RFR: 8256818: SSLSocket that is never bound or connected leaks socket resources [v3]

[Xue-Lei Andrew Fan]

On Wed, 2 Dec 2020 18:42:36 GMT, Christoph Langer <clanger at openjdk.org> wrote:

>> test/jdk/sun/security/ssl/SSLSocketImpl/SSLSocketLeak.java line 37:
>> 
>>> 35:  *          will not leave leaking socket file descriptors
>>> 36:  * @library /test/lib
>>> 37:  * @run main/othervm SSLSocketLeak
>> 
>> See bellow comment, I may suggest to have it as a manual test case if you agree the test case could be impacted.
>> @run main/manual SSLSocketLeak
>
> Hm, I think it's fine as it is. Running it in othervm will make sure the test runs in its own vm (see http://openjdk.java.net/jtreg/command-help.html). So within the VM process there should not be any interference by other workload. And we check open files before testing and afterwards, and allow for some margin.
> 
> The test has been running in our test setup for several days now, so I think it should be ok. And if worst comes to worse, and we see test noise, we might change the test to manual later on.

Sounds good to me.  Thanks!

-------------

PR: https://git.openjdk.java.net/jdk/pull/1363