHT_DEBUG=9
TPORT_DUMP=/tmp/sipdhtlog/$x.log sipdht -a $c$x -p $p -s$t sip:$c$z:$p &
			sleep $s
			echo "node $x has Joined"
		fi
	done



	


-----Original Message-----
From: Enrico Marocco [mailto:enr...@te...] 
Sent: 20 February 2008 07:59
To: Oredope, Adetola
Cc: sip...@li...
Subject: Re: [Sipdht-devel] SIPDHT Experiments

Adetola Oredope wrote:
> I am very aware that you no longer support the first version of
sipdht..
> But, is it possible that anyone may know how increase the maximum
number
> of tcp connections in the sipdht version 1, cos at the moment it is
> limiting me to just 20 connection after which there is a callback
> failure.

As far as I remember we didn't impose a limit to the number of tcp
connections.  I suspect it is something related to sofia-sip, but
probably some debug information could help us to better understand
what's going on.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Enrico M. <enr...@te...>]

Oredope, Adetola wrote:
> Please attached the logs

Could you also tell us where to look for the error?

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Adetola O. <ao...@es...>]

According to my last email,

the errors starts to occur from the creation of the 20th node
its a callback failure error

Regards,

Tola

On Wed, 2008-02-20 at 14:16 +0100, Enrico Marocco wrote:
> Oredope, Adetola wrote:
> > Please attached the logs
> 
> Could you also tell us where to look for the error?
> 

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Enrico M. <enr...@te...>]

Adetola Oredope wrote:
> According to my last email,
> 
> the errors starts to occur from the creation of the 20th node
> its a callback failure error

Yes, I got it, I'm just having an hard time skimming half a meg of
compressed logs.  Can you please indicate in which file and at which
line I can see the error?

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Adetola O. <ao...@es...>]

In the output.log file, the last successful connection was indicated in
Line 2420. I think the error probably occured after this line because
sipdht was unable to invoke a callback. i.e node 20, 21, 22, 23, 24, and
25 refused to join the existing dht network
The sip log files for the corresponding nodes are 20.log  - 25. log

Kind Regards,

Tola 


On Wed, 2008-02-20 at 14:39 +0100, Enrico Marocco wrote:
> Adetola Oredope wrote:
> > According to my last email,
> > 
> > the errors starts to occur from the creation of the 20th node
> > its a callback failure error
> 
> Yes, I got it, I'm just having an hard time skimming half a meg of
> compressed logs.  Can you please indicate in which file and at which
> line I can see the error?
> 

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Enrico M. <enr...@te...>]

Adetola Oredope wrote:
> In the output.log file, the last successful connection was indicated in
> Line 2420. I think the error probably occured after this line because
> sipdht was unable to invoke a callback. i.e node 20, 21, 22, 23, 24, and
> 25 refused to join the existing dht network
> The sip log files for the corresponding nodes are 20.log  - 25. log

It seems like node 21 sends its REGISTER to node 20 when node 20 still
has not joined the overlay (it keeps getting redirected, probably
because the overlay isn't consistent yet).  This is probably due to the
fact that the interval between successive joins is much smaller than the
stabilization interval.  I'd suggest to try and use the same value for
both intervals, just to be sure it is not a problem related to tcp.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Oredope, A. <ao...@es...>]

Thanks a lot Enrico,

I have tried it as suggested but it does not still seem to fin node 19,
please look below

Trying with node sip:127.0.0.19:5060...
Trying with node sip:80@127.0.0.1:5060;user=node...
Trying with node sip:88@127.0.0.9:5060;user=node...
Trying with node sip:b0@127.0.0.10:5060;user=node...
Trying with node sip:b1@127.0.0.11:5060;user=node...
Trying with node sip:b2@127.0.0.12:5060;user=node...
callback failure invoked: 0

Regards,

Tola

-----Original Message-----
From: Enrico Marocco [mailto:enr...@te...] 
Sent: 20 February 2008 14:18
To: Oredope, Adetola
Cc: sip...@li...
Subject: Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

