Closing the loop, FTR https://issues.apache.org/jira/browse/HTTPCORE-729
Gary On 2022/11/14 14:31:43 Gary Gregory wrote: > I will provide a fix for the NPE. Ah, now I see you did. > > I will eventually update VFS to avoid deprecation warnings. > > This issue was more about testing HC 5.2 as a drop in replacement for > 5.1.x, which it isn't, but 5.2.1 should be. > > Gary > > On Mon, Nov 14, 2022, 09:10 Oleg Kalnichevski <ol...@apache.org> wrote: > > > On Mon, 2022-11-14 at 13:51 +0000, Gary D. Gregory wrote: > > > On 2022/11/14 13:17:53 Oleg Kalnichevski wrote: > > > > On Mon, 2022-11-14 at 13:10 +0000, Gary D. Gregory wrote: > > > > > What did you test exactly? Note that running the 5.2 async > > > > > example > > > > > from the HC repo is _not_ the test to perform. > > > > > > > > > > The failure is not specific to my environment since the PR to > > > > > update > > > > > the HC version also fails on GitHub: > > > > > https://github.com/apache/commons-vfs/pull/332 > > > > > > > > > > The failure is not apparent but comes down to the fact that the > > > > > behavior of 5.2 vs 5.1.3/5.1.4 is not compatible. I'm not sure > > > > > what > > > > > the best path forward is. > > > > > > > > > > > > > I see no evidence of any regression in HttpCore. The best path > > > > forward > > > > is to find out what commons-vfs does incorrectly. > > > > > > > > For instance, I still see no exception logging. Does commons-vfs > > > > register an exception callback with HttpAsyncServer and logs > > > > exceptions > > > > in that callback? > > > > > > Thanks for the exception callback tip Oleg! I've updated the VFS code > > > and now I see: > > > > > > [INFO] --- exec-maven-plugin:3.1.0:java (default-cli) @ commons-vfs2 > > > --- > > > Mon, 14 Nov 2022 13:41:29 GMT | Serving . on /0:0:0:0:0:0:0:0:65036 > > > java.lang.NullPointerException > > > at > > > org.apache.hc.core5.http.impl.nio.ServerHttp1IOEventHandlerFactory.cr > > > eateHandler(ServerHttp1IOEventHandlerFactory.java:77) > > > at > > > org.apache.hc.core5.reactor.SingleCoreIOReactor.processPendingChannel > > > s(SingleCoreIOReactor.java:217) > > > at > > > org.apache.hc.core5.reactor.SingleCoreIOReactor.doExecute(SingleCoreI > > > OReactor.java:137) > > > at > > > org.apache.hc.core5.reactor.AbstractSingleCoreIOReactor.execute(Abstr > > > actSingleCoreIOReactor.java:86) > > > at > > > org.apache.hc.core5.reactor.IOReactorWorker.run(IOReactorWorker.java: > > > 44) > > > at java.lang.Thread.run(Thread.java:750) > > > > > > This is so much better. Indeed, there is a missing null check. > > > > Have you, though, fixed all deprecation warnings after the upgrade? > > > > Oleg > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: dev-unsubscr...@hc.apache.org > > For additional commands, e-mail: dev-h...@hc.apache.org > > > > > --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@hc.apache.org For additional commands, e-mail: dev-h...@hc.apache.org
