Displaying #maven-dev/2017-03-05.log:

Sun Mar 5 00:01:43 2017  Michael-O:tibor_, I am off. Need some rest. See you tomorrow.
Sun Mar 5 00:01:56 2017  tibor_:Good night
Sun Mar 5 05:02:30 2017  mbenson_:Joined the channel
Sun Mar 5 08:05:39 2017  tibor_:Joined the channel
Sun Mar 5 10:30:32 2017  Michael-O:Joined the channel
Sun Mar 5 10:45:26 2017  hboutemy:Joined the channel
Sun Mar 5 10:58:42 2017  tibor_:Joined the channel
Sun Mar 5 11:02:58 2017  Michael-O:Joined the channel
Sun Mar 5 11:05:18 2017  Michael-O:hboutemy, let's completely drop library.jansi.path
Sun Mar 5 11:12:11 2017  hboutemy:I really don't understand you
Sun Mar 5 11:12:25 2017  hboutemy:having the native libs in distribution is not so hard
Sun Mar 5 11:12:54 2017  hboutemy:and in fact, should be a natural way for applications integrating JAnsi
Sun Mar 5 11:14:06 2017  hboutemy:in fact, when I see https://github.com/fusesource/jansi/issues/67, I fear some people in constrained environments, with anti-virus and so on will have issues if we don't do it
Sun Mar 5 11:14:30 2017  Michael-O:It did not say it is hard, but HawtJNI doesn't make it easy unless exploded dirs are supported.
Sun Mar 5 11:14:45 2017  Michael-O:As for #67, people still can expand the .so to a safe place.
Sun Mar 5 11:15:01 2017  Michael-O:You can't make it up for everyone. That's the issue with native libs
Sun Mar 5 11:15:30 2017  hboutemy:yes you can
Sun Mar 5 11:15:42 2017  hboutemy:it's up to the packaging to prepare things
Sun Mar 5 11:15:50 2017  hboutemy:and I proved it's not that hard
Sun Mar 5 11:16:10 2017  hboutemy:(and with HawtJNI improved, it will even be easier)
Sun Mar 5 11:16:35 2017  Michael-O:Did you try on windows and linux?
Sun Mar 5 11:16:47 2017  hboutemy:I tried on Linux and CI did on Windows
Sun Mar 5 11:16:51 2017  hboutemy:and there is no magic
Sun Mar 5 11:16:58 2017  hboutemy:it's quite natural, seriously
Sun Mar 5 11:17:21 2017  hboutemy:you find a lib/jansi-native dir in your distribution
Sun Mar 5 11:17:49 2017  Michael-O:Yes, yes. Consider that we distribute the native libs twice, though.
Sun Mar 5 11:17:53 2017  hboutemy:the only thing HawtJNI should change is to avoid us to have to calculate the exact dir containing the native lib
Sun Mar 5 11:18:06 2017  Michael-O:Correct
Sun Mar 5 11:18:10 2017  hboutemy:you mean inside the .jar and in the distribution?
Sun Mar 5 11:18:15 2017  hboutemy:I don't care
Sun Mar 5 11:18:36 2017  hboutemy:seriously, I won't be so much picky
Sun Mar 5 11:19:16 2017  hboutemy:jansi.jar will support both well integrated situation (with native libs expanded in distribution) and ad-hoc where it has to do the job itself
Sun Mar 5 11:19:58 2017  hboutemy:and for freebsd user (or any other non-immediately supported platform), it gives a natural directory to put the lib
Sun Mar 5 11:20:04 2017  hboutemy:I really see only benefits
Sun Mar 5 11:20:28 2017  hboutemy:and this anti-virus concern on Windows really makes me nervous
Sun Mar 5 11:20:43 2017  Michael-O:The antivirus is ugly and a false positive.
Sun Mar 5 11:21:03 2017  hboutemy:yes, but users are users
Sun Mar 5 11:21:12 2017  Michael-O:I haven't had this problem with our scanner at work.
Sun Mar 5 11:21:19 2017  hboutemy:you're really talking like a FeeBSD extremist :)
Sun Mar 5 11:21:37 2017  Michael-O:You code includes a bug. There is not arch on macOS
Sun Mar 5 11:21:47 2017  Michael-O:the jnilib includes 32 and 64 bit code
Sun Mar 5 11:21:51 2017  hboutemy:this is a copy of HawtJNI
Sun Mar 5 11:21:53 2017  Michael-O:Please revise the code
Sun Mar 5 11:22:02 2017  hboutemy:I won't change anything on this part
Sun Mar 5 11:22:15 2017  Michael-O:The macOS library won't load
Sun Mar 5 11:22:26 2017  hboutemy:since I hope it will disappear once HawtJNI has a better support for our scenario
Sun Mar 5 11:22:31 2017  Michael-O:take a closer look. There is only one directory for macOS
Sun Mar 5 11:23:04 2017  hboutemy:you're reasoning on something I just copy/pasted
Sun Mar 5 11:23:14 2017  hboutemy:I don't have a Mac
Sun Mar 5 11:23:31 2017  hboutemy:if there is an issue in HawtJNI, let's work with HawtJNI people
Sun Mar 5 11:23:53 2017  hboutemy:you're mixing concerns
Sun Mar 5 11:29:45 2017  Michael-O:stephenc, can you test on macOS?
Sun Mar 5 11:29:58 2017  Michael-O:I expect it to fail, i.e., it will extract the jnilib
Sun Mar 5 11:31:27 2017  tibor_:Michael-O: what is the problem with native libs?
Sun Mar 5 11:32:08 2017  tibor_:mvn -B disables the feature or?
Sun Mar 5 11:33:16 2017  Michael-O:tibor_, it disables it completely.
Sun Mar 5 11:34:04 2017  tibor_:it caused any serious issues on your system?
Sun Mar 5 11:34:12 2017  Michael-O:No, none.
Sun Mar 5 11:34:24 2017  Michael-O:hboutemy, I think we can live we that until HawtJNI is improved.
Sun Mar 5 11:58:35 2017  hboutemy:Michael-O: it won't erxtract the lib because if it extracts the lib, it will do the exact same algorithm
Sun Mar 5 11:59:08 2017  hboutemy:perhaps it won't find the lib, yes
Sun Mar 5 11:59:15 2017  Michael-O:Iam confused now, didn't you do unpack-dependencies?
Sun Mar 5 11:59:21 2017  hboutemy:yes I did
Sun Mar 5 12:00:23 2017  hboutemy:then lib/jansi-native == jansi.jar!META-INF/native
Sun Mar 5 12:14:11 2017  Michael-O:tibor_, I will run the branches on another machine, 64 bit and see if there is any difference
Sun Mar 5 12:15:04 2017  tibor_:I am working on our test code and i will check it out on my Fedora.
Sun Mar 5 12:15:18 2017  Michael-O:Good!
Sun Mar 5 12:37:27 2017  Michael-O:tibor_, I have fired up an old VM with FreeBSD 9.3-RELEASE-p44 on i386 and will see wether the tests behave different
Sun Mar 5 12:37:43 2017  tibor_:ok
Sun Mar 5 12:37:49 2017  Michael-O:on my 64 bit VM all tests pass from ForkModeIT compared to the real hardware on i386 at work
Sun Mar 5 12:53:05 2017  tibor_:What is the VM "64 bit VM" specifics?
Sun Mar 5 12:53:20 2017  tibor_:and specifics of i386?
Sun Mar 5 12:54:29 2017  tibor_:CPU only or operating system version?
Sun Mar 5 13:01:06 2017  Michael-O:Joined the channel
Sun Mar 5 13:20:06 2017  rfscholte:Joined the channel
Sun Mar 5 13:22:55 2017  Michael-O:operating system only
Sun Mar 5 13:43:26 2017  Michael-O:tibor_, I have now updated my old machine to the latest kernel. Install Utils 0.9.x
Sun Mar 5 13:43:49 2017  tibor_:ok
Sun Mar 5 13:47:18 2017  Michael-O:build started for Surefire
Sun Mar 5 13:51:23 2017  Michael-O:tibor_, you won't believe it. The very same test on amd64 fails on i386
Sun Mar 5 13:51:27 2017  Michael-O:ForkModeIT
Sun Mar 5 13:52:06 2017  tibor_:so amd64 is fine, and i386 fails?
Sun Mar 5 13:52:07 2017  Michael-O:hold on, not fully true.
Sun Mar 5 13:52:15 2017  Michael-O:[INFO] SureFire JUnitCore Runner .......................... FAILURE [ 58.028 s]
Sun Mar 5 13:52:24 2017  Michael-O:Results :
Sun Mar 5 13:52:24 2017  Michael-O:Tests in error:
Sun Mar 5 13:52:24 2017  Michael-O: separatePoolsWithSuiteAndSequentialClasses(org.apache.maven.surefire.junitcore.pc.ParallelComputerBuilderTest): (..)
Sun Mar 5 13:52:24 2017  Michael-O: parallelMethodsReuseOneOrTwoThreads(org.apache.maven.surefire.junitcore.pc.ParallelComputerBuilderTest): (..)
Sun Mar 5 13:52:24 2017  Michael-O:Tests run: 137, Failures: 0, Errors: 2, Skipped: 0
Sun Mar 5 13:52:27 2017  tibor_:performance tests
Sun Mar 5 13:52:31 2017  Michael-O:tibor_, basically yes
Sun Mar 5 13:52:33 2017  tibor_:CPU is overloaded
Sun Mar 5 13:52:52 2017  tibor_:small Thread.slee( small delay )
Sun Mar 5 13:52:56 2017  Michael-O:My big machine is a Xeon one. Do you really think that this is the issue?
Sun Mar 5 13:53:05 2017  Michael-O:java.lang.AssertionError:
Sun Mar 5 13:53:06 2017  Michael-O:Expected: between <1450L> and <1750L>
Sun Mar 5 13:53:06 2017  Michael-O: but: was <1924L>
Sun Mar 5 13:53:18 2017  tibor_:yes this happens
Sun Mar 5 13:53:21 2017  Michael-O:Yes, the delay
Sun Mar 5 13:53:36 2017  tibor_:200 mills diff
Sun Mar 5 13:53:39 2017  tibor_:try again
Sun Mar 5 13:54:22 2017  Michael-O:My VM is 1 core, 2 GiB of RAM
Sun Mar 5 13:54:34 2017  Michael-O:Skylake i5
Sun Mar 5 13:55:05 2017  tibor_:could be GC overhead
Sun Mar 5 13:56:50 2017  Michael-O:Have you tried to run the tests on a i386 VM, any OS?
Sun Mar 5 13:58:38 2017  tibor_:never
Sun Mar 5 13:59:13 2017  Michael-O:The JUnit47SuiteTest never passes due to small intervals
Sun Mar 5 13:59:14 2017  tibor_:i386 is old CPU I guess
Sun Mar 5 13:59:23 2017  tibor_:do you run master?
