beepcore-java-users — Discussion list for developers using beepcore-java library

Re: [Beepcore-java-users] Tuning Profiles and Beepcore

[Will B. <wbe...@od...>]

So I looked at some more of what was going on, the session is being
terminated on the client side by a BeeepException being thrown from
Channel.postFrame(), stating:
Incorrect sequence number: was 306; expecting
119

This occurs (with those exact numbers) whenever I run Bing to connect to
Beepd and after the Tunnel profile has been used, before the first
nullecho message has been sent.

thanks,

William Berriel

On Mon, 10 Mar 2003, William Berriel wrote:

> Yep, I'm getting those profiles sent, the client sends out a message to
> start the new channel, the server replies to that message with the name
> of the profile (as it does with Beepd and Bing using TLS or no privacy,
> when the nullecho actually works) and then the client does nothing with it.
>
> William Berriel
>
> William J. Mills wrote:
>
>  > Are you getting the expected profiles in the greeting sent
>  > after the tuning reste?  It's possible you don't
>  > have the profile set to be availabel in the new state...
>  >
>  > -bill
>  >
>  > On Mon, Mar 10, 2003 at 01:33:28AM -0800, William Berriel wrote:
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Beepcore-java-users mailing list
> Bee...@li...
> https://lists.sourceforge.net/lists/listinfo/beepcore-java-users
>

Re: [Beepcore-java-users] Tuning Profiles and Beepcore

[William J. M. <wm...@es...>]

Ok, so that was my only good idea for the moment :)


On Mon, Mar 10, 2003 at 04:14:53PM -0800, William Berriel wrote:
> Yep, I'm getting those profiles sent, the client sends out a message to 
> start the new channel, the server replies to that message with the name 
> of the profile (as it does with Beepd and Bing using TLS or no privacy, 
> when the nullecho actually works) and then the client does nothing with it.
> 
> William Berriel
> 
> William J. Mills wrote:
> 
>  > Are you getting the expected profiles in the greeting sent
>  > after the tuning reste?  It's possible you don't
>  > have the profile set to be availabel in the new state...
>  >
>  > -bill
>  >
>  > On Mon, Mar 10, 2003 at 01:33:28AM -0800, William Berriel wrote:
> 
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Beepcore-java-users mailing list
> Bee...@li...
> https://lists.sourceforge.net/lists/listinfo/beepcore-java-users

Re: [Beepcore-java-users] Tuning Profiles and Beepcore

[William B. <wbe...@hm...>]

Yep, I'm getting those profiles sent, the client sends out a message to 
start the new channel, the server replies to that message with the name 
of the profile (as it does with Beepd and Bing using TLS or no privacy, 
when the nullecho actually works) and then the client does nothing with it.

William Berriel

William J. Mills wrote:

 > Are you getting the expected profiles in the greeting sent
 > after the tuning reste?  It's possible you don't
 > have the profile set to be availabel in the new state...
 >
 > -bill
 >
 > On Mon, Mar 10, 2003 at 01:33:28AM -0800, William Berriel wrote:

Re: [Beepcore-java-users] Tuning Profiles and Beepcore

[William J. M. <wm...@es...>]

Are you getting the expected profiles in the greeting sent
after the tuning reste?  It's possible you don't
have the profile set to be availabel in the new state...

-bill

On Mon, Mar 10, 2003 at 01:33:28AM -0800, William Berriel wrote:
> Hi,
> 
> For the purposes of creating a Tunnel Profile, I need to have the 
> session perform a tuning reset while not actually changing the socket. I 
> have used pretty much the exact code found in the two TLS profiles for 
> the tuning reset, yet when I use Beepd and Bing (unchanged, except for 
> the names of the profiles used), the client does not recognize the reply 
> to it's request to start an echo profile after the the tuning reset has 
> occurred. Can anyone give me hints as to why this is happening?
> 
> The code for the startTunnel and startChannel functions are enclosed. 
> startTunnel is used by a client to start a tunnel, startChannel is 
> called by the session, I believe.
> 
> thanks,
> 
> William Berriel

