[Design of JBoss Web Services] - Missing BasicInvocationContext class
by thomas.diesler@jboss.com
Thomas Diesler [14/Nov/07 06:35 AM]
As of jbossws-spi-1.0.0, you should be using
org.jboss.wsf.spi.invocation.InvocationContext
Mark Little [14/Nov/07 06:44 AM]
Thomas, unless this was previously documented as a non-public API then it cannot be removed in the 2.0.x codebase. You can support it elsewhere, but you need to be backwardly compatible. Thanks.
Thomas Diesler [14/Nov/07 07:01 AM]
jbossws-2.0.1 was the first release that used jbossws-spi-1.0.0.GA as a
standalone project. Before that the SPI was still work in progress and
an integral part of the jbossws core code base.
I understand your backward compatibility concerns and yes, everything in
jbossws-spi-1.0.0.GA is public and can be used (from that release
onward)
I hope we can resolve possible API/SPI incompatibility issues between
2.0.0 and 2.0.1 in an unbureaucratic way and jbossws-spi-1.0.0.GA can be
our baseline to move forward
Alternatively, we would need to create a SP branch for jbossws-spi-1.0.0 where we restore the API that the ESB is using.
For many a reason I would recommend the ESB uses the 1.0.0.GA SPI/API
Mark Little [14/Nov/07 08:26 AM]
Thomas, this isn't bureaucratic: we have a message around the platforms that there will be no backward compatibility issues within minor version numbers of the individual projects. Unless you made this change because of a bug, then this has to be changed as well.
If something was available in JBossWS 2.0.0GA it needs to continue to be available in JBossWS 2.0.1GA and CPs. We've been over this before and if the other projects can work within this approach I'd expect the same from your project.
What is the problem with adding jbossws-spi-1.0.0GA and maintaining backward compatibility with JBossWS 2.0.0GA?
View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4104564#4104564
Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4104564
18 years, 4 months
[Design of JBoss Web Services] - jUDDI tables dropped at server shutdown
by thomas.diesler@jboss.com
Kurt Stam [12/Nov/07 08:54 AM]
what do you think is causing the tables to be dropped? I have never seen this. How do I reproduce it? We have no code that does that, apart from code in the unittests that is.
Kevin Conner [12/Nov/07 11:14 AM]
This seems to be an issue with the SOA platform, there is a jUDDI configuration in juddi-service.sar. This configuration sets DropOnStop as true.
I don't think this is one for us. :)
Kevin Conner [12/Nov/07 11:17 AM]
Kurt, can you take a look at the SOA platform and advise Trevor on this? Is it safe for him just to remove this sar file?
Kurt Stam [12/Nov/07 11:30 AM]
Yes it should be removed. The juddi that comes with AS is much older then the one we use in ESB.
Mark Little [12/Nov/07 11:33 AM]
JBossWS uses jUDDI, so I assume it's coming from there. Someone needs to talk to Thomas Diesler first.
Kurt Stam [12/Nov/07 11:36 AM]
I guess we (ESB) should deploy the juddi.war by default, so we bring up the juddi WSs and then run whatever tests JBossAS has for this to make sure it all still works.
View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4104506#4104506
Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4104506
18 years, 4 months
[Design of Messaging on JBoss (Messaging/JBoss)] - Re: New wireformat and client side remoting
by timfox
"jmesnil" wrote :
| This 3rd case is meant to be covered by sendOneWay(AbstractPacket packet, PacketHandler callback).
|
That's right. But I was just making the point that in the wiki page, you mention that this mode would be used for non persistent messages which is not right :)
anonymous wrote :
| "# send a packet and set a PacketHandler? to be called back later on (e.g. send a non persistent message)
|
| * using sendOneWay(AbstractPacket? packet, PacketHandler? callbackHandler)
|
Whereas the real use case is persistent non blocking messages.
anonymous wrote :
| For now, the response is handled by the same thread which sent the request but I plan to use an ExecutorService to have other threads take care of handling the packet
|
I'm not sure that will be necessary - we want to avoid as much context switching as possible.
AFAIK Mina has two thread pools - once for the selectors and one to actually execute the requests (responses). Should be ok to just use the MINA thread surely?
View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4104456#4104456
Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4104456
18 years, 4 months