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

[Beepcore-java-users] Re: SessionEventListener question

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

Mike:

Looks like you're still using the old version. I'm not sure
if event driven callback was working in it. Only time 'fireEvent'
seems to be called in 'Session' is when it is being terminated.

I'd suggest you to use new version which has recently been released.

Speaking of new version. I downloaded it recently but I don't see
java doc for '...core.event' package.

If you insist on using old version you may want to register your
main class with profile rather than with 'Session'. If your think
multiple classes may register with profile for receiving start channel
event, you can introduce and interface/abstract class which are
implemented/derived by classes which are interested in such an event.
Or just have a 'Collection' of registered callbacks.

Hope that helps,

HD
----------------------------------------------------------------------
Harsh Daharwal
IOS Networks, Inc.
----------------------------------------------------------------------

---Original Message----------

 Hi,
       I have a class that listens for connections using a Profile from
 another class.  After a client connects to start the session I would like
to
 be notified when the Profile receives a startChannel event.  I do not want
 to do any hard coding in my profile that would require the name of my main
 class, for obvious reasons.  So, I've done the following...

 In my main class that listens for connections:
 After I establish a BEEP session, I make a call to registerForEvent like so
 (also, my main class implements SessionEventListener):
 session.registerForEvent(this, SessionEvent.CHANNEL_OPENED_EVENT_CODE);

 This compiles fine.  Now in my profile class I added the following code to
 startChannel:
 channel.getSession().fireEvent(SessionEvent.CHANNEL_OPENED_EVENT_CODE,
 null);

 My hope was that when my profile receives a startChannel event it can then
 notify my main class that a channel has started.  Unfortunatly, I get a
 compile error because fireEvent is protected:
 fireEvent(int,java.lang.Object) has protected access in
 org.beepcore.beep.core.Session

 channel.getSession().fireEvent(SessionEvent.CHANNEL_OPENED_EVENT_CODE,
 null);


 Perhaps I'm going at this the wrong way?  Any suggestions/work-arounds?

 Thanks for your time,
 -Mike

[Beepcore-java-users] SessionEventListener question

[Riggio, M. <Mic...@un...>]

Hi,
      I have a class that listens for connections using a Profile from
another class.  After a client connects to start the session I would like to
be notified when the Profile receives a startChannel event.  I do not want
to do any hard coding in my profile that would require the name of my main
class, for obvious reasons.  So, I've done the following...

In my main class that listens for connections:
After I establish a BEEP session, I make a call to registerForEvent like so
(also, my main class implements SessionEventListener):
session.registerForEvent(this, SessionEvent.CHANNEL_OPENED_EVENT_CODE);

This compiles fine.  Now in my profile class I added the following code to
startChannel:
channel.getSession().fireEvent(SessionEvent.CHANNEL_OPENED_EVENT_CODE,
null);

My hope was that when my profile receives a startChannel event it can then
notify my main class that a channel has started.  Unfortunatly, I get a
compile error because fireEvent is protected:
fireEvent(int,java.lang.Object) has protected access in
org.beepcore.beep.core.Session
 
channel.getSession().fireEvent(SessionEvent.CHANNEL_OPENED_EVENT_CODE,
null);


Perhaps I'm going at this the wrong way?  Any suggestions/work-arounds?

Thanks for your time,
-Mike

Re: [Beepcore-java-users] Need for queueing MSGs

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

There is a 1 to 1 relationship between MSG frames and RPY frames.  It is
up to the framework to keep track of the MSGs sent out and wait for the
corresponding RPY/ERR/ANS frames.  Even if there are no reply listeners
the frameworks needs to keep track of which messages is has sent to make
sure the session integrity is still in place.

--KMK

On Sun, 2002-09-15 at 14:43, Harsh Daharwal wrote:
>=20
> When a MSG is sent out corresponding MessageStatus is added to
> sentMessageQueue. I'm wondering if it's really necessary if
> ReplyListener for MSG is NullReplyListener or null. The fact
> that one is registering a null reply listener means one does
> not care about RPY. Why should framework?
>=20
> Harsh Daharwal
>=20
>=20
>=20
>=20
> -------------------------------------------------------
> 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
>=20
--=20
Kevin Kress
kk...@my...

GnuPG Key ID: 92949032
Fingerprint : B7F9 2B08 6FC8 35CA 5B64  BD34 6A8B 325C 9294 9032

See Keyserver: http://www.keyserver.net/en/findkey.html
(search for "Kevin Kress" or "0x92949032")

Get GnuPG or PGP at http://www.pgpi.org

[Beepcore-java-users] Need for queueing MSGs

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

When a MSG is sent out corresponding MessageStatus is added to
sentMessageQueue. I'm wondering if it's really necessary if
ReplyListener for MSG is NullReplyListener or null. The fact
that one is registering a null reply listener means one does
not care about RPY. Why should framework?

