Hi, I am using AS5.1 and Seam 2.2.0.GA. Some of the scheduled jobs in my application are started through an APPLICATION scoped seam component that has an @Observer("org.jboss.seam.postInitialization") . On this event it starts a bunch of QuartzTriggerHandle(s) with a start time and duration pluggen in. These jobs run as per its requested parameters to do various integration tasks. So far these work perfectly well in a single server mode.
My Question is what is the best was to convert it into a Cluster Wide Singleton. What I want is an HASingleton instance. But since it is a seam component, will adding @Depends("jboss.ha:service=HASingletonDeployer,type=Barrier") solve my problem or do I need to do something special? Is there a better approach specific to Seam? I know the latter part of the question is more appropriate to the Seam forum. But still, since there is a lot of overlap in support. So posting here.
AFAIR there isnt nor its planned to have cluster scope in seam, only application which is per-jvm.
I am guessing that adding the @Depends(..) should do the trick as it operates on POJOs as well. If not, you could leverage server/all/deploy-hasingleton by splitting that cluster-wide singleton part of the app to another deployment.