On Thu, 10 Sep 2026 13:18:37 GMT, Matthew Donovan <[email protected]> wrote:
> This PR updates AIACertTimeout.java to improve reliability: > > - Replaces the default HttpServer executor with a single-threaded > ExecutorService. One possible cause of the intermittent timeout is that that > HttpHandler passed to createContext() (line 137) has an explicit sleep() > call. The default executor executes the handler on the dispatcher thread. The > sleep() call could make the HttpServer unresponsive. > - Replace "localhost" with "127.0.0.1" > > > --------- > - [x] I confirm that I make this contribution in accordance with the [OpenJDK > Interim AI Policy](https://openjdk.org/legal/ai). test/jdk/sun/security/x509/URICertStore/AIACertTimeout.java line 55: > 53: import java.security.KeyPairGenerator; > 54: import java.security.spec.*; > 55: import java.util.*; Nit: widcard imports test/jdk/sun/security/x509/URICertStore/AIACertTimeout.java line 64: > 62: public class AIACertTimeout { > 63: > 64: private static final boolean logging = true; Do you think this should be a test parameter at `@run`? test/jdk/sun/security/x509/URICertStore/AIACertTimeout.java line 136: > 134: > 135: public void start() throws IOException { > 136: server.bind(new InetSocketAddress("127.0.0.1", 0), 0); I personally think it's better to have `InetAddress.getLoopbackAddress()...` instead of the hardcoded address. What do you think? ------------- PR Review Comment: https://git.openjdk.org/jdk/pull/32814#discussion_r3987276965 PR Review Comment: https://git.openjdk.org/jdk/pull/32814#discussion_r3987283580 PR Review Comment: https://git.openjdk.org/jdk/pull/32814#discussion_r3987311578