>  public void startChannel(Channel channel, String encoding, String data)
>         throws StartChannelException
>     {
>         TCPSession session = (TCPSession)channel.getSession();
>         String replyData = null;
>         if(data != null) {
>             if(data.equalsIgnoreCase(EMPTY) || data.equalsIgnoreCase(EMPTY2)){
>                 replyData = OK;
>                 try {
>                     this.begin(channel, URI, replyData);
> 
>                     // Consider the Profile Registry
>                     ProfileRegistry preg = session.getProfileRegistry();
>                     preg.removeStartChannelListener(URI);
> 
>                     Hashtable hash = new Hashtable();
>                     hash.put("TUNNEL", "true");
> 
>                     SessionTuningProperties tuning =
>                         new SessionTuningProperties(hash);
> 
>                     // Cause the session to be recreated and reset
>                     this.complete(channel, session.getLocalCredential(),
>                           session.getPeerCredential(), tuning,
>                                   preg, session.getSocket());
> 
>                 } catch (Exception e) {
>                     System.err.println("We got a problem in tuning reset : " +
>                                        e.getMessage());
> 
>                 }
>                 System.out.println("tunnel made");
>             }
>         }
>         System.err.println("Setting message listener to this");
>         channel.setMessageListener(this);
> 
>     }
> 
>   public TCPSession startTunnel(TCPSession session, String data)
>         throws BEEPException
>     {
>         // final argument is the messagelistener to handle the incoming
>         // messages, change this to ourselves once our profile can handle
>         // non piggy-backed messages
>         Channel ch = startChannel(session, URI, false, data, null);
> 
>         String startData = ch.getStartData();
> 
>         if (startData != null){
>             if(!startData.equalsIgnoreCase(OK)){
>                 System.err.println("We got back a bad message: " +
>                                    startData);
>                 throw new BEEPException("Bad Start Data : " + startData);
>             }
> 
>             Hashtable hash = new Hashtable();
>             hash.put("TUNNEL", "true");
>             SessionTuningProperties tuning=
>                 new SessionTuningProperties(hash);
> 
>             return (TCPSession) reset(session,
>                                       session.getLocalCredential(),
>                                       session.getPeerCredential(),
>                                       tuning,
>                                       session.getProfileRegistry(),
>                                       session.getSocket());
> 
> 
>         }
>         System.err.println("Didn't get an OK! ");
>         throw new BEEPException("Didn't get ok!");
>     }
> 
> 

[Beepcore-java-users] Tuning Profiles and Beepcore

[William B. <wbe...@hm...>]

Hi,

For the purposes of creating a Tunnel Profile, I need to have the 
session perform a tuning reset while not actually changing the socket. I 
have used pretty much the exact code found in the two TLS profiles for 
the tuning reset, yet when I use Beepd and Bing (unchanged, except for 
the names of the profiles used), the client does not recognize the reply 
to it's request to start an echo profile after the the tuning reset has 
occurred. Can anyone give me hints as to why this is happening?

The code for the startTunnel and startChannel functions are enclosed. 
startTunnel is used by a client to start a tunnel, startChannel is 
called by the session, I believe.

thanks,

William Berriel

Re: [Beepcore-java-users] Tunnel Profile

[Tunnel T. <ae...@cs...>]

On Sat, Mar 08, 2003 at 03:32:29PM -0600, Harsh Daharwal wrote:
> Huston,
> 
> Is Tunnel Profile implemented into beepcore? If not, is there
> any plan for doing so?
> 
> I'm trying to solve a peer to peer communication issue when peers
> are behind NATs/Firewalls. Tunnel Profile is one of things I'm
> looking at, although it does not solve the problem completely.
> 
> Regards,
> -- 
> Harsh Daharwal
> IOS Networks Inc.
> http://www.iosnetworks.com
> 

Hi,

The Tunnel Profile for beepcore-Java is currently under development.
We are working hard to get something working.  Expect something more solid
around May of this year.

--
----------------------------
|       Nick Hertl         |
|   Harvey Mudd College    |
| Aerospace Clinic Manager |
----------------------------

[Beepcore-java-users] Tunnel Profile

[Harsh D. <hda...@io...>]

Huston,

Is Tunnel Profile implemented into beepcore? If not, is there
any plan for doing so?

I'm trying to solve a peer to peer communication issue when peers
are behind NATs/Firewalls. Tunnel Profile is one of things I'm
looking at, although it does not solve the problem completely.

Regards,
-- 
Harsh Daharwal
IOS Networks Inc.
http://www.iosnetworks.com

Re: [Beepcore-java-users] Pbm with TLS Session

[Huston <hu...@us...>]

I'm thinking the library should do this for you, what do you think?