Sun Mar 5 13:59:26 2017  Michael-O:Yes
Sun Mar 5 13:59:30 2017  Michael-O:trying master
Sun Mar 5 13:59:44 2017  tibor_:maybe we should tune GC
Sun Mar 5 13:59:45 2017  tibor_:-server -Xmx768m -XX:MaxMetaspaceSize=864m
Sun Mar 5 14:00:49 2017  Michael-O:trying Java 7
Sun Mar 5 14:00:50 2017  tibor_:Parent POM has <argLine>${jacoco.agent}</argLine>
Sun Mar 5 14:01:04 2017  tibor_:then you can use MaxPermSize
Sun Mar 5 14:01:21 2017  tibor_:in argLine we can add Xmx
Sun Mar 5 14:01:41 2017  tibor_:not too small, but not too big
Sun Mar 5 14:02:17 2017  tibor_:maybe Jacoco agent should be switched off
Sun Mar 5 14:03:13 2017  tibor_:aha, sorry jenkins profile activates jacoco, so not issue for jacoco here
Sun Mar 5 14:03:45 2017  tibor_:-Xmx512m is enough
Sun Mar 5 14:06:04 2017  Michael-O:trying now
Sun Mar 5 14:09:43 2017  Michael-O:Increased to -Xmx768M
Sun Mar 5 14:09:54 2017  Michael-O:passed the test, continuing
Sun Mar 5 14:10:26 2017  Michael-O:rfscholte, did you read my update on the Wagon issue?
Sun Mar 5 14:11:37 2017  tibor_:-Xmx768M for entire build or the unit tests only?
Sun Mar 5 14:12:42 2017  Michael-O:MAVEN_OPTS
Sun Mar 5 14:12:46 2017  Michael-O:so in general
Sun Mar 5 14:13:12 2017  rfscholte:Michael-O: yes, that works. I'll create a branch for it
Sun Mar 5 14:15:41 2017  Michael-O:Please consider that people need to be able to enable debug logging via -D on httpclient as well.
Sun Mar 5 14:18:52 2017  Michael-O:tibor_, test just passed with the MAVEN_OPTS and Java 7
Sun Mar 5 14:18:55 2017  Michael-O:retrying Java 8 now
Sun Mar 5 14:19:16 2017  tibor_:and all IT test so fast or only one clas?
Sun Mar 5 14:20:46 2017  Michael-O:Only ForkModeIt yet
Sun Mar 5 14:22:21 2017  tibor_:@rfscholte: Is Jenkins having still the issue with Windows paths cutted as stehpenc mentioned cca 2 weeks ago? Our Maven jenkinsfile builds.
Sun Mar 5 14:22:37 2017  tibor_:cut
Sun Mar 5 14:22:47 2017  Michael-O:tibor_, passed with Java 8
Sun Mar 5 14:22:52 2017  Michael-O:I will run entire master on my VM
Sun Mar 5 14:22:55 2017  tibor_:cool
Sun Mar 5 14:23:00 2017  tibor_:ok
Sun Mar 5 14:28:39 2017  Michael-O:This really confusing. the same VM, same arch: VM passes, real box at work fails.
Sun Mar 5 14:29:19 2017  Michael-O:running all ITs on my VM now
Sun Mar 5 14:29:23 2017  Michael-O:with surefire master
Sun Mar 5 14:34:28 2017  tibor_:What VM you use VMware or VirtualBox? Could it be NFS issue or OS or virtualization?
Sun Mar 5 14:35:23 2017  tibor_:by real box you mean no VM nothing but real h/w and installation?
Sun Mar 5 14:42:37 2017  Michael-O:VirtualBox
Sun Mar 5 14:42:46 2017  Michael-O:Yes, real box is a Xeon-based HPE server
Sun Mar 5 14:43:11 2017  Michael-O:I have moved the repo off NFS to a local disk on the real box
Sun Mar 5 14:45:33 2017  Michael-O:If now all tests pass on the VM, we can safely release Surefire 2.20
Sun Mar 5 14:45:37 2017  Michael-O:and check for issues later
Sun Mar 5 14:46:50 2017  tibor_:no hurry. Anyway it was good time spent because I found what can be improved.
Sun Mar 5 14:47:28 2017  Michael-O:Are you going to commit any improvements to master?
Sun Mar 5 14:47:31 2017  Michael-O:before 2.20?
Sun Mar 5 14:47:39 2017  tibor_:a lot
Sun Mar 5 14:47:57 2017  tibor_:I can write a list to mail.
Sun Mar 5 14:49:10 2017  tibor_:i am facing sockets interesting and you find jni interesting.
Sun Mar 5 14:49:54 2017  tibor_:e.g. starting jvm takes 35 millis which is long time when forking each class without reusing jvm. So sockets might be faster.
Sun Mar 5 14:50:47 2017  Michael-O:Mail ist good.
Sun Mar 5 14:50:56 2017  tibor_:I remember one user had a lot of junit tests but each made some algorithms and did not spend too much time in test method. So the time the jvm started was longer.
Sun Mar 5 14:50:59 2017  Michael-O:Sockets are portable, but have more overhead than named pipes
Sun Mar 5 14:51:16 2017  tibor_:memory overhead or cpu?
Sun Mar 5 14:51:47 2017  Michael-O:At the end, if you have the interface with interacts with streams only, you can pick either one. I think it was memory. I read several articles on the net about it.
Sun Mar 5 14:53:25 2017  tibor_:you know if you strat Wildfly server it prints a lot of logs to stdout in boot time. Do you think this might be the overhead too?
Sun Mar 5 14:53:33 2017  tibor_:start
Sun Mar 5 14:54:01 2017  tibor_:could be some packets lost on 127.0.0.1?
Sun Mar 5 14:56:08 2017  Michael-O:Overhead for who/what? Does it affect your Surefire builds?
Sun Mar 5 14:56:21 2017  Michael-O:TCP guarantees packets to arrive
Sun Mar 5 14:56:31 2017  Michael-O:this is a big difference to UDP, but also an overhead of the porotocl
Sun Mar 5 14:57:11 2017  Michael-O:http://stackoverflow.com/questions/1235958/ipc-performance-named-pipe-vs-socket
Sun Mar 5 14:59:31 2017  Michael-O:Running org.apache.maven.surefire.its.ForkModeIT
Sun Mar 5 14:59:31 2017  Michael-O:Tests run: 12, Failures: 0, Errors: 0, Skipped: 0, Time elapsed: 62.785 sec
Sun Mar 5 14:59:41 2017  Michael-O:i386, 9.3-RELEASE
Sun Mar 5 15:04:01 2017  tibor_:Isn't i386, 9.3-RELEASE the real box?
Sun Mar 5 15:06:17 2017  Michael-O:the real box runs 10.3-STABLE
Sun Mar 5 15:06:22 2017  Michael-O:my VM runs 9.3-RELEASE
Sun Mar 5 15:14:46 2017  Michael-O:Joined the channel
Sun Mar 5 15:15:40 2017  Michael-O:tibor_, I have also another real box, very old: P4 2,4 GHz, 2 GiB RAM, running FreeBS 11.0-STABLE, i386
Sun Mar 5 15:15:43 2017  Michael-O:I will try here also
Sun Mar 5 15:16:30 2017  tibor_:uf P4, I remember very old CPU. high temperatures :)
Sun Mar 5 15:17:09 2017  Michael-O:Very good during hard winter
Sun Mar 5 15:17:27 2017  tibor_::) that's true
Sun Mar 5 15:24:05 2017  hboutemy:rfscholte: you know that you can put a commit message with CMS? :)
Sun Mar 5 15:24:31 2017  hboutemy:and you can also commit locally on your computer to test before committing basically through svn
Sun Mar 5 15:24:58 2017  hboutemy:the only thing where CMS can't be avoided is to put generated site in production from staging
Sun Mar 5 15:26:49 2017  rfscholte:hboutemy: for the MNGSITE pages I'd prefer CMS. But it's like you should add comment to every action: save, stage, production. Just to much :|
Sun Mar 5 15:33:26 2017  Michael-O:tibor_, ForkModeIT passed on my P4 box
Sun Mar 5 15:33:31 2017  Michael-O:will run the entire set now
Sun Mar 5 15:34:04 2017  Michael-O:tibor_, my FreeBSD 9.3 VM, i386 passed all tests.
Sun Mar 5 15:34:19 2017  tibor_:thx
Sun Mar 5 15:37:18 2017  hboutemy:rfscholte: not, just do a "quick commit" with the log message and ignore other messages
Sun Mar 5 15:37:42 2017  hboutemy:you'll just have a comment on the source commit, which is in general useful and not too much :)
Sun Mar 5 15:53:30 2017  Michael-O:Joined the channel
Sun Mar 5 16:04:50 2017  Michael-O:Joined the channel
Sun Mar 5 16:06:13 2017  tibor_:rfscholte: maven-jar-plugin references to Archiver, but archiver Version is 3.1.1
Sun Mar 5 16:06:34 2017  tibor_:it is different approach to create Class-Path in manifest
Sun Mar 5 16:07:00 2017  tibor_:classpathMavenRepositoryLayout does not exist
Sun Mar 5 16:11:39 2017  rfscholte:tibor_: I'm missing the context...
Sun Mar 5 16:12:27 2017  tibor_:I use V 3.0.2 and it is harder to create class-path in comparison with Version 2.6
Sun Mar 5 16:13:08 2017  tibor_:for instance I cannot get path to dependency artifact from local repo in classpath in manifest
Sun Mar 5 16:14:05 2017  tibor_:The configuration is different and previous version 2.6 has a bug (a space in path in new line).
Sun Mar 5 16:14:32 2017  tibor_:path broken to two lines and second line starts with space
Sun Mar 5 16:18:25 2017  rfscholte:classpath should simply use the file reference of the dependency (of from MavenProject if its part of the reactor)
Sun Mar 5 16:18:38 2017  rfscholte:I'm not aware of any differences
Sun Mar 5 16:37:52 2017  rfscholte:tibor_: http://maven.apache.org/shared/maven-archiver/examples/classpath.html#Repository
Sun Mar 5 16:38:04 2017  rfscholte:NOTE: Deprecated in version 2.4. Use 'classpathLayoutType' instead.
Sun Mar 5 16:51:06 2017  tibor_:@rfscholte: I've got it working. My bad. Sorry.
Sun Mar 5 17:14:22 2017  Michael-O:Joined the channel
Sun Mar 5 17:16:07 2017  Michael-O:hboutemy, will you create the issues with HawtJNI?
Sun Mar 5 17:24:35 2017  tibor_:Michael-O: The Utils pass three elements in String[] of cmd to new ProcessBuilder(cmd). Maybe this is the difference between Utils and our simple test.
Sun Mar 5 17:24:58 2017  tibor_:the last contains java -jar etc
Sun Mar 5 17:25:03 2017  tibor_:with spaces
Sun Mar 5 17:41:24 2017  Michael-O:do you think we could use utils in our simple testcase?
Sun Mar 5 17:42:03 2017  Michael-O:we defintively need to get rid off cd
Sun Mar 5 17:45:22 2017  Michael-O:tets just finished on a 64 bit 10.3 vm. One ForkModeIT test failed
Sun Mar 5 17:45:54 2017  Michael-O:P4 box with 11.0 is still running
Sun Mar 5 18:00:24 2017  Michael-O:forget about the failure, it was a read timeout to Central
Sun Mar 5 18:04:46 2017  Michael-O:I will run all ITs on the problematic box one more time, but this time with -Xmx1G
Sun Mar 5 18:30:27 2017  Michael-O:Joined the channel
Sun Mar 5 18:35:26 2017  tibor_:Michael-O: I have reproduced the issue. Even on Windows.
Sun Mar 5 18:47:36 2017  Michael-O:Joined the channel
Sun Mar 5 18:49:21 2017  Michael-O:tibor_, tests completed on my old P4 box
Sun Mar 5 18:49:27 2017  Michael-O:Tests run: 712, Failures: 4, Errors: 9, Skipped: 136
Sun Mar 5 18:50:58 2017  tibor_:I have also some finding in the test code.
Sun Mar 5 18:51:42 2017  Michael-O:Tell me!
