tag:blogger.com,1999:blog-5862547309732289650.post3204326610759024541..comments2023-07-06T00:55:14.122-07:00Comments on Ade On Middleware: The new JMS flow in ServiceMix 4Adehttp://www.blogger.com/profile/00743810504276237137noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-5862547309732289650.post-46605647846758097732010-08-02T00:27:51.553-07:002010-08-02T00:27:51.553-07:00Hi Vinod! I still believe that you can get HA and ...Hi Vinod! I still believe that you can get HA and Load Balancing in your system without the need for the classic NMR JMS and JCA flows. I recommend using Camel routes for the integration, and putting JMS queues between each route to ensure that messages are either processed successfully or replayed. <br /><br />As for your XML content not allowed in prolog error, I've seen this kind of thing before: it's usually that the payload in the JBI message isn't valid XML. Can you send a reproducible test case to the servicemix users list? Alternatively, you could post a message on the forums at fusesource.com.Adehttps://www.blogger.com/profile/00743810504276237137noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-77871720061969963342010-08-01T17:21:42.782-07:002010-08-01T17:21:42.782-07:00Hi Ade,
I am a follower of your blog articles, and...Hi Ade,<br />I am a follower of your blog articles, and I have benefited many times out of your articles. Recently I read abt the overhead of JMS and JCA flow and you recommended SEDA flow. I find it tough to convince others for SEDA flow, as we loose the HA and fail over capability. Also we are finding tough time making JMS flow alone to work. It gives some "Content not allowed in prolog" error. we are using servicemix 3. Pls advisevinodhttps://www.blogger.com/profile/16214753051278464448noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-58900749101675987962010-08-01T17:21:33.462-07:002010-08-01T17:21:33.462-07:00Hi Ade,
I am a follower of your blog articles, and...Hi Ade,<br />I am a follower of your blog articles, and I have benefited many times out of your articles. Recently I read abt the overhead of JMS and JCA flow and you recommended SEDA flow. I find it tough to convince others for SEDA flow, as we loose the HA and fail over capability. Also we are finding tough time making JMS flow alone to work. It gives some "Content not allowed in prolog" error. we are using servicemix 3. Pls advisevinodhttps://www.blogger.com/profile/16214753051278464448noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-8508637292819633142010-06-03T06:07:59.575-07:002010-06-03T06:07:59.575-07:00Here ya go! Have posted the code up at the FUSESou...Here ya go! Have posted the code up at the FUSESource FUSE ESB forum http://fusesource.com/forums/thread.jspa?threadID=1933. I haven't added configuration, but the tips at http://trenaman.blogspot.com/2010/04/four-things-you-need-to-know-about-new.html should be enough. Let me know how you get on! <br /><br />Cheers,<br />Ade.Adehttps://www.blogger.com/profile/00743810504276237137noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-40329315970071930142010-06-03T05:47:31.445-07:002010-06-03T05:47:31.445-07:00Ade,
By 'hype' I simply mean ' ... h...Ade, <br /><br />By 'hype' I simply mean ' ... here is how this feature should work' vs. proof (show me the code). Servicemix 4 has a huge amount of potential. But for guys who don't have alot of hours to burn trying to deliver a solution, useful examples are everything. I<br /><br />Would you consider posting the complete code and configuration of your working example to go along with your article? That would be a tremendous help. Does your example demonstrate networked Servicemix instances (not simply in-VM clustering)? <br /><br />Thanks!nedpepperhttps://www.blogger.com/profile/10844443859862238183noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-37716590751683596802010-06-03T03:14:40.774-07:002010-06-03T03:14:40.774-07:00Hi Nedpepper,
Thanks for the feedback! However, ...Hi Nedpepper, <br /><br />Thanks for the feedback! However, please don't accuse me of hype; I hate that. In fact, I've spoken *against* JBI and JBI clustering on this blog in the past. And, if you check out the article at http://trenaman.blogspot.com/2010/04/four-things-you-need-to-know-about-new.html, you'll see that I have indeed got the clustering working; there are however some tricky and non-obvious things you need to do to make it work.Adehttps://www.blogger.com/profile/00743810504276237137noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-14841778634368344822010-06-02T06:43:38.226-07:002010-06-02T06:43:38.226-07:00Yes ... great article and a really an essential fe...Yes ... great article and a really an essential feature in Servicemix 4. However, if you look at the Servicemix or FUSE forums, it appears that no one has been able to get it working.<br /><br />Rather than continue to spew out the hype, I challenge anyone of the Servicemix 4.x commiters to post a complete, working example of this.nedpepperhttps://www.blogger.com/profile/10844443859862238183noreply@blogger.comtag:blogger.com,1999:blog-5862547309732289650.post-44256077592228705532009-03-27T06:01:00.000-07:002009-03-27T06:01:00.000-07:00Nice article Ade...It gives a good insight into th...Nice article Ade...<BR/><BR/>It gives a good insight into the reasons for the design choice and how the new implementation of JMS flow will work.<BR/><BR/>Kudos to Guillaume for a stellar solution. Very nice indeed.Ashwin Karpehttps://www.blogger.com/profile/04086079195258423880noreply@blogger.com