Compatibility certification request for Apusic AAS for Jakarta EE 9.1 Full Platform

[PatrickHuang888]

• Organization Name ("Organization") and, if applicable, URL:
  
  Kingdee Apusic Cloud Computing
• Product Name, Version and download URL (if applicable):
  
  AAS V10-EE9
• Specification Name, Version and download URL:
  
  Jakarta EE Platform 9.1
• TCK Version, digital SHA-256 fingerprint and download URL:
  
  Jakarta EE Platform TCK, 9.1
  SHA256SUM: 2157f31cfc49b8e6c3747270b4cc2a71fd63e79e8066c26a645db8d36305bf6a
• Public URL of TCK Results Summary:
  
  TCK results summary
• Any Additional Specification Certification Requirements:
  
  Jakarta Contexts and Dependency Injection cdi-tck-3.0.1-dist.zip, SHA-256: f0a3bdd81ea552ddf2c2a6cd2576f0d5ca45026665cb4a5c42606a58bf1c133d

Jakarta Bean Validation beanvalidation-tck-dist-3.0.0.zip, SHA-256: c975fd229df0c40947a9f0a69b779ec92bebb3d21e05fdc65fccc1d11ef5525b

Dependency Injection jakarta.inject-tck-2.0.1-bin.zip, SHA-256: 7853d02d372838f8300f5a18cfcc23011c9eb9016cf3980bba9442e4b1f8bfc6

Debugging jakarta-debugging-tck-2.0.0.zip, SHA-256: 71999815418799837dc6f3d0dc40c3dcc4144cd90c7cdfd06aa69270483d78bc

• Java runtime used to run the implementation:
  
  Oracle JDK 1.8.0_201-b09
• Summary of the information for the certification environment, operating system, cloud, ...:
  
  Apache Derby, Linux, Ubuntu 20.04.2 LTS
• By checking this box I acknowledge that the Organization I represent accepts the terms of the EFTL.
• By checking this box I attest that all TCK requirements have been met, including any compatibility rules.

[kwsutter]

The TCK Results seem to have one error being reported...

ejb30/lite/singleton

[runcts] OUT => [javatest.batch] ********************************************************************************
[runcts] OUT => [javatest.batch] Completed running 230 tests.
[runcts] OUT => [javatest.batch] Number of Tests Passed = 219
[runcts] OUT => [javatest.batch] Number of Tests Failed = 10
[runcts] OUT => [javatest.batch] Number of Tests with Errors = 1
[runcts] OUT => [javatest.batch] ********************************************************************************

[PatrickHuang888]

Hi, kwsutter
as I understand, after running test harness would run failed and errors test automatically again, and after following running log show like this

   [runcts] OUT => [javatest.batch] ********************************************************************************
   [runcts] OUT => [javatest.batch] Completed running 11 tests.
   [runcts] OUT => [javatest.batch] Number of Tests Passed      = 11
   [runcts] OUT => [javatest.batch] Number of Tests Failed      = 0
   [runcts] OUT => [javatest.batch] Number of Tests with Errors = 0
   [runcts] OUT => [javatest.batch] ********************************************************************************

and I checked ejb30/lite/singleton test report
It shows like this:

Tests that passed	230	plain view	grouped view

[kwsutter]

Hi, kwsutter
as I understand, after running test harness would run failed and errors test automatically again, and after following running log show like this

   [runcts] OUT => [javatest.batch] ********************************************************************************
   [runcts] OUT => [javatest.batch] Completed running 11 tests.
   [runcts] OUT => [javatest.batch] Number of Tests Passed      = 11
   [runcts] OUT => [javatest.batch] Number of Tests Failed      = 0
   [runcts] OUT => [javatest.batch] Number of Tests with Errors = 0
   [runcts] OUT => [javatest.batch] ********************************************************************************

and I checked ejb30/lite/singleton test report
It shows like this:
Tests that passed 230 plain view grouped view

Okay. Now I see the re-execution of the tests. I see that you have to do this re-execution on three separate buckets to get zero failures/errors:

• ejb30/lite/packaging
• ejb30/lite/singleton
• ejb30/lite/tx

You are correct that the re-execution of tests are considered okay. Personally, I would dive a bit deeper into these failures to understand why the bucket needs to be re-run to get them successful. Could be something in your product, or it could be something in the test buckets themselves. Whatever the reason, it would be good to understand. Thanks.

[scottmarlow]

@PatrickHuang888, if you do determine the cause of test failure and find that the test needs to be improved, please consider creating a pull request with improvements via github.com/eclipse-ee4j/jakartaee-tck. Help with making test improvements is greatly appreciated!

Anyone working on improving the TCK tests might want to join the https://accounts.eclipse.org/mailing-list/jakartaee-tck-dev mailing list so that you have a place to ask questions.

[PatrickHuang888]

@kwsutter @scottmarlow
I will look into this, thanks for the advice.

[kwsutter]

I'm ready to accept this CCR. +1 Thanks for answering all of the questions, @PatrickHuang888!

[PatrickHuang888]

thanks @kwsutter
I found one error case caused by HeapSize not enough, and OOM when running, can be fixed by setting -Xmx.

[kwsutter]

thanks @kwsutter
I found one error case caused by HeapSize not enough, and OOM when running, can be fixed by setting -Xmx.

Sounds good! Although the process allows for the test buckets to be immediately re-run to show success, it's a solid practice to figure out why that second run was required. The goal should be to execute the buckets just once to show success. Thanks!