Sun Mar 5 18:54:11 2017  Michael-O:tibor_, I have uploaded a new tarball for the P4 machine
Sun Mar 5 19:00:06 2017  Michael-O:reuploading the tarball, the log file was wrong
Sun Mar 5 19:05:15 2017  Michael-O:All tests, but this one "testForkCountTwoReuse(org.apache.maven.surefire.its.ForkModeMultiModuleIT): Exit code was non-zero: 1; command line and log = (..)" fail due to network issues.
Sun Mar 5 19:05:19 2017  Michael-O:So we have only one real issue
Sun Mar 5 19:05:46 2017  Michael-O:I am rerunning ForkMode*IT.java
Sun Mar 5 19:06:06 2017  tibor_:I have pushed to our test in my github repo
Sun Mar 5 19:06:14 2017  tibor_:it uses two parallel forks
Sun Mar 5 19:06:20 2017  tibor_:and maven-shared-utils
Sun Mar 5 19:06:21 2017  Michael-O:I personally get the feeling that something is flawed with the installation on the real box at work
Sun Mar 5 19:06:26 2017  Michael-O:OK, let me check
Sun Mar 5 19:06:46 2017  tibor_:The Utils used command with spaces.
Sun Mar 5 19:07:45 2017  tibor_:Which is what Christian was pointing to. So I have commented code which splits correctly. We can try both.
Sun Mar 5 19:08:35 2017  tibor_:The application should run cca 3 seconds.
Sun Mar 5 19:09:00 2017  tibor_:java -jar exec-1.0-main.jar
Sun Mar 5 19:15:04 2017  Michael-O:from target?
Sun Mar 5 19:21:57 2017  tibor_:yes from target
Sun Mar 5 19:23:29 2017  tibor_:ForkModeMultiModuleIT_testForkCountTwoReuse does not lokk good
Sun Mar 5 19:34:09 2017  tibor_:ForkModeMultiModuleIT_testForkCountTwoReuse runs the reactor with -T2. The JVM exited with 0 which looks like the issue with missing ACK. So that plugin should acknowledge the good-bye event arrived from fork.
Sun Mar 5 19:35:35 2017  tibor_:This is one of the improvements in my list.
Sun Mar 5 19:44:28 2017  Michael-O:Just ran from target, the forks don't start, likely due to the missing JARs in testForkCountTwoReuse
Sun Mar 5 19:44:38 2017  Michael-O:I need to rerun master and change Main.java, right?
Sun Mar 5 19:46:22 2017  tibor_:you don't need to change Main, only git pull and mvn package
Sun Mar 5 19:46:32 2017  tibor_:what is the content of target?
Sun Mar 5 19:47:04 2017  tibor_:exec-1.0-main.jar, exec-1.0-fork1.jar, exec-1.0-fork2.jar
Sun Mar 5 19:48:31 2017  Michael-O:that's the content
Sun Mar 5 19:48:53 2017  tibor_:and what is the problem?
Sun Mar 5 19:49:00 2017  tibor_:any exception?
Sun Mar 5 19:49:18 2017  Michael-O:I see only ProcessBuilder.start()
Sun Mar 5 19:49:20 2017  Michael-O:that's it
Sun Mar 5 19:49:29 2017  Michael-O:startProcess() never returns obviously
Sun Mar 5 19:49:37 2017  tibor_:wait
Sun Mar 5 19:49:53 2017  tibor_:1,488,743,283,387 millis Before ProcessBuilder.start().
Sun Mar 5 19:49:54 2017  tibor_:1,488,743,283,388 millis Before ProcessBuilder.start().
Sun Mar 5 19:50:02 2017  tibor_:this is what you see?
Sun Mar 5 19:50:43 2017  tibor_:and then the command is [cmd]
Sun Mar 5 19:50:46 2017  tibor_:and then
Sun Mar 5 19:50:47 2017  tibor_:1,488,743,283,440 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 19:51:59 2017  Michael-O:Almost, there is not start() method has returnd
Sun Mar 5 19:52:05 2017  Michael-O:Only three output lines
Sun Mar 5 19:53:27 2017  tibor_:exactly
Sun Mar 5 19:53:29 2017  tibor_:and
Sun Mar 5 19:53:43 2017  tibor_:I don't say it's problem of the platfrom
Sun Mar 5 19:53:53 2017  tibor_:I see problem with Utils
Sun Mar 5 19:54:01 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 19:54:01 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 19:54:01 2017  Michael-O:1.488.747.333.689 millis Before ProcessBuilder.start().
Sun Mar 5 19:54:01 2017  Michael-O:1.488.747.333.709 millis Before ProcessBuilder.start().
Sun Mar 5 19:54:18 2017  tibor_:did you modify something?
Sun Mar 5 19:54:33 2017  Michael-O:replaced cmd with /bin/sh
Sun Mar 5 19:54:45 2017  tibor_:ah, my ba, sorry
Sun Mar 5 19:54:53 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ git diff
Sun Mar 5 19:54:53 2017  Michael-O:diff --git a/pom.xml b/pom.xml
Sun Mar 5 19:54:53 2017  Michael-O:index 2c697b2..1a2a800 100644
Sun Mar 5 19:54:53 2017  Michael-O:--- a/pom.xml
Sun Mar 5 19:54:53 2017  Michael-O:+++ b/pom.xml
Sun Mar 5 19:54:54 2017  Michael-O:@@ -17,7 +17,7 @@
Sun Mar 5 19:54:56 2017  Michael-O: <dependency>
Sun Mar 5 19:54:58 2017  Michael-O: <groupId>org.apache.maven.shared</groupId>
Sun Mar 5 19:55:00 2017  Michael-O: <artifactId>maven-shared-utils</artifactId>
Sun Mar 5 19:55:02 2017  Michael-O:- <version>0.9-SNAPSHOT</version>
Sun Mar 5 19:55:04 2017  Michael-O:+ <version>0.9.1-SNAPSHOT</version>
Sun Mar 5 19:55:06 2017  Michael-O: <exclusions>
Sun Mar 5 19:55:08 2017  Michael-O: <exclusion>
Sun Mar 5 19:55:12 2017  Michael-O: <groupId>com.google.code.findbugs</groupId>
Sun Mar 5 19:55:14 2017  Michael-O:@@ -94,4 +94,4 @@
Sun Mar 5 19:55:16 2017  Michael-O: </plugins>
Sun Mar 5 19:55:17 2017  tibor_:what is 0.9.1?
Sun Mar 5 19:55:18 2017  Michael-O: </build>
Sun Mar 5 19:55:20 2017  Michael-O:-</project>
Sun Mar 5 19:55:22 2017  Michael-O:\ No newline at end of file
Sun Mar 5 19:55:24 2017  Michael-O:+</project>
Sun Mar 5 19:55:26 2017  Michael-O:diff --git a/src/main/java/Main.java b/src/main/java/Main.java
Sun Mar 5 19:55:28 2017  Michael-O:index 76a0d71..ae246db 100644
Sun Mar 5 19:55:30 2017  Michael-O:--- a/src/main/java/Main.java
Sun Mar 5 19:55:32 2017  Michael-O:+++ b/src/main/java/Main.java
Sun Mar 5 19:55:34 2017  Michael-O:@@ -71,13 +71,8 @@ redirectTo(process.getOutputStream(), process.getInputStream());*/
Sun Mar 5 19:55:36 2017  Michael-O: String jar = new File(userDir, "exec-1.0-fork" + idx + ".jar").getCanonicalPath();
Sun Mar 5 19:55:38 2017  Michael-O: String[] cmd =
Sun Mar 5 19:55:42 2017  Michael-O: {
Sun Mar 5 19:55:44 2017  Michael-O:- /*
Sun Mar 5 19:55:46 2017  Michael-O: "/bin/sh",
Sun Mar 5 19:55:48 2017  Michael-O: "-c",
Sun Mar 5 19:55:50 2017  Michael-O:- */
Sun Mar 5 19:55:52 2017  Michael-O:- "CMD",
Sun Mar 5 19:55:54 2017  Michael-O:- "/X",
Sun Mar 5 19:55:56 2017  Michael-O:- "/C",
Sun Mar 5 19:55:58 2017  Michael-O: "cd",
Sun Mar 5 19:56:00 2017  Michael-O: quoteIfHasSpace(userDir.getAbsolutePath()),
Sun Mar 5 19:56:02 2017  tibor_:I am aware of this
Sun Mar 5 19:56:06 2017  tibor_:I wil fix it
Sun Mar 5 19:57:14 2017  Michael-O:There is on 0.9-SNAPSHOT
Sun Mar 5 19:57:21 2017  Michael-O:The bugfix branch contains 0.9.1-SNAPSHOT
Sun Mar 5 19:59:14 2017  tibor_:ok, try to rever pls. and you will have /bin/sh and 0.9.1
Sun Mar 5 19:59:20 2017  tibor_:git pull
Sun Mar 5 19:59:46 2017  tibor_:Till now cmd was created upon Utils.
Sun Mar 5 20:00:00 2017  tibor_:Now I committed our own cmd array.
Sun Mar 5 20:00:19 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 20:00:19 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 20:00:19 2017  Michael-O:1.488.747.702.005 millis Before ProcessBuilder.start().
Sun Mar 5 20:00:19 2017  Michael-O:1.488.747.701.986 millis Before ProcessBuilder.start().
Sun Mar 5 20:00:19 2017  Michael-O:[/bin/sh, -c, cd, /usr/home/mosipov/Projekte/freebsd-issue/target, &&, /usr/local/openjdk7/jre/bin/java, -jar, /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar, /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/, 2017-03-04T11-14-12_766-jvmRun2, surefire3463035433514775088tmp, surefire_11427250543530051033tmp]
Sun Mar 5 20:00:20 2017  tibor_:Maybe this will change the behavior.
Sun Mar 5 20:00:22 2017  Michael-O:[/bin/sh, -c, cd, /usr/home/mosipov/Projekte/freebsd-issue/target, &&, /usr/local/openjdk7/jre/bin/java, -jar, /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar, /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/, 2017-03-04T11-14-12_766-jvmRun2, surefire3463035433514775088tmp, surefire_11427250543530051033tmp]
Sun Mar 5 20:00:27 2017  Michael-O:1.488.747.702.074 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 20:00:29 2017  Michael-O:1.488.747.702.075 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 20:00:31 2017  Michael-O:1.488.747.702.077 millis After process.waitFor()
Sun Mar 5 20:00:33 2017  Michael-O:1.488.747.702.078 millis After process.waitFor()
Sun Mar 5 20:00:35 2017  Michael-O:1.488.747.702.079 millis After sub-process has completed with error code: 0.
Sun Mar 5 20:00:38 2017  Michael-O:1.488.747.702.081 millis After sub-process has completed with error code: 0.
Sun Mar 5 20:00:41 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$
Sun Mar 5 20:00:43 2017  tibor_:ok
Sun Mar 5 20:00:58 2017  tibor_:Do you see what I changed in last commit?
Sun Mar 5 20:01:23 2017  tibor_:Can you alter the "cmd" arrays?
Sun Mar 5 20:02:42 2017  tibor_:but here the process should last for 3 seconds, and it does not.
Sun Mar 5 20:03:38 2017  tibor_:So it's better but still not good.
Sun Mar 5 20:03:57 2017  tibor_:waitFor() should take 3 sec.
Sun Mar 5 20:05:03 2017  Michael-O:Right, the sleep does not work.
Sun Mar 5 20:05:21 2017  tibor_:InterruptedException?