Harsh Daharwal

[Beepcore-java-users] sending a broadcast message

[Michael J. R. <mjr...@co...>]

Hi,
      Is there an easy way to broadcast a message to all active
sessions/channels, aside from going through some variation of a loop and
sending the message one at a time?

Thanks!
-Mike

Re: [Beepcore-java-users] beepd dies on user logout

[Andrew N. <an...@ec...>]

Huston wrote:
> 
> I use debian and haven't seen any problems. I would try turning the logging
> level up to see if that gives any indication as to what is going wrong. If
> that doesn't help I will try to recreate (and fix) the problem if you give
> me the details.

Yeah, well it is the first time I've ever seen this is a java program. 
It has got something to do with a Bastille setting.  I have a laptop 
running Mandrake 8.2 in lax security mode and I don't see this problem.
However, I also don't see this problem with Jakarta Tomcat on the 
problem box.  Go figure!

Anyway, I'll look at turning the logging up.  I guess that is done by 
setting a log service and then setting the verbosity up to high level?

-andy

Re: [Beepcore-java-users] beepd dies on user logout

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

> Ulf Tidstrand wrote:
> > I suspect you are using Windows NT/2000/XP and are trying to run the
> > server as a Windows NT service?
>
> Actually, this is on a Mandrake Linux 8.2 server in high security mode.
>   It may be something to do with Mandrake's version of Bastille, but I
> couldn't find anything that looked wrong.

I use debian and haven't seen any problems. I would try turning the logging
level up to see if that gives any indication as to what is going wrong. If
that doesn't help I will try to recreate (and fix) the problem if you give
me the details.

--Huston

Re: [Beepcore-java-users] Concurrency

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

> I have a 'Profile' (implementing MessageListener) that can be used
> in multiple 'Session's. Since there is a thread per 'Session', I'm
> assuming same 'Profile' object will be used by multiple session
> threads. So the responsibility of thread-safety falls on profile
> author. Right?

Yes, if you share the MessageListener between multiple channels.

> Is there a way to have a separate instance of profile per
> session/thread. If not, could somebody educate me on it's pros/cons?

Profile, StartChannelListener, and MessageListener are separate interfaces
so you can have separate instances per listener, session, and channel
respectfully.

If you want to avoid locking in your MessageListener then you will need to
allocate a MessageListener per channel since messages can be received in
parallel in different channels.

--Huston

Re: [Beepcore-java-users] beepd dies on user logout

[Ulf T. <d9...@dt...>]

On Mon, 9 Sep 2002, Andrew Newton wrote:

> Actually, this is on a Mandrake Linux 8.2 server in high security mode. 
>   It may be something to do with Mandrake's version of Bastille, but I 
> couldn't find anything that looked wrong.

OK - I don't have any experience of Mandrake (have used Debian mostly).
But if anyone dares to run Beepd under Windows, my advice could possibly
help... :)

/Ulf

Re: [Beepcore-java-users] beepd dies on user logout

[Andrew N. <an...@ec...>]

Ulf Tidstrand wrote:
> I suspect you are using Windows NT/2000/XP and are trying to run the
> server as a Windows NT service?

Actually, this is on a Mandrake Linux 8.2 server in high security mode. 
  It may be something to do with Mandrake's version of Bastille, but I 
couldn't find anything that looked wrong.

-andy

[Beepcore-java-users] Concurrency

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

I have a 'Profile' (implementing MessageListener) that can be used
in multiple 'Session's. Since there is a thread per 'Session', I'm
assuming same 'Profile' object will be used by multiple session
threads. So the responsibility of thread-safety falls on profile
author. Right?

Is there a way to have a separate instance of profile per
session/thread. If not, could somebody educate me on it's pros/cons?

Thanks,
Harsh

Re: [Beepcore-java-users] beepd dies on user logout

[Ulf T. <d9...@dt...>]

I suspect you are using Windows NT/2000/XP and are trying to run the
server as a Windows NT service?

If that is the case, this is a common behaviour for many Java
applications. As soon as you try to run them as a service under Windows
and then logout, the OS sends a signal to terminate the JVM. This has to
be handled in some way - for example by using some sort of service
wrapper that catches the signals from the OS before it reaches the JVM.