Adetola Oredope wrote:
> In the output.log file, the last successful connection was indicated
in
> Line 2420. I think the error probably occured after this line because
> sipdht was unable to invoke a callback. i.e node 20, 21, 22, 23, 24,
and
> 25 refused to join the existing dht network
> The sip log files for the corresponding nodes are 20.log  - 25. log

It seems like node 21 sends its REGISTER to node 20 when node 20 still
has not joined the overlay (it keeps getting redirected, probably
because the overlay isn't consistent yet).  This is probably due to the
fact that the interval between successive joins is much smaller than the
stabilization interval.  I'd suggest to try and use the same value for
both intervals, just to be sure it is not a problem related to tcp.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Enrico M. <enr...@te...>]

Oredope, Adetola wrote:
> I have tried it as suggested but it does not still seem to fin node 19,
> please look below
> 
> Trying with node sip:127.0.0.19:5060...
> Trying with node sip:80@127.0.0.1:5060;user=node...
> Trying with node sip:88@127.0.0.9:5060;user=node...
> Trying with node sip:b0@127.0.0.10:5060;user=node...
> Trying with node sip:b1@127.0.0.11:5060;user=node...
> Trying with node sip:b2@127.0.0.12:5060;user=node...
> callback failure invoked: 0

After a second look, I noticed that node 20 has the same id as node 11
("b1" for both, in the user part of the SIP URI).  That is due to the
weakness of the hash function and of course makes the join fail.  I've
made available the latest version of libsipdht and sipdht-tools where
the support for SHA-1 was certainly implemented; unfortunately I can't
remember if it was enabled by default or if some tricks were needed.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Oredope, A. <ao...@es...>]

Hello Enrico,

Thanks a lot for your help, it is highly appreciated.
I have tried compiling the new version for sipdht tools and there seems
to be an error with an undefined parameter (sipdht_peer_set_routing).
Please see error below

[root@psr106 sipdht-tools-0.4.1]# make
make  all-recursive
make[1]: Entering directory `/root/sipdht-tools-0.4.1'
Making all in src
make[2]: Entering directory `/root/sipdht-tools-0.4.1/src'
gcc  -g -O2   -o sipdht  sipdht.o  -lsipdht 
sipdht.o: In function `main':
/root/sipdht-tools-0.4.1/src/sipdht.c:286: undefined reference to
`sipdht_peer_set_routing'
collect2: ld returned 1 exit status
make[2]: *** [sipdht] Error 1
make[2]: Leaving directory `/root/sipdht-tools-0.4.1/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/root/sipdht-tools-0.4.1'
make: *** [all] Error 2

Regards,

Tola

-----Original Message-----
From: Enrico Marocco [mailto:enr...@te...] 
Sent: 20 February 2008 15:58
To: Oredope, Adetola
Cc: sip...@li...
Subject: Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

Oredope, Adetola wrote:
> I have tried it as suggested but it does not still seem to fin node
19,
> please look below
> 
> Trying with node sip:127.0.0.19:5060...
> Trying with node sip:80@127.0.0.1:5060;user=node...
> Trying with node sip:88@127.0.0.9:5060;user=node...
> Trying with node sip:b0@127.0.0.10:5060;user=node...
> Trying with node sip:b1@127.0.0.11:5060;user=node...
> Trying with node sip:b2@127.0.0.12:5060;user=node...
> callback failure invoked: 0

After a second look, I noticed that node 20 has the same id as node 11
("b1" for both, in the user part of the SIP URI).  That is due to the
weakness of the hash function and of course makes the join fail.  I've
made available the latest version of libsipdht and sipdht-tools where
the support for SHA-1 was certainly implemented; unfortunately I can't
remember if it was enabled by default or if some tricks were needed.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Enrico M. <enr...@te...>]

Oredope, Adetola wrote:
> Thanks a lot for your help, it is highly appreciated.
> I have tried compiling the new version for sipdht tools and there seems
> to be an error with an undefined parameter (sipdht_peer_set_routing).
> Please see error below

Which version of libsipdht are you using?

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Oredope, A. <ao...@es...>]

Hello Enrico,

I am using sipdht-tools-0.4.1 and libsipdht-0.7.0.
I have actually just done a clean install using the .deb packages on an
Ubuntu machine but I'm still encountering the same errors.
What are your opinions on this?
Also can please point me to the files responsible for generating the
nodes and creating the hash?