Sun Mar 5 20:07:35 2017  tibor_:I have pushed more logs in Fork.java see <number>.log, and delete the old ones.
Sun Mar 5 20:09:23 2017  Michael-O:no change: mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 20:09:24 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 20:09:24 2017  Michael-O:1.488.748.197.800 millis Before ProcessBuilder.start().
Sun Mar 5 20:09:24 2017  Michael-O:1.488.748.197.819 millis Before ProcessBuilder.start().
Sun Mar 5 20:09:24 2017  Michael-O:[/bin/sh, -c, cd, /usr/home/mosipov/Projekte/freebsd-issue/target, &&, /usr/local/openjdk7/jre/bin/java, -jar, /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar, /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/, 2017-03-04T11-14-12_766-jvmRun2, surefire3463035433514775088tmp, surefire_11427250543530051033tmp]
Sun Mar 5 20:09:27 2017  Michael-O:[/bin/sh, -c, cd, /usr/home/mosipov/Projekte/freebsd-issue/target, &&, /usr/local/openjdk7/jre/bin/java, -jar, /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar, /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/, 2017-03-04T11-14-12_766-jvmRun2, surefire3463035433514775088tmp, surefire_11427250543530051033tmp]
Sun Mar 5 20:09:32 2017  Michael-O:1.488.748.197.888 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 20:09:34 2017  Michael-O:1.488.748.197.888 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 20:09:36 2017  Michael-O:1.488.748.197.891 millis After process.waitFor()
Sun Mar 5 20:09:38 2017  Michael-O:1.488.748.197.892 millis After process.waitFor()
Sun Mar 5 20:09:42 2017  Michael-O:1.488.748.197.893 millis After sub-process has completed with error code: 0.
Sun Mar 5 20:09:44 2017  Michael-O:1.488.748.197.896 millis After sub-process has completed with error code: 0.
Sun Mar 5 20:09:46 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ tree
Sun Mar 5 20:10:49 2017  tibor_:yes and when you open the log file
Sun Mar 5 20:11:14 2017  tibor_:we should see main() started
Sun Mar 5 20:11:16 2017  tibor_: main() returned
Sun Mar 5 20:11:44 2017  tibor_:if it exists the jvm was alive
Sun Mar 5 20:13:48 2017  tibor_:The number of jvms should not matter, but we can check it out with one but i dont know if it is important to test multiple forks.
Sun Mar 5 20:17:12 2017  tibor_:These last logs from console dont hang however they do not wait 3 seconds.
Sun Mar 5 20:17:16 2017  Michael-O:there is no log file
Sun Mar 5 20:17:22 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ tree
Sun Mar 5 20:17:23 2017  Michael-O:.
Sun Mar 5 20:17:23 2017  Michael-O:├── classes
Sun Mar 5 20:17:23 2017  Michael-O:│   ├── Exec.class
Sun Mar 5 20:17:23 2017  Michael-O:│   ├── Fork.class
Sun Mar 5 20:17:23 2017  Michael-O:│   ├── Main.class
Sun Mar 5 20:17:26 2017  Michael-O:│   ├── Main$1.class
Sun Mar 5 20:17:28 2017  Michael-O:│   ├── Main$2.class
Sun Mar 5 20:17:30 2017  Michael-O:│   └── Main$3.class
Sun Mar 5 20:17:31 2017  Michael-O:├── exec-1.0-fork1.jar
Sun Mar 5 20:17:33 2017  Michael-O:├── exec-1.0-fork2.jar
Sun Mar 5 20:17:36 2017  Michael-O:├── exec-1.0-main.jar
Sun Mar 5 20:17:36 2017  tibor_:so it means the jvm did not start
Sun Mar 5 20:17:38 2017  Michael-O:├── exec-1.0.jar
Sun Mar 5 20:17:42 2017  Michael-O:├── generated-sources
Sun Mar 5 20:17:43 2017  Michael-O:│   └── annotations
Sun Mar 5 20:17:46 2017  Michael-O:├── maven-archiver
Sun Mar 5 20:17:48 2017  Michael-O:│   └── pom.properties
Sun Mar 5 20:17:50 2017  Michael-O:└── maven-status
Sun Mar 5 20:17:51 2017  Michael-O: └── maven-compiler-plugin
Sun Mar 5 20:17:53 2017  Michael-O: └── compile
Sun Mar 5 20:17:55 2017  Michael-O: └── default-compile
Sun Mar 5 20:17:58 2017  Michael-O: ├── createdFiles.lst
Sun Mar 5 20:18:00 2017  Michael-O: └── inputFiles.lst
Sun Mar 5 20:18:02 2017  Michael-O:8 directories, 13 files
Sun Mar 5 20:18:18 2017  Michael-O:exactly
Sun Mar 5 20:18:28 2017  tibor_:Did the log exist before my last change when we used CLI fromUtils?
Sun Mar 5 20:18:36 2017  Michael-O:No
Sun Mar 5 20:20:20 2017  tibor_:So let's try again when with next push when we use Utils again.
Sun Mar 5 20:20:28 2017  tibor_:git pull
Sun Mar 5 20:21:28 2017  Michael-O:Even worse: [mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 20:21:28 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 20:21:28 2017  Michael-O:1.488.748.978.308 millis Before ProcessBuilder.start().
Sun Mar 5 20:21:28 2017  Michael-O:1.488.748.978.327 millis Before ProcessBuilder.start().
Sun Mar 5 20:21:33 2017  tibor_:You said that this was wrog: cd &&
Sun Mar 5 20:21:40 2017  tibor_:but this ok: cd ;
Sun Mar 5 20:22:00 2017  Michael-O:I said we can replace cd && with ProcessBuilder#directory()
Sun Mar 5 20:22:08 2017  Michael-O:I did an isolated test, cd && works fine here
Sun Mar 5 20:22:09 2017  tibor_:aha
Sun Mar 5 20:22:18 2017  tibor_:lets try out
Sun Mar 5 20:22:28 2017  tibor_:now you still do not see the logs?
Sun Mar 5 20:22:48 2017  tibor_:<num>log file
Sun Mar 5 20:23:19 2017  tibor_:I started in cd target, and all is in there
Sun Mar 5 20:23:51 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ tree
Sun Mar 5 20:23:51 2017  Michael-O:.
Sun Mar 5 20:23:51 2017  Michael-O:├── classes
Sun Mar 5 20:23:51 2017  Michael-O:│   ├── Exec.class
Sun Mar 5 20:23:51 2017  Michael-O:│   ├── Fork.class
Sun Mar 5 20:23:52 2017  Michael-O:│   ├── Main.class
Sun Mar 5 20:23:54 2017  Michael-O:│   ├── Main$1.class
Sun Mar 5 20:23:56 2017  Michael-O:│   ├── Main$2.class
Sun Mar 5 20:23:58 2017  Michael-O:│   └── Main$3.class
Sun Mar 5 20:24:00 2017  Michael-O:├── exec-1.0-fork1.jar
Sun Mar 5 20:24:02 2017  Michael-O:├── exec-1.0-fork2.jar
Sun Mar 5 20:24:04 2017  Michael-O:├── exec-1.0-main.jar
Sun Mar 5 20:24:06 2017  Michael-O:├── exec-1.0.jar
Sun Mar 5 20:24:08 2017  Michael-O:├── generated-sources
Sun Mar 5 20:24:12 2017  Michael-O:│   └── annotations
Sun Mar 5 20:24:14 2017  Michael-O:├── maven-archiver
Sun Mar 5 20:24:16 2017  Michael-O:│   └── pom.properties
Sun Mar 5 20:24:18 2017  Michael-O:└── maven-status
Sun Mar 5 20:24:20 2017  Michael-O: └── maven-compiler-plugin
Sun Mar 5 20:24:22 2017  Michael-O: └── compile
Sun Mar 5 20:24:24 2017  Michael-O: └── default-compile
Sun Mar 5 20:24:26 2017  Michael-O: ├── createdFiles.lst
Sun Mar 5 20:24:28 2017  Michael-O: └── inputFiles.lst
Sun Mar 5 20:24:30 2017  Michael-O:8 directories, 13 files
Sun Mar 5 20:24:32 2017  Michael-O:JULI requires configuration most of the times to log to stdout
Sun Mar 5 20:27:05 2017  tibor_:what you mean by JULI
Sun Mar 5 20:27:18 2017  Michael-O:java.util.logging interface
Sun Mar 5 20:27:29 2017  Michael-O:just pulled: [mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 20:27:29 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 20:27:29 2017  Michael-O:1.488.749.325.127 millis Before ProcessBuilder.start().
Sun Mar 5 20:27:29 2017  Michael-O:1.488.749.325.146 millis Before ProcessBuilder.start().
Sun Mar 5 20:28:13 2017  tibor_:I have pushed using only .directory(userDir)
Sun Mar 5 20:29:25 2017  tibor_:You think the logger threw an exception?
Sun Mar 5 20:31:07 2017  tibor_:now the logger is removed
Sun Mar 5 20:31:10 2017  tibor_:git pull
Sun Mar 5 20:32:08 2017  tibor_:SecurityManager does not permite creating file IO?
Sun Mar 5 20:32:19 2017  tibor_:permit
Sun Mar 5 20:32:49 2017  Michael-O:This is not the issue, JULI does not log per default to the console until you configure it. This is not SLF4J simple
Sun Mar 5 20:32:58 2017  Michael-O:use sysouts
Sun Mar 5 20:33:39 2017  Michael-O:http://stackoverflow.com/a/36081830/696632
Sun Mar 5 20:35:17 2017  Michael-O:The first issue we have is that the Utils code does not work compared to your manual code
Sun Mar 5 20:35:26 2017  Michael-O:your manual code does start the VM
Sun Mar 5 20:35:32 2017  Michael-O:Utils silently fails
Sun Mar 5 20:35:36 2017  Michael-O:we should focus on this
Sun Mar 5 20:35:38 2017  Michael-O:first
Sun Mar 5 20:37:00 2017  tibor_:I agree. I wanted to avoid issues connected to JULI. So I removed it in HEAD commit.
Sun Mar 5 20:39:38 2017  Michael-O:Lets focus on a working code with your custom starter
Sun Mar 5 20:39:47 2017  Michael-O:When this works, we can go on to Utils
Sun Mar 5 20:39:54 2017  tibor_:Maybe we should test only with String[] cmd created by us and not the Utils. The command is printed in console.
Sun Mar 5 20:40:17 2017  tibor_::) telepathy
Sun Mar 5 20:40:35 2017  Michael-O::d
Sun Mar 5 20:41:40 2017  tibor_:our custom starter was pushed
Sun Mar 5 20:41:53 2017  Michael-O:Revert the change to Utils please in HEAD
Sun Mar 5 20:41:58 2017  tibor_:in console we will see the array elements
Sun Mar 5 20:42:41 2017  tibor_:done
Sun Mar 5 20:43:44 2017  tibor_:any plan?
Sun Mar 5 20:43:50 2017  Michael-O:checking...
Sun Mar 5 20:47:15 2017  tibor_:Do we really need to have cd ?
Sun Mar 5 20:47:25 2017  Michael-O:No, you don't.
Sun Mar 5 20:52:15 2017  Michael-O:I am adding some code. Give me a couple of minutes
Sun Mar 5 20:55:02 2017  tibor_:sure, i will be doing something in the kitchen. So PING me by then.
Sun Mar 5 20:57:02 2017  Michael-O:tibor_, look here:
Sun Mar 5 20:57:03 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 20:57:03 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 20:57:03 2017  Michael-O:1.488.751.102.232 millis Before ProcessBuilder.start().