--Huston

----- Original Message -----
From: "Thierry Hanot" <th...@in...>
To: <bee...@li...>
Sent: Thursday, March 06, 2003 1:17 AM
Subject: [Beepcore-java-users] Pbm with TLS Session


> Hello
> I'm currently using a TLS JSEE Profile for secure my channels in a
sessions.
> In the server side , i want to detect session close. The problem is when
the
> session
> is created with the TCPSessionCreator.listen() the channel with the TLS
> profile is open
> And the session is replace by a new secure session .
> Like i have no access to this session , i cannot add a session listener on
> it for being informed of session close.
> Like the other channels are optionals depending of the services requested
by
> the client i cannot simply manage the session close by the channel
> creation/destruction .
> The simplest way is, i think, to set the listener of the old session ( the
> TCP session )
> to the new  secure session ( in the TLSJSSEProfile ). But perhaps  there
is
> a simplest way
> To do it . Is somebody can help me with that ?
>
> Best regard
>
> Thierry Hanot
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Etnus, makers of TotalView, The
debugger
> for complex code. Debugging C/C++ programs can leave you feeling lost and
> disoriented. TotalView can help you find your way. Available on major UNIX
> and Linux platforms. Try it free. www.etnus.com
> _______________________________________________
> Beepcore-java-users mailing list
> Bee...@li...
> https://lists.sourceforge.net/lists/listinfo/beepcore-java-users
>
>

[Beepcore-java-users] Pbm with TLS Session

[Thierry H. <th...@in...>]

Hello 
I'm currently using a TLS JSEE Profile for secure my channels in a sessions.
In the server side , i want to detect session close. The problem is when the
session 
is created with the TCPSessionCreator.listen() the channel with the TLS
profile is open 
And the session is replace by a new secure session . 
Like i have no access to this session , i cannot add a session listener on
it for being informed of session close. 
Like the other channels are optionals depending of the services requested by
the client i cannot simply manage the session close by the channel
creation/destruction .
The simplest way is, i think, to set the listener of the old session ( the
TCP session ) 
to the new  secure session ( in the TLSJSSEProfile ). But perhaps  there is
a simplest way 
To do it . Is somebody can help me with that ?

Best regard 

Thierry Hanot 

Re: [Beepcore-java-users] Apex

[Jason G. <ja...@in...>]

http://netlab.cis.temple.edu/apex/  this seems to be a
partial java implementation of it.

At 03:02 PM 3/5/2003, Girish Wadhwani wrote:

>I am interested in developing open source Apex & Apex
>services on Beepcore. I would prefer to join forces
>with anyone already working or interested in working
>on Apex.

Re: [Beepcore-java-users] Apex

[Girish W. <gw...@ya...>]

I am interested in developing open source Apex & Apex
services on Beepcore. I would prefer to join forces
with anyone already working or interested in working
on Apex.

-Girish 


--- Huston <hu...@us...> wrote:
> > Is anyone implenting/considering implementing Apex
> or
> > any
> > of the other profiles for beepcore java or C?
> 
> I know there are a few under development (based on
> emails I have received)
> but I don't keep track of them.
> 
> > Also, I haven't been able to glean much
> information
> > about the status of the project other than it is
> beta.
> > Are there any parts of BEEP unimplemented?
> 
> The core functionality is complete in both
> libraries. There are things like
> "features" support which has not been added.
> 
> --Huston
> 
> 


__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

Re: [Beepcore-java-users] Apex

[Huston <hu...@us...>]

> Is anyone implenting/considering implementing Apex or
> any
> of the other profiles for beepcore java or C?

I know there are a few under development (based on emails I have received)
but I don't keep track of them.

> Also, I haven't been able to glean much information
> about the status of the project other than it is beta.
> Are there any parts of BEEP unimplemented?

The core functionality is complete in both libraries. There are things like
"features" support which has not been added.

--Huston

[Beepcore-java-users] Apex

[Girish W. <gw...@ya...>]

Hello,

Is anyone implenting/considering implementing Apex or
any 
of the other profiles for beepcore java or C?

Also, I haven't been able to glean much information
about the status of the project other than it is beta.
Are there any parts of BEEP unimplemented?

TIA,
Girish  

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com

Re: [Beepcore-java-users] Multiple sends doesn't seem to work because of ThreadedMessageListener?