King Regards,

Tola

-----Original Message-----
From: Enrico Marocco [mailto:enr...@te...] 
Sent: 21 February 2008 07:41
To: Oredope, Adetola
Cc: sip...@li...
Subject: Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

Oredope, Adetola wrote:
> Thanks a lot for your help, it is highly appreciated.
> I have tried compiling the new version for sipdht tools and there
seems
> to be an error with an undefined parameter (sipdht_peer_set_routing).
> Please see error below

Which version of libsipdht are you using?

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Enrico M. <enr...@te...>]

Oredope, Adetola wrote:
> I am using sipdht-tools-0.4.1 and libsipdht-0.7.0.
> I have actually just done a clean install using the .deb packages on an
> Ubuntu machine but I'm still encountering the same errors.
> What are your opinions on this?
> Also can please point me to the files responsible for generating the
> nodes and creating the hash?

The hash function used by sipdht is implemented in the C function
"hash_function" in sipdht.c (sipdht-tools), and then set to the peer
through the "sipdht_peer_set_hash_cb" function.  How it works is quite
well documented in the "peer" module documentation, available at
http://sipdht.sourceforge.net/api/index.html.

I'd suggest to implement a better hash function (the current one just
sums the char values and then casts the result in 8 bits) and then
replace the current one with it.

-- 
Ciao,
Enrico

Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

[Oredope, A. <ao...@es...>]

Hello Enrico,

Thanks a lot for you help on this. I have been able to play around with
the hash function in which I discovered it parameters had char
properties which is quite small, thus causing repetition. I have change
this to long and also used the system time as a seed to generate random
numbers. Although there some errors while compling (of course!), it
still worked. Please see below the revised code.

static
void hash_function(url_t *url, unsigned long *buf, void *user_data) {
	unsigned long ans;
	const unsigned long *ptr;
	unsigned long tmp_yixuan;

	srand(time());

	for (ans = 0, ptr = url->url_host; *ptr != '\0'; ptr++)
		tmp_yixuan=1+(100000.0*rand()/(RAND_MAX+1.0));
		ans ^= (tmp_yixuan + *ptr);
	if (url->url_user)
		for (ptr = url->url_user; *ptr != '\0'; ptr++)
			tmp_yixuan=1+(100000.0*rand()/(RAND_MAX+1.0));
			ans ^= (tmp_yixuan + *ptr);
	if (url->url_port)
		for (ptr = url->url_port; *ptr != '\0'; ptr++)
			tmp_yixuan=1+(100000.0*rand()/(RAND_MAX+1.0));
			ans ^= (tmp_yixuan + *ptr);

	buf[0] = ans;
}

At the moment this is working but there are like one duplicate out of
every 30, which is fair enough for now. I have already now simulate over
200 nodes
Do you think this changes may affect anytime else in the full
architecture? Or is there anything else I am missing or should what out
for?

Thanks a lot again for you help..

Kind Regards

Tola


-----Original Message-----
From: Enrico Marocco [mailto:enr...@te...] 
Sent: 21 February 2008 10:34
To: Oredope, Adetola
Cc: sip...@li...
Subject: Re: [Sipdht-devel] SIPDHT Experiments -- with attachment

Oredope, Adetola wrote:
> I am using sipdht-tools-0.4.1 and libsipdht-0.7.0.
> I have actually just done a clean install using the .deb packages on
an
> Ubuntu machine but I'm still encountering the same errors.
> What are your opinions on this?
> Also can please point me to the files responsible for generating the
> nodes and creating the hash?

The hash function used by sipdht is implemented in the C function
"hash_function" in sipdht.c (sipdht-tools), and then set to the peer
through the "sipdht_peer_set_hash_cb" function.  How it works is quite
well documented in the "peer" module documentation, available at
http://sipdht.sourceforge.net/api/index.html.

I'd suggest to implement a better hash function (the current one just
sums the char values and then casts the result in 8 bits) and then
replace the current one with it.

-- 
Ciao,
Enrico