Sun Mar 5 20:57:03 2017  Michael-O:1.488.751.102.251 millis Before ProcessBuilder.start().
Sun Mar 5 20:57:04 2017  Michael-O:This is our JAR: /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar
Sun Mar 5 20:57:05 2017  Michael-O:This is our JAR: /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar
Sun Mar 5 20:57:07 2017  Michael-O:This is our userDir: /usr/home/mosipov/Projekte/freebsd-issue/target
Sun Mar 5 20:57:11 2017  Michael-O:This is our userDir: /usr/home/mosipov/Projekte/freebsd-issue/target
Sun Mar 5 20:57:13 2017  Michael-O:[/bin/sh, -c, /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 20:57:18 2017  Michael-O:[/bin/sh, -c, /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 20:57:21 2017  Michael-O:1.488.751.102.356 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 20:57:24 2017  Michael-O:1.488.751.102.358 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 20:57:25 2017  Michael-O:1.488.751.105.781 millis After process.waitFor()
Sun Mar 5 20:57:28 2017  Michael-O:1.488.751.105.784 millis After sub-process has completed with error code: 0.
Sun Mar 5 20:57:29 2017  Michael-O:1.488.751.105.843 millis After process.waitFor()
Sun Mar 5 20:57:31 2017  Michael-O:1.488.751.105.844 millis After sub-process has completed with error code: 0.
Sun Mar 5 20:57:34 2017  Michael-O:It waits now for 3 seconds
Sun Mar 5 20:57:57 2017  Michael-O:This is a common mistake when /bin/sh -c is invoked
Sun Mar 5 20:58:08 2017  Michael-O:-c expects only one argument
Sun Mar 5 20:58:09 2017  Michael-O:@@ -82,29 +83,15 @@ redirectTo(process.getOutputStream(), process.getInputStream());*/
Sun Mar 5 20:58:10 2017  Michael-O: /*"cd",
Sun Mar 5 20:58:10 2017  Michael-O: quoteIfHasSpace(userDir.getAbsolutePath()),
Sun Mar 5 20:58:12 2017  Michael-O: "&&",*/
Sun Mar 5 20:58:14 2017  Michael-O:- quoteIfHasSpace(executable.getAbsolutePath()),
Sun Mar 5 20:58:15 2017  Michael-O:- "-jar",
Sun Mar 5 20:58:18 2017  Michael-O:- quoteIfHasSpace(jar),
Sun Mar 5 20:58:20 2017  Michael-O:- "/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/",
Sun Mar 5 20:58:22 2017  Michael-O:- "2017-03-04T11-14-12_766-jvmRun2",
Sun Mar 5 20:58:25 2017  Michael-O:- "surefire3463035433514775088tmp",
Sun Mar 5 20:58:27 2017  Michael-O:- "surefire_11427250543530051033tmp",
Sun Mar 5 20:58:29 2017  Michael-O:+ executable.getAbsolutePath() + " " +
Sun Mar 5 20:58:31 2017  Michael-O:+ "-jar" + " " +
Sun Mar 5 20:58:33 2017  Michael-O:+ jar + " " +
Sun Mar 5 20:58:35 2017  Michael-O:+ "/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/" + " " +
Sun Mar 5 20:58:38 2017  Michael-O:+ "2017-03-04T11-14-12_766-jvmRun2" + " " +
Sun Mar 5 20:58:42 2017  Michael-O:+ "surefire3463035433514775088tmp" + " " +
Sun Mar 5 20:58:44 2017  Michael-O:+ "surefire_11427250543530051033tmp"
Sun Mar 5 20:58:46 2017  Michael-O: };
Sun Mar 5 20:58:48 2017  Michael-O:- // CMD from maven-shared-utils
Sun Mar 5 20:58:50 2017  Michael-O:- Commandline cli = new Commandline();
Sun Mar 5 20:58:52 2017  Michael-O:- cli.setExecutable(executable.getCanonicalPath());//"/usr/local/openjdk7/jre/bin/java"
Sun Mar 5 20:58:54 2017  Michael-O:- //cli.setWorkingDirectory(userDir);
Sun Mar 5 20:58:56 2017  Michael-O:- cli.createArg().setValue("-jar");
Sun Mar 5 20:58:58 2017  Michael-O:- cli.createArg().setValue(jar);
Sun Mar 5 20:59:00 2017  Michael-O:- // dummy arguments
Sun Mar 5 20:59:02 2017  Michael-O:- cli.createArg().setValue("/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/");
Sun Mar 5 20:59:05 2017  Michael-O:- cli.createArg().setValue("2017-03-04T11-14-12_766-jvmRun2");
Sun Mar 5 20:59:07 2017  Michael-O:- cli.createArg().setValue("surefire3463035433514775088tmp");
Sun Mar 5 20:59:09 2017  Michael-O:- cli.createArg().setValue("surefire_11427250543530051033tmp");
Sun Mar 5 20:59:13 2017  Michael-O:- //CommandLineUtils.executeCommandLine()
Sun Mar 5 20:59:15 2017  Michael-O:- List<String> cliCmd = cli.getShell().getShellCommandLine(cli.getArguments());
Sun Mar 5 20:59:17 2017  Michael-O:- cmd = cliCmd.toArray(new String[cliCmd.size()]);
Sun Mar 5 20:59:19 2017  Michael-O: System.out.println(Arrays.toString(cmd));
Sun Mar 5 20:59:21 2017  Michael-O:from the manpage:
Sun Mar 5 20:59:23 2017  Michael-O:The -c option causes the commands to be read from the string operand
Sun Mar 5 20:59:25 2017  Michael-O: instead of from the standard input. Keep in mind that this option only
Sun Mar 5 20:59:27 2017  Michael-O: accepts a single string as its argument, hence multi-word strings must be
Sun Mar 5 20:59:29 2017  Michael-O: quoted.
Sun Mar 5 21:00:59 2017  tibor_:and what mistDoes not Utils do that? I guess they have whole string after -c surrounded by quotation marks.
Sun Mar 5 21:03:50 2017  tibor_:So let's print both commands from our code and Utils below.
Sun Mar 5 21:03:58 2017  tibor_:How is it different?
Sun Mar 5 21:04:07 2017  tibor_:I mean the arrays
Sun Mar 5 21:04:23 2017  Michael-O:The issue is that Utils assumes a regular executable is run, e.g., java, curl, git
Sun Mar 5 21:04:40 2017  Michael-O:invoking /bin/sh is special, it always expects a SINGLE argument
Sun Mar 5 21:04:50 2017  Michael-O:Utils breaks up in multiple ones
Sun Mar 5 21:06:29 2017  tibor_:so Utils does not make it with ayyar size 3?
Sun Mar 5 21:06:39 2017  tibor_:array
Sun Mar 5 21:07:12 2017  Michael-O:Yes, we need size 3
Sun Mar 5 21:07:20 2017  Michael-O:here is the manpage of ubuntu: http://manpages.ubuntu.com/manpages/xenial/en/man1/sh.1posix.html
Sun Mar 5 21:07:30 2017  tibor_:You say this? "/bin/sh", "-c", "/path/to/java -jar my.jar my args"
Sun Mar 5 21:08:11 2017  Michael-O:Exactly
Sun Mar 5 21:08:29 2017  tibor_:how come it worked on Jenkins Ubuntu
Sun Mar 5 21:08:42 2017  Michael-O:if you try "bin/sh", "-c", "sleep", "3", it should fail
Sun Mar 5 21:08:53 2017  Michael-O:maybe GNU sh is not strict
Sun Mar 5 21:08:57 2017  tibor_:now I understand
Sun Mar 5 21:09:30 2017  tibor_:Pls print the array from Utils, I wnat to know what wrong is with Utils
Sun Mar 5 21:09:52 2017  tibor_:i did not want to relase Utils, but I guess this will happen
Sun Mar 5 21:10:12 2017  tibor_:Nobody used Utils on Unix?
Sun Mar 5 21:10:44 2017  tibor_:because on Windows the array size is 3
Sun Mar 5 21:11:11 2017  Michael-O:Hold on a second
Sun Mar 5 21:11:18 2017  tibor_:I have it in your logs
Sun Mar 5 21:11:37 2017  tibor_:1488665652885 OutputStreamFlushableCommandline#15923711 CLI = [/bin/sh, -c, cd /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse && /usr/local/openjdk7/jre/bin/java -jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/surefirebooter8768130629591811092.jar /usr/home/osipovmi/Projekte/maven-surefi
Sun Mar 5 21:11:56 2017  tibor_:it is size 3
Sun Mar 5 21:12:06 2017  tibor_:but is has cd as well
Sun Mar 5 21:12:21 2017  tibor_:and this is from real ITs
Sun Mar 5 21:12:33 2017  Michael-O:I will try cd as next
Sun Mar 5 21:14:45 2017  Michael-O:bad news, as soon as I have this "Commandline cli = new Commandline();" enabled, nothing works
Sun Mar 5 21:15:06 2017  Michael-O:It sticks to:
Sun Mar 5 21:15:07 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 21:15:07 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:15:07 2017  Michael-O:1.488.752.152.457 millis Before ProcessBuilder.start().
Sun Mar 5 21:15:07 2017  Michael-O:1.488.752.152.464 millis Before ProcessBuilder.start().
Sun Mar 5 21:16:38 2017  Michael-O:I do not even get to that point to print cmd[]
Sun Mar 5 21:18:24 2017  Michael-O:I have now introduced cd &&
Sun Mar 5 21:18:25 2017  Michael-O:lets see
Sun Mar 5 21:18:48 2017  Michael-O:it perfectly works:
Sun Mar 5 21:18:48 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 21:18:48 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:18:48 2017  Michael-O:1.488.752.407.027 millis Before ProcessBuilder.start().
Sun Mar 5 21:18:48 2017  Michael-O:1.488.752.407.047 millis Before ProcessBuilder.start().
Sun Mar 5 21:18:48 2017  Michael-O:This is our JAR: /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar
Sun Mar 5 21:18:50 2017  Michael-O:This is our JAR: /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar
Sun Mar 5 21:18:53 2017  Michael-O:This is our userDir: /usr/home/mosipov/Projekte/freebsd-issue/target
Sun Mar 5 21:18:55 2017  Michael-O:This is our userDir: /usr/home/mosipov/Projekte/freebsd-issue/target
Sun Mar 5 21:18:56 2017  Michael-O:[/bin/sh, -c, cd /usr/home/mosipov/Projekte/freebsd-issue/target && /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 21:19:01 2017  Michael-O:[/bin/sh, -c, cd /usr/home/mosipov/Projekte/freebsd-issue/target && /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 21:19:07 2017  Michael-O:1.488.752.407.152 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 21:19:09 2017  Michael-O:1.488.752.407.166 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 21:19:12 2017  Michael-O:1.488.752.410.633 millis After process.waitFor()
Sun Mar 5 21:19:15 2017  Michael-O:1.488.752.410.633 millis After process.waitFor()
Sun Mar 5 21:19:16 2017  Michael-O:1.488.752.410.636 millis After sub-process has completed with error code: 0.
Sun Mar 5 21:19:19 2017  Michael-O:1.488.752.410.639 millis After sub-process has completed with error code: 0.