[Huston <hu...@us...>]

Jon,

> I've just subscribed and I note from the archive that a problem like
> this has been noted, but perhaps not resolved.
>
> My code tries:
>
> channel.sendMSG(msg1,reply1);
> channel.sendMSG(msg2,reply2);
> channel.sendMSG(msg3,reply3);
>
> then I try to read the replies. I've also tried using reply1 three times.
>
> The problem appears to be with ThreadedMessageListener, which assigns
> the received message to a field before spawning a thread to process the
> message.
>
> public void receiveMSG(Message message) // simplified
> {
>    this.message = (MessageMSG)message;
>    callbackQueue.execute(this);
> }
>
> If another message is received before the first is processed it
> overwrites the field. For this reason, I get "Incorrect message number:
> was 3; expecting 1".
>
> I'm pulling down the CVS copy to try this fix:
>
> public void receiveMSG(Message message) // simplified
> {
>    callbackQueue.execute(new MessageProcessor(message));
> }
>
> where a new MessageProcessor inner class holds the original run().
>
> Comments welcome; should this parallel messaging in a single channel
> work in the first place? - I can't find any examples.

Thanks for the email! This is a known problem and I have a solution in mind.

What you are doing is valid, it is ok to send another request before
receiving the reply. The problem is that the listener is suppose to process
the requests and reply to them in order and Channel is currently not
enforcing this. This is a bug that was introduced when
ThreadedMessageListener was added.

--Huston

[Beepcore-java-users] Multiple sends doesn't seem to work because of ThreadedMessageListener?

[Jon P. <jo...@el...>]

I've just subscribed and I note from the archive that a problem like 
this has been noted, but perhaps not resolved.

My code tries:

channel.sendMSG(msg1,reply1);
channel.sendMSG(msg2,reply2);
channel.sendMSG(msg3,reply3);

then I try to read the replies. I've also tried using reply1 three times.

The problem appears to be with ThreadedMessageListener, which assigns 
the received message to a field before spawning a thread to process the 
message.

public void receiveMSG(Message message) // simplified
{
   this.message = (MessageMSG)message;
   callbackQueue.execute(this);
}

If another message is received before the first is processed it 
overwrites the field. For this reason, I get "Incorrect message number: 
was 3; expecting 1".

I'm pulling down the CVS copy to try this fix:

public void receiveMSG(Message message) // simplified
{
   callbackQueue.execute(new MessageProcessor(message));
}

where a new MessageProcessor inner class holds the original run().

Comments welcome; should this parallel messaging in a single channel 
work in the first place? - I can't find any examples.

Regards, Jon.

Re: [Beepcore-java-users] BEEPException.getCause()

[Huston <hu...@us...>]

> It there any reason why the core code does not set the cause for
> BEEPException's raised?

There is a reason, just not a very good one ;-) cause was added to
BEEPException after that code was written. I have just been adding the cause
to the code as I find places where it is missing. I will added it to
TCPSession.

--Huston

[Beepcore-java-users] BEEPException.getCause()

[Andrew K. <ak...@Sm...>]

It there any reason why the core code does not set the cause for
BEEPException's raised? An issue I have is that a telnet connection to the
BEEP server will cause the TCPSessionCreator to throw an execption, but I
can't easily determine what the cause was, so I have to loop and recreate
the TCPSessionCreator.

    while (keepRunning)
    {
      try
      {
        TCPSessionCreator.listen(listenPort, reg);
      }
      catch (BEEPException be)
      {
        log.error("BEEP exception thrown", be);
      }
    }

Now if I run another instance of the server on the same port, I obviously
get a BindException thrown internally, but I can't trap that as all I
receive in the above code is a BEEPException with no cause set. This is one
example but there are probably many other different situations, where the
underlying cause might need to be available?

RE: [Beepcore-java-users] How do I get the IP address of the remo te client?

[Andrew K. <ak...@Sm...>]

Doh! Apologies, it's late.... (not). Cast the session as a TCPSession then
getSocket().........

-----Original Message-----
From: Andrew Keedle [mailto:ak...@Sm...]
Sent: 23 January 2003 17:31
To: 'bee...@li...'
Subject: [Beepcore-java-users] How do I get the IP address of the remote
client?



I'm struggling to determine the IP address of the remote client from within
my server profile. How do I do it? 

[Beepcore-java-users] How do I get the IP address of the remote client?

[Andrew K. <ak...@Sm...>]

I'm struggling to determine the IP address of the remote client from within
my server profile. How do I do it?

