Re: unstable master branch

2020-03-21 Thread Ralph Goers
I guess if I had understood I could have told you that. I believe the default-jar configuration is built into the maven jar plugin. Ralph > On Mar 21, 2020, at 8:00 AM, Matt Sicker wrote: > > Great find! I did not know there were two different properties involved > there. Thanks for digging.

Re: unstable master branch

2020-03-21 Thread Matt Sicker
Great find! I did not know there were two different properties involved there. Thanks for digging. On Sat, Mar 21, 2020 at 09:38 Volkan Yazıcı wrote: > I've finally found and solved the problem! It is right there in my > post: -Dmaven.test.skip=true. This causes test-jar generation to be > skipp

Re: unstable master branch

2020-03-21 Thread Volkan Yazıcı
I've finally found and solved the problem! It is right there in my post: -Dmaven.test.skip=true. This causes test-jar generation to be skipped and hence modules with dependencies to test-jars cannot resolve them. One should instead use -DskipTests, which will skip tests, but will still generate the

Re: unstable master branch

2020-03-06 Thread Ralph Goers
The Jenkins build of master is now working on Windows. It seems ResolverUtil had a bug that I fixed on the release-2.x branch last year but forgot to migrate to master. That fix has now been applied. Ralph > On Mar 6, 2020, at 8:48 AM, Ralph Goers wrote: > > I compile master several times a d

Re: unstable master branch

2020-03-06 Thread Ralph Goers
I compile master several times a day on MacOS without problems. Jenkins shows a single test failing on Windows. I have determined that there is one test method that is not properly closing a stream and is preventing a file from being deleted but I have not yet located the code that is causing th

Re: unstable master branch

2020-03-06 Thread Volkan Yazıcı
I am pretty confident that the "master" is broken: $ /home/vy/Software/share/java/jdk/8/bin/java -version openjdk version "1.8.0_232" OpenJDK Runtime Environment (AdoptOpenJDK)(build 1.8.0_232-b09) OpenJDK 64-Bit Server VM (AdoptOpenJDK)(build 25.232-b09, mixed mode) $ /home/vy/Software/share/jav

Re: unstable master branch

2020-03-05 Thread Volkan Yazıcı
Nope, neither "install" works; fails with the same message. On Thu, Mar 5, 2020 at 4:25 PM Matt Sicker wrote: > > It could be an issue in the pom somewhere. Does "install" instead of > "package" work for you? > > On Thu, 5 Mar 2020 at 02:27, Volkan Yazıcı wrote: > > > > Even after a clean checko

Re: unstable master branch

2020-03-05 Thread Matt Sicker
It could be an issue in the pom somewhere. Does "install" instead of "package" work for you? On Thu, 5 Mar 2020 at 02:27, Volkan Yazıcı wrote: > > Even after a clean checkout: > > $ cd /tmp > $ git clone https://github.com/apache/logging-log4j2.git log4j2 > $ cd log4j2 > $ m -rf ~/.m2/repository/

Re: unstable master branch

2020-03-05 Thread Volkan Yazıcı
Even after a clean checkout: $ cd /tmp $ git clone https://github.com/apache/logging-log4j2.git log4j2 $ cd log4j2 $ m -rf ~/.m2/repository/org/apache/logging $ ./mvnw package -Dmaven.test.skip=true I still get the very same error. I see that Jenkins is still failing, though Travis has recovered

Re: unstable master branch

2020-03-04 Thread Ralph Goers
My typical development follow 1 of 2 paths. 1. I am doing a patch or new feature targeted at the 2.x release train. I do my initial development on the release-2.x branch. Once it is fully tested I cherry-pick it (when possible) to the master branch. If I can’t cherry-pick it I manually copy th

unstable master branch

2020-03-04 Thread Volkan Yazıcı
Hello, While developing it is really difficult to keep up with the master branch, it continuously gets broken. master and release-2.x diverge a lot as well, hence I am sitting on the fence for whether I should base my changes on master or release-2.x. This is my current state: $ git remote -v ori