Sun Mar 5 21:19:21 2017  Michael-O:FreeBSD's sh implemenation is perfectly fine
Sun Mar 5 21:19:31 2017  tibor_:yes
Sun Mar 5 21:19:49 2017  tibor_:but now I lost the context what has changed
Sun Mar 5 21:20:18 2017  tibor_:this is with Utils or our cmd
Sun Mar 5 21:21:44 2017  Michael-O:with our cmd
Sun Mar 5 21:21:48 2017  Michael-O:Utils does not work at all
Sun Mar 5 21:21:56 2017  Michael-O:the VM simply exits
Sun Mar 5 21:22:00 2017  Michael-O:not further sysouts
Sun Mar 5 21:22:17 2017  tibor_:yes but how the command from Utils looks like
Sun Mar 5 21:22:31 2017  tibor_:the constructor has no reason to hang
Sun Mar 5 21:22:31 2017  Michael-O:I can't tell you, the code does not get executed at all
Sun Mar 5 21:22:39 2017  Michael-O:this is what I am trying to tell
Sun Mar 5 21:23:23 2017  tibor_:but this does not make sense in Surefire
Sun Mar 5 21:23:23 2017  Michael-O:as soon as I add ""Commandline cli = new Commandline();"", the VM does nothing. As if System.exit() is called
Sun Mar 5 21:23:47 2017  tibor_:because in Surefire constructor gets called and then we created jar file afterwards
Sun Mar 5 21:24:21 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue]$ git status
Sun Mar 5 21:24:21 2017  Michael-O:Auf Branch master
Sun Mar 5 21:24:21 2017  Michael-O:Ihr Branch ist auf dem selben Stand wie 'origin/master'.
Sun Mar 5 21:24:21 2017  Michael-O:nichts zu committen, Arbeitsverzeichnis unverändert
Sun Mar 5 21:24:31 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 21:24:31 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:24:31 2017  Michael-O:1.488.752.746.230 millis Before ProcessBuilder.start().
Sun Mar 5 21:24:31 2017  Michael-O:1.488.752.746.249 millis Before ProcessBuilder.start().
Sun Mar 5 21:24:38 2017  Michael-O:this is with Utils
Sun Mar 5 21:25:10 2017  tibor_:Does Utils have any native lib inside?
Sun Mar 5 21:25:25 2017  tibor_:I will check it
Sun Mar 5 21:27:24 2017  tibor_:btw, check it out with 0.9-SNAPSHOT or simply 0.9 release.
Sun Mar 5 21:27:40 2017  tibor_:we made some changes in a branch
Sun Mar 5 21:27:54 2017  Michael-O:OK, wait
Sun Mar 5 21:28:21 2017  Michael-O:this patch makes it run smoothly:
Sun Mar 5 21:28:22 2017  Michael-O:diff --git a/src/main/java/Main.java b/src/main/java/Main.java
Sun Mar 5 21:28:22 2017  Michael-O:index 7b86fdd..e97ced9 100644
Sun Mar 5 21:28:22 2017  Michael-O:--- a/src/main/java/Main.java
Sun Mar 5 21:28:22 2017  Michael-O:+++ b/src/main/java/Main.java
Sun Mar 5 21:28:22 2017  Michael-O:@@ -82,17 +82,17 @@ redirectTo(process.getOutputStream(), process.getInputStream());*/
Sun Mar 5 21:28:23 2017  Michael-O: /*"cd",
Sun Mar 5 21:28:25 2017  Michael-O: quoteIfHasSpace(userDir.getAbsolutePath()),
Sun Mar 5 21:28:27 2017  Michael-O: "&&",*/
Sun Mar 5 21:28:29 2017  Michael-O:- quoteIfHasSpace(executable.getAbsolutePath()),
Sun Mar 5 21:28:31 2017  Michael-O:- "-jar",
Sun Mar 5 21:28:33 2017  Michael-O:- quoteIfHasSpace(jar),
Sun Mar 5 21:28:35 2017  Michael-O:- "/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/",
Sun Mar 5 21:28:38 2017  Michael-O:- "2017-03-04T11-14-12_766-jvmRun2",
Sun Mar 5 21:28:42 2017  Michael-O:- "surefire3463035433514775088tmp",
Sun Mar 5 21:28:44 2017  Michael-O:- "surefire_11427250543530051033tmp",
Sun Mar 5 21:28:46 2017  Michael-O:+ "cd " + userDir.getAbsolutePath() + " && " + executable.getAbsolutePath() + " " +
Sun Mar 5 21:28:49 2017  Michael-O:+ "-jar" + " " +
Sun Mar 5 21:28:51 2017  Michael-O:+ jar + " " +
Sun Mar 5 21:28:53 2017  Michael-O:+ "/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/" + " " +
Sun Mar 5 21:28:56 2017  Michael-O:+ "2017-03-04T11-14-12_766-jvmRun2" + " " +
Sun Mar 5 21:28:58 2017  Michael-O:+ "surefire3463035433514775088tmp" + " " +
Sun Mar 5 21:29:00 2017  Michael-O:+ "surefire_11427250543530051033tmp"
Sun Mar 5 21:29:02 2017  Michael-O: };
Sun Mar 5 21:29:04 2017  Michael-O: // CMD from maven-shared-utils
Sun Mar 5 21:29:06 2017  Michael-O:- Commandline cli = new Commandline();
Sun Mar 5 21:29:08 2017  Michael-O:+ /*Commandline cli = new Commandline();
Sun Mar 5 21:29:12 2017  Michael-O: cli.setExecutable(executable.getCanonicalPath());//"/usr/local/openjdk7/jre/bin/java"
Sun Mar 5 21:29:14 2017  Michael-O: //cli.setWorkingDirectory(userDir);
Sun Mar 5 21:29:16 2017  Michael-O: cli.createArg().setValue("-jar");
Sun Mar 5 21:29:18 2017  Michael-O:@@ -104,12 +104,12 @@ redirectTo(process.getOutputStream(), process.getInputStream());*/
Sun Mar 5 21:29:20 2017  Michael-O: cli.createArg().setValue("surefire_11427250543530051033tmp");
Sun Mar 5 21:29:22 2017  Michael-O: //CommandLineUtils.executeCommandLine()
Sun Mar 5 21:29:24 2017  Michael-O: List<String> cliCmd = cli.getShell().getShellCommandLine(cli.getArguments());
Sun Mar 5 21:29:26 2017  Michael-O:- cmd = cliCmd.toArray(new String[cliCmd.size()]);
Sun Mar 5 21:29:28 2017  Michael-O:+ cmd = cliCmd.toArray(new String[cliCmd.size()]);*/
Sun Mar 5 21:29:30 2017  Michael-O: System.out.println(Arrays.toString(cmd));
Sun Mar 5 21:29:32 2017  Michael-O: return new ProcessBuilder(cmd)
Sun Mar 5 21:29:34 2017  Michael-O:- .directory(userDir)
Sun Mar 5 21:29:36 2017  Michael-O:+ //.directory(userDir)
Sun Mar 5 21:29:38 2017  Michael-O: .start();
Sun Mar 5 21:29:42 2017  Michael-O: }
Sun Mar 5 21:29:58 2017  Michael-O:Just tried 0.9
Sun Mar 5 21:30:02 2017  Michael-O:same result:
Sun Mar 5 21:30:23 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ java -jar exec-1.0-main.jar
Sun Mar 5 21:30:23 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:30:23 2017  Michael-O:1.488.753.085.543 millis Before ProcessBuilder.start().
Sun Mar 5 21:30:23 2017  Michael-O:1.488.753.085.562 millis Before ProcessBuilder.start().
Sun Mar 5 21:31:57 2017  Michael-O:I tried Java 7 and 8
Sun Mar 5 21:32:21 2017  Michael-O:so you code is correct if you pass 3 args and don't use Commandline
Sun Mar 5 21:32:35 2017  tibor_::)
Sun Mar 5 21:32:49 2017  tibor_:but constructor should not hang
Sun Mar 5 21:33:36 2017  tibor_:how you know that the program did not step behind the constructor?
Sun Mar 5 21:33:47 2017  tibor_:below
Sun Mar 5 21:33:58 2017  tibor_:did you println?
Sun Mar 5 21:34:03 2017  Michael-O:I added sysouts before and after
Sun Mar 5 21:34:05 2017  Michael-O:Let me try again
Sun Mar 5 21:35:19 2017  tibor_:this is still in an async Thread. So maybe some exceptino was consumed by ThreadPool
Sun Mar 5 21:35:24 2017  Michael-O:diff --git a/src/main/java/Main.java b/src/main/java/Main.java
Sun Mar 5 21:35:24 2017  Michael-O:index 7b86fdd..d602fe4 100644
Sun Mar 5 21:35:24 2017  Michael-O:--- a/src/main/java/Main.java
Sun Mar 5 21:35:24 2017  Michael-O:+++ b/src/main/java/Main.java
Sun Mar 5 21:35:24 2017  Michael-O:@@ -82,18 +82,20 @@ redirectTo(process.getOutputStream(), process.getInputStream());*/
Sun Mar 5 21:35:25 2017  Michael-O: /*"cd",
Sun Mar 5 21:35:27 2017  Michael-O: quoteIfHasSpace(userDir.getAbsolutePath()),
Sun Mar 5 21:35:28 2017  Michael-O: "&&",*/
Sun Mar 5 21:35:31 2017  Michael-O:- quoteIfHasSpace(executable.getAbsolutePath()),
Sun Mar 5 21:35:33 2017  Michael-O:- "-jar",
Sun Mar 5 21:35:35 2017  Michael-O:- quoteIfHasSpace(jar),
Sun Mar 5 21:35:37 2017  Michael-O:- "/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/",
Sun Mar 5 21:35:42 2017  Michael-O:- "2017-03-04T11-14-12_766-jvmRun2",
Sun Mar 5 21:35:44 2017  Michael-O:- "surefire3463035433514775088tmp",
Sun Mar 5 21:35:46 2017  Michael-O:- "surefire_11427250543530051033tmp",
Sun Mar 5 21:35:48 2017  Michael-O:+ "cd " + userDir.getAbsolutePath() + " && " + executable.getAbsolutePath() + " " +
Sun Mar 5 21:35:51 2017  Michael-O:+ "-jar" + " " +
Sun Mar 5 21:35:53 2017  Michael-O:+ jar + " " +
Sun Mar 5 21:35:55 2017  Michael-O:+ "/usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/" + " " +
Sun Mar 5 21:35:57 2017  Michael-O:+ "2017-03-04T11-14-12_766-jvmRun2" + " " +
Sun Mar 5 21:35:59 2017  Michael-O:+ "surefire3463035433514775088tmp" + " " +
Sun Mar 5 21:36:01 2017  Michael-O:+ "surefire_11427250543530051033tmp"
Sun Mar 5 21:36:03 2017  Michael-O: };
Sun Mar 5 21:36:05 2017  Michael-O:+ System.out.println("Before Commandline");
Sun Mar 5 21:36:07 2017  Michael-O: // CMD from maven-shared-utils
Sun Mar 5 21:36:11 2017  Michael-O: Commandline cli = new Commandline();
Sun Mar 5 21:36:13 2017  Michael-O:- cli.setExecutable(executable.getCanonicalPath());//"/usr/local/openjdk7/jre/bin/java"
Sun Mar 5 21:36:15 2017  Michael-O:+ System.out.println("After Commandline");
Sun Mar 5 21:36:17 2017  Michael-O:+ cli.setExecutable(executable.getCanonicalPath());//"/usr/local/openjdk7/jre/bin/java"
Sun Mar 5 21:36:19 2017  Michael-O: //cli.setWorkingDirectory(userDir);
Sun Mar 5 21:36:21 2017  Michael-O: cli.createArg().setValue("-jar");
Sun Mar 5 21:36:23 2017  Michael-O: cli.createArg().setValue(jar);
Sun Mar 5 21:36:25 2017  Michael-O:@@ -109,7 +111,7 @@ redirectTo(process.getOutputStream(), process.getInputStream());*/
Sun Mar 5 21:36:27 2017  Michael-O: System.out.println(Arrays.toString(cmd));
Sun Mar 5 21:36:29 2017  Michael-O: return new ProcessBuilder(cmd)
Sun Mar 5 21:36:31 2017  Michael-O:- .directory(userDir)
Sun Mar 5 21:36:33 2017  Michael-O:+ //.directory(userDir)
Sun Mar 5 21:36:35 2017  Michael-O::
Sun Mar 5 21:36:37 2017  Michael-O:this is my patch
Sun Mar 5 21:36:41 2017  Michael-O:Do you see the Before and After?