Re: [Beepcore-java-users] Patch to use a ThreadPool in TCPSession

[Huston <hu...@us...>]

> I've posted a patch to use a ThreadPool for processNextFrame() in
> TCPSession. It uses the concurrent.jar library. It seems to improve things
> on my Linux box, the CPU usage drops, new processes aren't created, and
the
> I/O rate seems to improve. It's a bit of a noddy patch but any thoughts
> anybody. I couldn't see any problems it might create, but I don't know the
> core internals.

Thanks! I'll give it a try and if it tests out I'll add it to cvs.

--Huston

[Beepcore-java-users] Patch to use a ThreadPool in TCPSession

[Andrew K. <ak...@Sm...>]

I've posted a patch to use a ThreadPool for processNextFrame() in
TCPSession. It uses the concurrent.jar library. It seems to improve things
on my Linux box, the CPU usage drops, new processes aren't created, and the
I/O rate seems to improve. It's a bit of a noddy patch but any thoughts
anybody. I couldn't see any problems it might create, but I don't know the
core internals.

Re: [Beepcore-java-users] More on the reply problem

[Huston <hu...@us...>]

I was able to reproduce this error, I will also take a look at this.

Thanks for the test code!

--Huston


----- Original Message -----
From: <136...@su...>
To: "Kevin Kress" <kk...@my...>
Cc: "beepcore-java-users" <bee...@li...>
Sent: Monday, January 06, 2003 7:29 AM
Subject: [Beepcore-java-users] More on the reply problem


Kevin,

I have run the program on two other machines, a Pentium 4 1.8 Ghz running
Windows XP and a Pentium 4 800 Mhz running Linux. All the machines that I
have tested on have J2SDK 1.4.1 installed. I have looked further into the
problem and am more confused now than ever. I have narrowed it down to one
line of code, the recieveFrame method recieves a frame (of MSG type) and
then processes it. It then calls the listener.recieveMSG(m) method. Just
before this line is executed I output the message number and it is correct,
but after in the message listener (TestBombProfile) the message recieved by
recieveMSG has an incorrect number.

The error occurs at about line 500 in the Channel.java file (receiveFrame
method). I added three debug outputs two in the receiveFrame method - one at
the begining of the MSG type part (it outputs the message number) and the
other just before the listener.recieveMSG(m) method is called (about line
500). I also put an output at the begining of the recieveMSG method (Also
outputs the message number). Here follows what output should be recieved
(for message number 4):
                 recieveFrame: recieved frame no 4
                 Channel: removed message 4
                 TestBombProfile: recieved message no. 4

I will attatch the results that I got as well as the example files. You will
notice that I got very strange results, the first few messages don't seem to
call the listener but the after a few they call the listener but with the
wrong object (they call it the number of times that it was missed with the
latest message object). To me it looks like a standard method call passing a
parameter, it does not make sence to me that it does what it does. The error
I revieved on the client side is:
          SEVERE: Incorrect message number: was 3; expected 1
and then the client aborted the session.

Before you think that it has something to to do with the listener not being
registered in time, I have seen it happen where the first few messages are
correct but then the problem occurs later on.

What pc are you running, what operating system, and what version of JDK is
installed? I connot understand how I get the same problem everytime but you
don't.

Please try the test again and let me know how it goes.

Thanks,
Sean

Attached files:
bombServer.java => the server for the test
bombCient.java => the client for the test
TestBombProfile.java => the profile for the test
out_server.txt => the output I recieved on the server side
out_client.txt => the output I recieved on the client side

[Beepcore-java-users] More on the reply problem

[De K. SJ <136...@su...>]

