|
At the very least we should update the Scheduler.getCurrentlyExecutingJobs() Javadoc to make it clear that we are only returning jobs that are executing in this instance of the cluster.
The alternative is to make the method cluster aware (or add a new explicit method that is). In this case we could use the FIRED_TRIGGERS table to determine what is being processed in other instances, and merge the results with what is running in this instance. Like for RMI, the JobExecutionContexts returned from other instances would have no Scheduler or Job references.
|