A couple of wrappers available:

JNT:
http://www.eworksmart.com/JNT/

Javants (available in the JBoss CVS repository):
http://www.jboss.org/

Wrapper:
http://sourceforge.net/projects/wrapper/

Personally, I have only tried the Javants wrapper in combination with the
JBoss EJB server. Although it required some modifications to work, it did
what it should do. There are probably more wrappers available.

/Ulf Tidstrand

On Mon, 9 Sep 2002, Andrew Newton wrote:

> Ok.  I've got a bit of an odd problem I was wondering if anybody else
> has seen.  I'm developing a server using beepcore.  Whenever I start the
> server, and then logout, it terminates.  I even use the nohup command
> and that still doesn't stop it.   This even happens if I'm root.
> 
> I reproduced this with the Beepd example as well.
> 
> I even switched out the JDK's (1.3 for 1.4).  Same problem.  So I
> downloaded Jakarta Tomcat to see if it was a general Java problem.
> Tomcat does not exhibit this behaviour.
> 
> As anybody seen this?
> 
> -andy
> 

/Ulf

--
More info: http://www.dtek.chalmers.se/~d97ulf/

"Hey, we're different!" - Jay Minor, Creator of the Amiga
http://www.amiga.com/

[Beepcore-java-users] beepd dies on user logout

[Andrew N. <an...@ec...>]

Ok.  I've got a bit of an odd problem I was wondering if anybody else
has seen.  I'm developing a server using beepcore.  Whenever I start the
server, and then logout, it terminates.  I even use the nohup command
and that still doesn't stop it.   This even happens if I'm root.

I reproduced this with the Beepd example as well.

I even switched out the JDK's (1.3 for 1.4).  Same problem.  So I
downloaded Jakarta Tomcat to see if it was a general Java problem.
Tomcat does not exhibit this behaviour.

As anybody seen this?

-andy

[Beepcore-java-users] Secure Your Future

[Smith <sm...@ew...>]

 



 
 
 
Don't Wait any longer!
Get Started Today!  
       

AmeriOneCard - this card is perfect for you if you have no credit or damaged credit and allows you to improve your financial freedom. No credit check or checking account is required for this card. 
Buy Right.com® - a membership program offering a $5,000 line of credit for purchasing any of thousands of great brand-name products from Buy Right catalogs online and from the convenience of home - all at 0 % APR for 20 months. A $150 bonus certificate is applied toward the member's minimum payments as built-in assurance of a good credit "quick start".


 We have a lot more to offer. To know more about other products like Home Mortgage, Emergency Loans Credit Cards,  please visit www.ewaytowealth.com.   



************
To Unsubscribe, Click the link below 
Unsubscribe from mailing list
************
© Copyright 2001-2002, All Rights Reserved.
Terms and conditions - Privacy Policy
Violators will be Prosecuted

Re: [Beepcore-java-users] Trouble compiling

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

Do you recommend downloading it separately. Which dir it should reside
in if I want to have it in the same dir where beepcore framework to
distribution typically keeps it?

What I'm trying to do is to build a beep kit (but with latest code)
similar to the first release(0.9.06) so that I can replace the latter?

Thanks

Huston said:
>> I synchronized my view with latest from CVS repository. Having trouble
>> compiling tls-ptls package.
>> Ant log is attached.
>
> It looks like you don't have the PureTLS jar in the classpath for your
> project. Also, you may need to rename the tls\com directory to tls\COM
> when using eclipse. I have entered a support request with the
> sourceforge team to rename this directory in cvs.
>
> --Huston
>
>
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by: OSDN - Tired of that same old
> cell phone?  Get a new here for FREE!
> https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390
> _______________________________________________
> Beepcore-java-users mailing list
> Bee...@li...
> https://lists.sourceforge.net/lists/listinfo/beepcore-java-users


Harsh Daharwal
(972)424-6931

Re: [Beepcore-java-users] Trouble compiling

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

> I synchronized my view with latest from CVS repository. Having trouble
> compiling tls-ptls package.
> Ant log is attached.

It looks like you don't have the PureTLS jar in the classpath for your
project. Also, you may need to rename the tls\com directory to tls\COM when
using eclipse. I have entered a support request with the sourceforge team to
rename this directory in cvs.

--Huston

[Beepcore-java-users] Trouble compiling

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

I synchronized my view with latest from CVS repository. Having trouble
compiling tls-ptls package.
Ant log is attached.

Thanks,
Harsh

Re: [Beepcore-java-users] Message as String

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

> When 'receiveMSG' is called, can payload be assumed to have been
> completely received? If not what are the steps client is expected
> to take so that a complete payload is available for processing?
>
> Could you also answer this question w.r.t. ReplyListener's receiveXXX
> methods?