Kevin,
=20
I have run the program on two other machines, a Pentium 4 1.8 Ghz =
running Windows XP and a Pentium 4 800 Mhz running Linux. All the =
machines that I have tested on have J2SDK 1.4.1 installed. I have looked =
further into the problem and am more confused now than ever. I have =
narrowed it down to one line of code, the recieveFrame method recieves a =
frame (of MSG type) and then processes it. It then calls the =
listener.recieveMSG(m) method. Just before this line is executed I =
output the message number and it is correct, but after in the message =
listener (TestBombProfile) the message recieved by recieveMSG has an =
incorrect number.
=20
The error occurs at about line 500 in the Channel.java file =
(receiveFrame method). I added three debug outputs two in the =
receiveFrame method - one at the begining of the MSG type part (it =
outputs the message number) and the other just before the =
listener.recieveMSG(m) method is called (about line 500). I also put an =
output at the begining of the recieveMSG method (Also outputs the =
message number). Here follows what output should be recieved (for =
message number 4):
                 recieveFrame: recieved frame no 4
                 Channel: removed message 4
                 TestBombProfile: recieved message no. 4
=20
I will attatch the results that I got as well as the example files. You =
will notice that I got very strange results, the first few messages =
don't seem to call the listener but the after a few they call the =
listener but with the wrong object (they call it the number of times =
that it was missed with the latest message object). To me it looks like =
a standard method call passing a parameter, it does not make sence to me =
that it does what it does. The error I revieved on the client side is:
          SEVERE: Incorrect message number: was 3; expected 1
and then the client aborted the session.
=20
Before you think that it has something to to do with the listener not =
being registered in time, I have seen it happen where the first few =
messages are correct but then the problem occurs later on.
=20
What pc are you running, what operating system, and what version of JDK =
is installed? I connot understand how I get the same problem everytime =
but you don't.
=20
Please try the test again and let me know how it goes.
=20
Thanks,
Sean
=20
Attached files:
bombServer.java =3D> the server for the test
bombCient.java =3D> the client for the test
TestBombProfile.java =3D> the profile for the test
out_server.txt =3D> the output I recieved on the server side
out_client.txt =3D> the output I recieved on the client side

[Beepcore-java-users] Bomb tests

[De K. SJ <136...@su...>]

Kevin,
=20
Thanks for checking out the problem. I ran the program on my pc again =
and got the same results as I got previously. I checked that I had the =
most up to date channel.java file (I did) - that is where the problem =
appears to be (in the processFrame method). What os are you running, I =
am using an old windows NT machine, I know that there are differences =
between some of the JDK implementations of threads on different =
platforms. I will try to run it on a linux machine in the other office. =
I'll look in the code and see if I can find out what is happening, maybe =
I will find it. The problem seems to be that the pc is running too fast =
because if I put a few println statements in the the problem goes away, =
I using a pentium 2 350Mhz with 320 MB RAM. I am sure that your computer =
is much faster so I would of thought that the problem would definately =
of appeared on your pc. Let me know if you have any further ideas.
=20
Thanks,
Sean

Re: [Beepcore-java-users] multiple messages

[Kevin K. <kk...@my...>]

I ran your example, but I did not get the errors that you reported.=20
This could be caused by a few different things.  If it is a race
condition (which it sounds like it is), my machine could be too
fast/slow and not able to reproduce the issue.  Another option is that
this has been fixed in CVS.  What version are you running?

While I was looking through the ethereal logs for these test results I
noticed something odd having to do with SEQ messages.  I am not very
familiar with RFC 3081 ( need to reread it), but 2 things struck me as
odd.  First is the fact that reply messages do not produce SEQ messages,
from Listener to Initiator, only Initiator to Listener, is that right?

Second was the fact that the SEQ messages were coming together in blocks
instead of directly following the message that was sent.  I have posted
the logs from this test so that people can view them and verify they are
correct, but it seems to me that the SEQ messages are coming later than
they should.  Now granted this could just be an artifact of the way the
capture occurs and it does not appear to effect the sequence numbers
correctness, but I figured I should mention it anyway.

log from ethereal for bombServer and bombClient (colored by me so there
might be a few mis colored lines):=20

http://www.myslo.com/~kkress/beeplog.html

--Kevin Kress

On Mon, 2002-12-23 at 00:43, De Kock SJ <136...@su...> wrote:
> Hi,
> =20
> I have been testing beep by flooding the session with messages and get SE=
VERE errors. I use a message listener that just sends a string back. On the=
 sending peer's side I register the NullReplyListener. I do not wait until =
I have recieved reply before I sending the next message. After a few messag=
es have been sent then I get a error
>           SEVERE: Incorrect message number: was 16; expection 1
> Is it correct to assume that I can send miltiple messages on a single cha=
nnel without waiting for a reply between every message?
> =20
> If I don't send any reply back then the error does not occur.
> =20
> I have attatched a demo program to illistrate the problem.
> =20
> Thanks,
> Sean de Kock
> =20
> Files:
>     bombClient.java (package: org.beepcore.beep.example) - client
>     bombServer.java (package: org.beepcore.beep.example) - server
>     TestBombProfile.java (package: org.beepcore.beep.profile.test) - prof=
ile
--=20
Kevin Kress <kk...@my...>