Re: kandula and sandesha

[Jaliya Ekanayake <jn...@gmail.com>]

Hi All,

Let me clarify some of the issues regarding discussed deployment problem.

1. The MalformedURI problem is fixed and please use the source from the svn.

2. Sandesha DOES NOT use a separate addressing implementation. It is the 
same addressing implementation that we have in apache. However when we 
initially develop Sandesha we did not have the addressing implementation and 
hence we did set the addressing headers directly by Sandesha. The main 
reason for doing this is to facilitate the request/response dual transport 
requirement for WS-RM as axis +addressing handlers alone cannot support 
this.

So let's try to get this two deployed correctly.

The main client side component in Sandesha is the RMSender which acts as a 
transport to the axis engine. So the global addressing handler in client 
side should not cause any problems. However the problem comes when the same 
client-config.wsdd is used again when the RMSender use Call object to send 
control messages such as CreateSequence yielding the addressing handler to 
get invoked twice.

So the first problem that we need to solve is to find a way to inform Call 
object to ignore the client-config.wsdd which is in our classpath for the 
second time. Does any of you  know how to do this? I remember there was a 
method but could not remember now.

Once this is fixed let's try to deploy them correctly.

Thanks,
Jaliya






----- Original Message ----- 
From: "Hannes Erven" <ha...@erven.at>
To: "Benjamin Schmeling" <Be...@gmx.de>
Cc: "Dasarath Weeratunge" <dw...@purdue.edu>; <ka...@ws.apache.org>
Sent: Thursday, July 13, 2006 4:16 PM
Subject: Re: kandula and sandesha


> Hi Benjamin,
>
>
>> Nevertheless there is a Nullpointer
>> Exception with that version when using global addressing handlers. I 
>> don't know why this exception is thrown.
>>
>> {http://xml.apache.org/axis/}stackTrace:java.lang.NullPointerException
>>    at 
>> org.apache.sandesha.server.RMMessageProcessorIdentifier.getMessageProcessor(RMMessageProcessorIdentifier.java:53)
>
> That doesn't sound like kandula has something to do with it, since there 
> is no occurence of "kandula" or "addressinghandlers" in the stack trace 
> ;-)
>
> What did the sandesha people tell you about this?
>
>
> -hannes
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: kandula-dev-unsubscribe@ws.apache.org
> For additional commands, e-mail: kandula-dev-help@ws.apache.org
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: axis-dev-unsubscribe@ws.apache.org
For additional commands, e-mail: axis-dev-help@ws.apache.org