Code Review Request 7142596: RMI JPRT tests are failing

Alan Bateman Alan.Bateman at oracle.com
Wed Jul 11 08:18:44 UTC 2012


On 05/07/2012 22:22, Darryl Mocek wrote:
> Hello core-libs.  Please review this webrev to fix Bugs #7142596 and 
> 7161503.  Webrev can be found here: 
> http://cr.openjdk.java.net/~dmocek/7142596/webrev.02. This commit 
> fixes concurrency issues with the RMI tests.
>
> - Added TestLibrary.createRegistryOnUnusedPort method.  This creates 
> an RMIRegistry on an unused port. It will try up to 10 times before 
> giving up.
> - Added a TestLibrary.getRegistryPort(Registry) method to get the port 
> number of the registry.
I don't have cycles to review this in any detail but what is the reason 
for the "try up to 10 times". I thought the plan was to just use 
LocateRegistry.createRegistry(0) and so use an ephemeral port.

Also one idea is to do:

public class TestRegistry extends Registry {
     public static TestRegistry create() { ... }
     public int port() { ... }
     :
}

so that the tests can simply do:

TestRegistry reg = TestRegistry.create();

and then use reg.port() instead of TestLibrary.getRegistryPort 
everywhere. It amounts to the same thing but may be a bit neater.

-Alan.



More information about the core-libs-dev mailing list