Sun Mar 5 21:36:52 2017  tibor_:I trust you, but some exception maybe
Sun Mar 5 21:36:56 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ JAVA_VERSION=1.7 java -jar exec-1.0-main.jar
Sun Mar 5 21:36:56 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:36:56 2017  Michael-O:1.488.753.399.122 millis Before ProcessBuilder.start().
Sun Mar 5 21:36:56 2017  Michael-O:1.488.753.399.142 millis Before ProcessBuilder.start().
Sun Mar 5 21:36:56 2017  Michael-O:Before Commandline
Sun Mar 5 21:36:57 2017  Michael-O:Before Commandline
Sun Mar 5 21:36:59 2017  Michael-O:this is the output
Sun Mar 5 21:37:15 2017  tibor_:yes
Sun Mar 5 21:37:29 2017  tibor_:you see the Executors class?
Sun Mar 5 21:38:02 2017  tibor_:Line 27
Sun Mar 5 21:38:11 2017  tibor_:this is Main Thread
Sun Mar 5 21:38:29 2017  Michael-O:yes
Sun Mar 5 21:38:31 2017  Michael-O:I know
Sun Mar 5 21:38:33 2017  tibor_:so if anything is thrown we should see it in console
Sun Mar 5 21:38:36 2017  Michael-O:fixed pool
Sun Mar 5 21:38:45 2017  Michael-O:but Commandline runs in a thread
Sun Mar 5 21:38:47 2017  Michael-O:separate
Sun Mar 5 21:39:03 2017  tibor_:yes, but what if exception was not printed
Sun Mar 5 21:39:26 2017  Michael-O:from the main thread?
Sun Mar 5 21:39:30 2017  tibor_:maybe some problem or bug in constructor
Sun Mar 5 21:39:37 2017  tibor_:yes main Thread
Sun Mar 5 21:39:55 2017  Michael-O:Let me add Thread#getCurrentThread()#getName()
Sun Mar 5 21:44:32 2017  Michael-O:with utils:
Sun Mar 5 21:44:33 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:44:33 2017  Michael-O:Thread: main
Sun Mar 5 21:44:33 2017  Michael-O:1.488.753.912.295 millis Before ProcessBuilder.start().
Sun Mar 5 21:44:33 2017  Michael-O:1.488.753.912.309 millis Before ProcessBuilder.start().
Sun Mar 5 21:44:33 2017  Michael-O:Thread: pool-1-thread-1
Sun Mar 5 21:44:34 2017  Michael-O:Thread: pool-1-thread-2
Sun Mar 5 21:44:36 2017  Michael-O:Before Commandline
Sun Mar 5 21:44:38 2017  Michael-O:Before Commandline
Sun Mar 5 21:44:42 2017  Michael-O:Thread: main
Sun Mar 5 21:44:44 2017  Michael-O:without utils:
Sun Mar 5 21:45:07 2017  tibor_:NoClassDefFound maube due to excluded findbug jsr305
Sun Mar 5 21:45:10 2017  Michael-O:his Java application forks JVM.
Sun Mar 5 21:45:10 2017  Michael-O:Thread: main
Sun Mar 5 21:45:10 2017  Michael-O:1.488.753.981.313 millis Before ProcessBuilder.start().
Sun Mar 5 21:45:10 2017  Michael-O:1.488.753.981.332 millis Before ProcessBuilder.start().
Sun Mar 5 21:45:11 2017  Michael-O:Thread: pool-1-thread-1
Sun Mar 5 21:45:13 2017  Michael-O:Thread: pool-1-thread-2
Sun Mar 5 21:45:15 2017  Michael-O:Thread: pool-1-thread-1
Sun Mar 5 21:45:16 2017  tibor_:or Error
Sun Mar 5 21:45:17 2017  Michael-O:Thread: pool-1-thread-2
Sun Mar 5 21:45:19 2017  Michael-O:[/bin/sh, -c, cd /usr/home/mosipov/Projekte/freebsd-issue/target && /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 21:45:24 2017  Michael-O:[/bin/sh, -c, cd /usr/home/mosipov/Projekte/freebsd-issue/target && /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 21:45:29 2017  Michael-O:1.488.753.981.437 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 21:45:31 2017  Michael-O:1.488.753.981.450 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 21:45:33 2017  Michael-O:Thread: main
Sun Mar 5 21:45:35 2017  Michael-O:1.488.753.984.862 millis After process.waitFor()
Sun Mar 5 21:45:37 2017  Michael-O:1.488.753.984.865 millis After sub-process has completed with error code: 0.
Sun Mar 5 21:45:41 2017  Michael-O:1.488.753.984.900 millis After process.waitFor()
Sun Mar 5 21:45:43 2017  Michael-O:1.488.753.984.901 millis After sub-process has completed with error code: 0.
Sun Mar 5 21:45:45 2017  Michael-O:Could be the case
Sun Mar 5 21:46:26 2017  Michael-O:let met ry
Sun Mar 5 21:47:39 2017  Michael-O:Removed the exclusion, no avail
Sun Mar 5 21:49:07 2017  Michael-O:Class-Path: file:/${maven.repo.local}/org/apache/maven/shared/maven-sh
Sun Mar 5 21:49:07 2017  Michael-O: ared-utils/0.9/maven-shared-utils-0.9.jar file:/${maven.repo.local}/c
Sun Mar 5 21:49:07 2017  Michael-O: om/google/code/findbugs/jsr305/2.0.1/jsr305-2.0.1.jar
Sun Mar 5 21:49:07 2017  Michael-O:Created-By: Apache Maven 3.3.9
Sun Mar 5 21:49:15 2017  Michael-O:not resolved
Sun Mar 5 21:49:27 2017  tibor_:ok
Sun Mar 5 21:49:49 2017  tibor_:Why this is not filled by default, maven.repo.local
Sun Mar 5 21:50:23 2017  tibor_:it is taken fromsettings.xml?
Sun Mar 5 21:50:57 2017  tibor_:settings.localRepository ?
Sun Mar 5 21:51:03 2017  Michael-O:added hardcoded value
Sun Mar 5 21:51:09 2017  tibor_:ok
Sun Mar 5 21:51:36 2017  Michael-O:it works onw
Sun Mar 5 21:51:37 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue/target]$ JAVA_VERSION=1.7 java -jar exec-1.0-main.jar
Sun Mar 5 21:51:37 2017  Michael-O:This Java application forks JVM.
Sun Mar 5 21:51:38 2017  Michael-O:Thread: main
Sun Mar 5 21:51:38 2017  Michael-O:1.488.754.370.949 millis Before ProcessBuilder.start().
Sun Mar 5 21:51:38 2017  Michael-O:1.488.754.370.968 millis Before ProcessBuilder.start().
Sun Mar 5 21:51:38 2017  Michael-O:Thread: pool-1-thread-1
Sun Mar 5 21:51:42 2017  Michael-O:Thread: pool-1-thread-2
Sun Mar 5 21:51:44 2017  Michael-O:Before Commandline
Sun Mar 5 21:51:46 2017  Michael-O:Before Commandline
Sun Mar 5 21:51:48 2017  Michael-O:After Commandline
Sun Mar 5 21:51:50 2017  Michael-O:After Commandline
Sun Mar 5 21:51:52 2017  Michael-O:Thread: pool-1-thread-1
Sun Mar 5 21:51:54 2017  Michael-O:Thread: pool-1-thread-2
Sun Mar 5 21:51:56 2017  Michael-O:[/bin/sh, -c, /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork1.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 21:52:00 2017  Michael-O:[/bin/sh, -c, /usr/local/openjdk7/jre/bin/java -jar /usr/home/mosipov/Projekte/freebsd-issue/target/exec-1.0-fork2.jar /usr/home/osipovmi/Projekte/maven-surefire/surefire-integration-tests/target/ForkModeIT_testForkCountTwoReuse/target/surefire/ 2017-03-04T11-14-12_766-jvmRun2 surefire3463035433514775088tmp surefire_11427250543530051033tmp]
Sun Mar 5 21:52:04 2017  Michael-O:1.488.754.371.101 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 21:52:06 2017  Michael-O:1.488.754.371.132 millis After ProcessBuilder.start() method has returned.
Sun Mar 5 21:52:08 2017  Michael-O:Thread: main
Sun Mar 5 21:52:12 2017  Michael-O:1.488.754.374.569 millis After process.waitFor()
Sun Mar 5 21:52:15 2017  Michael-O:1.488.754.374.572 millis After process.waitFor()
Sun Mar 5 21:52:16 2017  Michael-O:1.488.754.374.576 millis After sub-process has completed with error code: 0.
Sun Mar 5 21:52:18 2017  Michael-O:1.488.754.374.578 millis After sub-process has completed with error code: 0.
Sun Mar 5 21:52:20 2017  Michael-O:with utils 0.9
Sun Mar 5 21:52:22 2017  Michael-O:Utils does 3 args only
Sun Mar 5 21:52:24 2017  Michael-O:which is the right thing
Sun Mar 5 21:52:29 2017  tibor_:+1
Sun Mar 5 21:53:14 2017  tibor_:now we do not have Utils after constructor?
Sun Mar 5 21:53:48 2017  tibor_:ok so we should see how Utils create the cmd array
Sun Mar 5 21:54:27 2017  Michael-O:0.9.1-SNAPSHOT works too
Sun Mar 5 21:54:39 2017  Michael-O:so Utils does construct everything correctly
Sun Mar 5 21:54:39 2017  tibor_:very well
Sun Mar 5 21:54:58 2017  Michael-O:I see no issue with Utils since your minimal example does the right thing
Sun Mar 5 21:56:11 2017  tibor_:Did you uncomment the rest of Utils code?