It is the same for both receiveMSG and receiveXXX. The method is called when
the first frame of a message is received. The isComplete() method on
InputDataStream can be used to determine when the last frame is received
(also the waitForNextSegment and read methods will return null and 0
respectfully after the last frame has been 'read').

One thing to note is that if the channel's window fills up, more data will
not be received until the already received data is 'read'. So I would not
recommend waiting until all of the frames have been received before reading
any of the data unless you are absolutely sure that the message will be less
than the window size.

--Huston

Re: [Beepcore-java-users] Message as String

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

When 'receiveMSG' is called, can payload be assumed to have been
completely received? If not what are the steps client is expected
to take so that a complete payload is available for processing?

Could you also answer this question w.r.t. ReplyListener's receiveXXX
methods?

Thanks.

Huston said:
>> - I have a profile that implements 'MessageListener' interface.
>>   When receiveMSG is called, a 'Message' object is passed to
>>   this profile. How do I get the payload of the message in
>>  'String' format?
>
> The simplest solution is to get an InputStream from the InputDataStream
> and wrap that in a Reader. You could also get the BufferSegments from
> the InputDataStream and convert the bytes to characters.
>
>> - to compile the latest load as a 'release' version, I'm
>>   assuming I have to use 'dist' target for ant?
>
> dist will build the .tgz and .zip file that are uploaded to
> sourceforge. If by 'release' you mean a non-debug version you need to
> set the debug property in the ant build file appropriately.
>
> --Huston


Harsh Daharwal
(972)424-6931

Re: [Beepcore-java-users] Message as String

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

> - I have a profile that implements 'MessageListener' interface.
>   When receiveMSG is called, a 'Message' object is passed to
>   this profile. How do I get the payload of the message in
>  'String' format?

The simplest solution is to get an InputStream from the InputDataStream and
wrap that in a Reader. You could also get the BufferSegments from the
InputDataStream and convert the bytes to characters.

> - to compile the latest load as a 'release' version, I'm
>   assuming I have to use 'dist' target for ant?

dist will build the .tgz and .zip file that are uploaded to sourceforge. If
by 'release' you mean a non-debug version you need to set the debug property
in the ant build file appropriately.

--Huston

[Beepcore-java-users] Message as String

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

- I have a profile that implements 'MessageListener' interface.
  When receiveMSG is called, a 'Message' object is passed to
  this profile. How do I get the payload of the message in
 'String' format?

- to compile the latest load as a 'release' version, I'm
  assuming I have to use 'dist' target for ant?

Harsh

Re: [Beepcore-java-users] Beep exception about message number

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

> I am receiving this exception from Channel.postFrame, line 756:
>
>
>
> org.beepcore.beep.core.Session.terminate: Incorrect message number: was 6;
> expecting 5.

Probably a bug, is it reproducible and if so can you provide a description
of how to reproduce it?

Thanks,
--Huston

[Beepcore-java-users] Beep exception about message number

[Serge A. <sA...@in...>]

I am receiving this exception from Channel.postFrame, line 756:

 

org.beepcore.beep.core.Session.terminate: Incorrect message number: was 6;
expecting 5.

 

Is it a bug?

 

 

code:

                if (frame.getMsgno() != mstatus.getMsgno()) {

                    throw new BEEPException("Incorrect message number: was "

                                            + frame.getMsgno()

                                            + "; expecting "

                                            + mstatus.getMsgno());

                }

Serge

Re: [Beepcore-java-users] Waiting for a message in a client application

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

Huston, did my thread stuff ever get integrated into the main line?

-bill


On Thu, Aug 29, 2002 at 09:38:26AM -0600, Huston wrote:
> > Thank you; I have understood that now.
> > Is there any document describing the threading model implemented by the
> API?
> > I had to look in the sources to understand when a listening thread is
> > started.
> 
> There is not right now mostly because I'm in the process of improving it.
> Currently there is one thread per socket for receiving (until nio support is
> added) and a thread pool that is used for calling MessageListeners.
> 
> --Huston
> 
> 
> 
> -------------------------------------------------------
> 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] Waiting for a message in a client application

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

> Thank you; I have understood that now.
> Is there any document describing the threading model implemented by the
API?
> I had to look in the sources to understand when a listening thread is
> started.

There is not right now mostly because I'm in the process of improving it.
Currently there is one thread per socket for receiving (until nio support is
added) and a thread pool that is used for calling MessageListeners.

--Huston