aqa-tests: jdk_tools_1 jpackage tests in Windows JDK 17 & 21 fail due to signing issue
From examining: Failed run: https://ci.adoptium.net/job/Test_openjdk21_hs_extended.openjdk_x86-64_windows_testList_2/22/testReport/tools_jpackage_windows_WinConsoleTest/java/WinConsoleTest/ Successful run: https://ci.adoptium.net/job/Test_openjdk21_hs_extended.openjdk_x86-64_windows_testList_2/18/testReport/tools_jpackage_windows_WinConsoleTest/java/WinConsoleTest/
Almost all of the failures mention something similar to:
[01:24:25.973] TRACE: exec: Execute tool provider [jpackage --input .\test.1ac8b3a6\input --dest .\test.1ac8b3a6\output --name WinConsoleTest --type app-image --main-jar hello.jar --main-class Hello --win-console --verbose](15)...
[2023/10/19 01:24:32.336, jpackage.dll (PID: 5868, TID: 10192), jpackage.cpp:51 (Java_jdk_jpackage_internal_ExecutableRebrander_lockResource)]
ERROR: Exception with message 'ResourceEditor.cpp(37) at ResourceEditor::FileLock::FileLock(): BeginUpdateResource(C:\Jenkins\workspace\Test_openjdk21_hs_extended.openjdk_x86-64_windows_testList_2\aqa-tests\TKG\output_16976697789912\jdk_tools_1\work\scratch\0\.\test.1ac8b3a6\output\WinConsoleTest\WinConsoleTest.exe) failed. System error [1006](system error 1006 (The volume for a file has been externally altered so that the opened file is no longer valid))' caught
[...]
[01:24:32.337] java.lang.RuntimeException: Failed to lock: C:\Jenkins\workspace\Test_openjdk21_hs_extended.openjdk_x86-64_windows_testList_2\aqa-tests\TKG\output_16976697789912\jdk_tools_1\work\scratch\0\.\test.1ac8b3a6\output\WinConsoleTest\WinConsoleTest.exe
The notable output in the failure run that isn’t in the successful run is:
[01:24:32.336] Warning: Windows Defender may prevent jpackage from functioning. If there is an issue, it can be addressed by either disabling realtime monitoring, or adding an exclusion for the directory "C:\Users\jenkins\AppData\Local\Temp\jdk.jpackage7026907677576338441".
Test Info Test Name: jdk_tools_1 Test Duration: 18 min 1 sec Machine: build-azure-win2012r2-x64-2 TRSS link for the test output: http://20.90.182.165/output/test?id=65301ea720748f006f8f3b66
Build Info Build Name: Test_openjdk17_hs_extended.openjdk_x86-64_windows_testList_0 Jenkins Build start time: Oct 18 2023, 06:21 am Jenkins Build URL: https://ci.adoptium.net/job/Test_openjdk17_hs_extended.openjdk_x86-64_windows_testList_0/117/ TRSS link for the build: http://20.90.182.165/allTestsInfo?buildId=65301def20748f006f8f3776
Java Version openjdk version “17.0.9” 2023-10-17 OpenJDK Runtime Environment Temurin-17.0.9+9 (build 17.0.9+9) OpenJDK 64-Bit Server VM Temurin-17.0.9+9 (build 17.0.9+9, mixed mode, sharing)
This test has been failed 3 times since Sep 24 2023, 09:31 am Java Version when the issue first seen openjdk version “17.0.9-beta” 2023-10-17 OpenJDK Runtime Environment Temurin-17.0.9+7-202309241213 (build 17.0.9-beta+7-202309241213) OpenJDK 64-Bit Server VM Temurin-17.0.9+7-202309241213 (build 17.0.9-beta+7-202309241213, mixed mode, sharing) Jenkins Build URL: https://ci.adoptium.net/job/Test_openjdk17_hs_extended.openjdk_x86-64_windows_testList_0/115/
The test failed on machine build-azure-win2012r2-x64-2 1 times The test failed on machine test-azure-win2019-x64-1 1 times The test failed on machine test-azure-win2016-x64-1 1 times
About this issue
- Original URL
- State: closed
- Created 8 months ago
- Comments: 33 (32 by maintainers)
Re-builds in-progress for release: jdk-17: win_64 https://ci.adoptium.net/job/build-scripts/job/release-openjdk17-pipeline/43/ jdk-17: win_32 https://ci.adoptium.net/job/build-scripts/job/release-openjdk17-pipeline/46/ jdk-21: win_64 https://ci.adoptium.net/job/build-scripts/job/release-openjdk21-pipeline/14/
No, jpackageapplauncher is unique as only being present in the JMOD, the signing job signs signs the visible jdk bin exe’s the JMOD exe signing happens during the build.
jmod jdk.jpackages file: classes\jdk\jpackage\internal\resources\jpackageapplauncher.exe is 0 bytes long! Wonder if EF signing failure went un-checked… leaving 0 length file
Noting that Sophia has confirmed in an edit to her previous comment that Azul and Temurin are showing different results when the test is run on the same machine.