On 22/06/18 10:47, Konstantin Kolinko wrote: <snip/>
> 2. Unit tests (Java 8u172 32-bit x All connectors): OK, but two issues > 1). full run running serially was 5 hours (297 minutes), > and 1 hour of that was wasted by a single test, running for 20 minutes > * 3 connectors: > javax.websocket.TestContainerProviderPerformance I've renamed that test Tester... as it does not contain any assertions. <snip/> > 2) org.apache.tomcat.jni.TestSocketServer fails for all connectors. > [concat] Testsuites with failed tests: > [concat] TEST-org.apache.tomcat.jni.TestSocketServer.APR.txt > [concat] TEST-org.apache.tomcat.jni.TestSocketServer.NIO.txt > [concat] TEST-org.apache.tomcat.jni.TestSocketServer.NIO2.txt > > It is not a showstopper. It is JNI internals test, and failures are > apparently caused by timer accuracy. Some test have already been > corrected to be more tolerant, but apparently not all of them. I've improved the debug logging so we can see how far out the error margin is. Thanks for the feedback. Mark --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org For additional commands, e-mail: dev-h...@tomcat.apache.org
