(teststabilization) RFR: 8231506: Fix some instabilities in a few networking tests

[Daniel Fuchs]

Hi Alan,

I have renamed loopbackAddress() into loopbackSocketAddress(), thanks
for the suggestion!

http://cr.openjdk.java.net/~dfuchs/webrev_8231506/webrev.01

I also made a small modification to store the local port
in a final field in UnreferencedMulticastSockets::Server constructor
(one of my tests run got a NPE calling `Server::getPort())

best regards,

-- daniel

On 26/09/2019 16:43, Alan Bateman wrote:
> 
> 
> On 26/09/2019 15:16, Daniel Fuchs wrote:
>> Hi,
>>
>> Please find below a patch for:
>>
>> https://bugs.openjdk.java.net/browse/JDK-8231506
>> 8231506: Fix some instabilities in a few networking tests
>>
>> webrev:
>> http://cr.openjdk.java.net/~dfuchs/webrev_8231506/webrev.00/
>>
> In SocketImplCombinations, the update to testNewSocket2 looks good. My 
> fault that it didn't originally use the loopback address. I assume the 
> rest of the changes to the test aren't really needed as the changes are 
> to tests that don't involve a connection (the tests are mostly checking 
> that accept fails due to a SocketImpl mismatch). No objection to the 
> changes but if you do include them then loopbackAddress should probably 
> be renamed to loopbackSocketAddress to avoid confusing readers (as it 
> returns a SocketAddress rather than the InetAddress for the loopback 
> address).
> 
> -Alan