Sun Mar 5 21:56:36 2017  Michael-O:No, I left it intact:
Sun Mar 5 21:56:36 2017  Michael-O:[mosipov@bsd1home ~/Projekte/freebsd-issue]$ git diff
Sun Mar 5 21:56:36 2017  Michael-O:diff --git a/pom.xml b/pom.xml
Sun Mar 5 21:56:36 2017  Michael-O:index 3aea9fb..800f6d2 100644
Sun Mar 5 21:56:36 2017  Michael-O:--- a/pom.xml
Sun Mar 5 21:56:36 2017  Michael-O:+++ b/pom.xml
Sun Mar 5 21:56:38 2017  Michael-O:@@ -18,12 +18,6 @@
Sun Mar 5 21:56:42 2017  Michael-O: <groupId>org.apache.maven.shared</groupId>
Sun Mar 5 21:56:44 2017  Michael-O: <artifactId>maven-shared-utils</artifactId>
Sun Mar 5 21:56:46 2017  Michael-O: <version>0.9.1-SNAPSHOT</version>
Sun Mar 5 21:56:48 2017  Michael-O:- <exclusions>
Sun Mar 5 21:56:50 2017  Michael-O:- <exclusion>
Sun Mar 5 21:56:52 2017  Michael-O:- <groupId>com.google.code.findbugs</groupId>
Sun Mar 5 21:56:54 2017  Michael-O:- <artifactId>jsr305</artifactId>
Sun Mar 5 21:56:56 2017  Michael-O:- </exclusion>
Sun Mar 5 21:56:58 2017  Michael-O:- </exclusions>
Sun Mar 5 21:57:00 2017  Michael-O: </dependency>
Sun Mar 5 21:57:02 2017  Michael-O: </dependencies>
Sun Mar 5 21:57:04 2017  Michael-O:@@ -45,7 +39,7 @@
Sun Mar 5 21:57:06 2017  Michael-O: <manifest>
Sun Mar 5 21:57:08 2017  Michael-O: <mainClass>Main</mainClass>
Sun Mar 5 21:57:12 2017  Michael-O: <addClasspath>true</addClasspath>
Sun Mar 5 21:57:14 2017  Michael-O:- <classpathPrefix>file:/${maven.repo.local}/</classpathPrefix>
Sun Mar 5 21:57:16 2017  Michael-O:+ <classpathPrefix>file:///usr/home/mosipov/.m2/repository/</classpathPrefix>
Sun Mar 5 21:57:19 2017  Michael-O: <classpathLayoutType>repository</classpathLayoutType>
Sun Mar 5 21:57:21 2017  Michael-O: </manifest>
Sun Mar 5 21:57:23 2017  Michael-O: </archive>
Sun Mar 5 21:57:25 2017  Michael-O:@@ -94,4 +88,4 @@
Sun Mar 5 21:57:27 2017  Michael-O: </plugins>
Sun Mar 5 21:57:29 2017  Michael-O: </build>
Sun Mar 5 21:57:31 2017  Michael-O:-</project>
Sun Mar 5 21:57:33 2017  Michael-O:\ No newline at end of file
Sun Mar 5 21:57:35 2017  Michael-O:+</project>
Sun Mar 5 21:57:37 2017  Michael-O:to Java code changes
Sun Mar 5 21:58:52 2017  Michael-O:Since utils work, what now?
Sun Mar 5 22:02:54 2017  tibor_:We are in the begining.
Sun Mar 5 22:03:17 2017  tibor_:The real box still has not pass the test?
Sun Mar 5 22:03:34 2017  Michael-O:Correct#
Sun Mar 5 22:04:53 2017  Michael-O:I think we should retry as soon as you have the ACK
Sun Mar 5 22:06:02 2017  tibor_:Pls push a new PR.
Sun Mar 5 22:08:09 2017  tibor_:So what you think about it?
Sun Mar 5 22:08:29 2017  tibor_:nightmare for me
Sun Mar 5 22:08:51 2017  Michael-O:Sorry, what do you mean? What is exactly a nightmare for you?
Sun Mar 5 22:09:15 2017  tibor_:This issue Surefire and Unix.
Sun Mar 5 22:09:54 2017  tibor_:What platform was used now?
Sun Mar 5 22:10:02 2017  Michael-O:FreeBSD only
Sun Mar 5 22:10:10 2017  tibor_:yeah but real box?
Sun Mar 5 22:10:10 2017  Michael-O:FreeBSD bsd1home 11.0-STABLE FreeBSD 11.0-STABLE #0 r313898: Sat Feb 18 11:36:44 CET 2017 mosipov@bsd1home:/usr/obj/usr/src/sys/BSD1HOME i386
Sun Mar 5 22:10:17 2017  Michael-O:Yes, real box
Sun Mar 5 22:10:22 2017  Michael-O:on Pentinum 4
Sun Mar 5 22:11:01 2017  tibor_:Did not you say that P4 passed tests with some network issues and on IT broken?
Sun Mar 5 22:11:22 2017  tibor_:one IT issue
Sun Mar 5 22:11:36 2017  Michael-O:Exactly
Sun Mar 5 22:11:56 2017  Michael-O:I can run the same test code on the other real box where we have real problems
Sun Mar 5 22:11:56 2017  tibor_:I guess HP 10-STABLE was real box, or?
Sun Mar 5 22:12:05 2017  Michael-O:right
Sun Mar 5 22:12:08 2017  Michael-O:10-STALBE
Sun Mar 5 22:12:24 2017  Michael-O:let me start my laptop and access the corporate network
Sun Mar 5 22:12:32 2017  Michael-O:we will see results in 10 min
Sun Mar 5 22:12:43 2017  tibor_:You still have energy :)
Sun Mar 5 22:13:27 2017  Michael-O:Yes, final test for toda
Sun Mar 5 22:13:29 2017  Michael-O:y
Sun Mar 5 22:14:39 2017  tibor_:Don't forget the absolute path to Maven Local Repo is specified in POM.
Sun Mar 5 22:15:30 2017  Michael-O:I know
Sun Mar 5 22:20:33 2017  Michael-O:test done on the real box at work
Sun Mar 5 22:20:47 2017  Michael-O:works perfectly with the 3-second-delay
Sun Mar 5 22:25:57 2017  tibor_:I was away for a while.
Sun Mar 5 22:26:13 2017  tibor_:ok, so no issue.
Sun Mar 5 22:26:50 2017  Michael-O:Right, everything is fine
Sun Mar 5 22:29:53 2017  tibor_:If we do not add <groupId>com.google.code.findbugs</groupId> <artifactId>jsr305</artifactId>
Sun Mar 5 22:30:03 2017  Michael-O:Do you want me to remove it?
Sun Mar 5 22:30:11 2017  Michael-O:exclusion
Sun Mar 5 22:30:17 2017  tibor_:ClassLoader should not throw exception, right?
Sun Mar 5 22:30:43 2017  tibor_:If annotation is not found the CL should not throw exception.
Sun Mar 5 22:31:01 2017  tibor_:I am asking because this artifact is not on classpath in real Surefire
Sun Mar 5 22:31:16 2017  Michael-O:Added exclusion, rerunning JAR
Sun Mar 5 22:31:41 2017  Michael-O:Works without JSR305
Sun Mar 5 22:31:54 2017  tibor_:I know that JLS says that CL must not throw Exception if annotation is not on classpath.
Sun Mar 5 22:32:50 2017  Michael-O:and it doesn't
Sun Mar 5 22:32:58 2017  tibor_:yeah
Sun Mar 5 22:33:01 2017  tibor_:ok
Sun Mar 5 22:33:06 2017  tibor_:never mind
Sun Mar 5 22:33:17 2017  tibor_:let's go to sleep
Sun Mar 5 22:33:39 2017  tibor_:What is your plan for tomorrow?
Sun Mar 5 22:34:18 2017  Michael-O:Work is up tomorrow, I'd likely need to see a doctor first
Sun Mar 5 22:34:26 2017  Michael-O:Expect me to be online by evening
Sun Mar 5 22:35:44 2017  tibor_:I should visit doctor as well, because of some pain in stomach
Sun Mar 5 22:36:15 2017  Michael-O:I felt aweful half the day
Sun Mar 5 22:36:37 2017  tibor_:Did you drink too much? :)
Sun Mar 5 22:37:07 2017  Michael-O:Not at all, my daugher is pretty sick. We spent the entire day at the hospital yesterday
Sun Mar 5 22:37:11 2017  Michael-O:So we all are infected here
Sun Mar 5 22:37:29 2017  tibor_:Drink soemthing, may help.
Sun Mar 5 22:37:54 2017  Michael-O:Drinking hole day
Sun Mar 5 22:37:58 2017  tibor_::)
Sun Mar 5 22:38:06 2017  Michael-O:Commuting between restroom and computer
Sun Mar 5 22:39:12 2017  tibor_:oh, direct toilette access with computer
Sun Mar 5 22:39:39 2017  tibor_:some virus
Sun Mar 5 22:40:09 2017  Michael-O:Yes, a virus
Sun Mar 5 22:40:14 2017  tibor_:I dont have kids but my colleagues have this problem as well. All families are infected.
Sun Mar 5 22:41:54 2017  Michael-O:same here
Sun Mar 5 22:43:09 2017  tibor_:I will shave my very long beard. Takes time until I go to sleep today, tomorrow:)
Sun Mar 5 22:43:39 2017  tibor_:So thank you for your work.
Sun Mar 5 22:44:10 2017  tibor_:And heal with entire faimly.
Sun Mar 5 22:44:19 2017  Michael-O:Welcome. Do you think we can push your ideas to 2.20 or shall we wait for 2.20.1?
Sun Mar 5 22:44:30 2017  tibor_:2.20
Sun Mar 5 22:44:44 2017  tibor_:First of all
Sun Mar 5 22:44:51 2017  tibor_:I have to get back to Christian
Sun Mar 5 22:44:56 2017  tibor_:because I know
Sun Mar 5 22:45:08 2017  tibor_:that he has some PR with closing IO
Sun Mar 5 22:45:28 2017  Michael-O:This should land on 2.20?
Sun Mar 5 22:45:29 2017  tibor_:and I have started populating his changes from previous commit
Sun Mar 5 22:45:58 2017  tibor_:I think I have to do this because I dont want to ignore people around me.
Sun Mar 5 22:46:27 2017  tibor_:I could not touch with him because this issue was very important.
Sun Mar 5 22:46:49 2017  tibor_:And JUnit people need my feedback
Sun Mar 5 22:47:10 2017  Michael-O:https://github.com/apache/maven-surefire/pull/144
Sun Mar 5 22:47:12 2017  tibor_:because they want to have JUNit5 support. So I am busy guy with Surefire
Sun Mar 5 22:47:12 2017  Michael-O:Right?
Sun Mar 5 22:47:19 2017  tibor_:yes 144
Sun Mar 5 22:47:34 2017  tibor_:I have a self on my PC with much more changes
Sun Mar 5 22:47:45 2017  Michael-O:If you want to have it tested, ping me
Sun Mar 5 22:47:51 2017  tibor_:but Christian mixed several topics in this PR
Sun Mar 5 22:48:20 2017  tibor_:we should separate the concerns
Sun Mar 5 22:49:27 2017  tibor_:Thx, I will let you know, but wont be so soon. And lets see how about your helth is going to be next few days. I hope without high temperatures.
Sun Mar 5 22:50:18 2017  Michael-O:There are four commits on the PR
Sun Mar 5 22:50:29 2017  Michael-O:parts are unrelated and should be split in several PRs
Sun Mar 5 22:50:53 2017  tibor_:Yes but this is really minimum in comparison with other chnages .
Sun Mar 5 22:51:46 2017  tibor_:Writers do not need flush, but FileOutputStream needs. And this is actually our conflict.
Sun Mar 5 22:52:10 2017  tibor_:He does not want to put flush on FOS and I see opportunity.
Sun Mar 5 22:52:46 2017  tibor_:The Javadoc in close of FOS does not say close makes any flush. Opposite to Writer.
Sun Mar 5 22:53:39 2017  tibor_:So we have to be very careful what to remove, especially flush. I dont want to make it worst than bad.
Sun Mar 5 22:55:34 2017  Michael-O:I see, right
Sun Mar 5 22:55:48 2017  Michael-O:Ping him again
Sun Mar 5 22:56:01 2017  tibor_:yes but not now :)
Sun Mar 5 22:56:40 2017  tibor_:Good Night.
Sun Mar 5 22:56:58 2017  tibor_:See you next days. And take care!

Comments