Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update SharedClassesCacheChecker to only delete test-specific cache #74

Merged
merged 1 commit into from
Jan 17, 2019

Conversation

Mesbah-Alam
Copy link
Contributor

@Mesbah-Alam Mesbah-Alam commented Jan 17, 2019

@Mesbah-Alam
Copy link
Contributor Author

Mesbah-Alam commented Jan 17, 2019

5x Grinder on Linux passes: https://ci.adoptopenjdk.net/view/Test_grinder/job/Grinder/859/tapResults/

Tested on Windows using internal Jenkins Grinder.

@Mesbah-Alam
Copy link
Contributor Author

Mesbah-Alam commented Jan 17, 2019

Note: This PR fixes #73 and also employs destroyAll to destroy all test specific caches created in given non-default cacheDir (instead of individually destroying test caches from non-default location by cacheName). For test caches created in default location, it still destroys each cache by cache name; so that any cache in default location, that was not created by the test, are left un-touched.

@Mesbah-Alam
Copy link
Contributor Author

Reviewer : @pshipton

@pshipton pshipton merged commit d211b6a into adoptium:master Jan 17, 2019
@pshipton
Copy link
Contributor

@pshipton
Copy link
Contributor

Now it's failing for a different reason, an assertion in the code. Is it possible to see the commands being run?

https://ci.eclipse.org/openj9/job/Test-extended.system-JDK8-win_x86-64_cmprssptrs/140/tapTestReport

STF 17:43:37.304 - +------ Step 31 - DefinedLocationGroupAccessJava: Monitor workload processes
STF 17:43:37.304 - | Wait for processes to meet expectations
STF 17:43:37.304 - |   Processes: [WL1, WL2, WL3, WL4]
STF 17:43:37.304 - |
STF 17:43:37.304 - Monitoring processes: WL1 WL2 WL3 WL4
STF 17:43:37.436 - Found dump at: C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\core.20190117.174337.2916.0001.dmp
STF 17:43:37.476 - Found dump at: C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\core.20190117.174337.7228.0001.dmp
STF 17:43:39.106 - Found dump at: C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\javacore.20190117.174337.7228.0002.txt
STF 17:43:42.090 - Found dump at: C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\Snap.20190117.174337.7228.0003.trc
STF 17:43:44.403 - Found dump at: C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\Snap.20190117.174337.2916.0003.trc
STF 17:43:44.462 - Found dump at: C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\javacore.20190117.174337.2916.0002.txt
STF 17:43:44.718 - **FAILED** Process WL3 ended with exit code (4294967295) and not the expected exit code/s (0)
STF 17:43:44.719 - Monitoring Report Summary:
STF 17:43:44.719 -   o Process WL1 has crashed unexpectedly
STF 17:43:44.720 -   o Process WL2 ended with the expected exit code (0)
STF 17:43:44.720 -   o Process WL3 has crashed unexpectedly
STF 17:43:44.720 -   o Process WL4 has crashed unexpectedly

Found this in 30.WL4.stderr

JVMSHRC777W The softmx limit for shared cache usage is smaller than the minimum feasible value 3980759048 bytes.
23:43:37.434 0xee9a00   j9shr.1012   *   ** ASSERTION FAILED ** at CompositeCache.cpp:6276: (((0 != _theca) && hasWriteMutex(currentThread) && (getTotalSize() >= softMaxBytes) && (softMaxBytes >= getUsedBytes())))

Found this in the javacore

1XECTHTYPE     Current thread history (J9VMThread:0x00000000017D9A00)
3XEHSTTYPE     23:43:37:474859273 GMT j9shr.1012 - * ** ASSERTION FAILED ** at CompositeCache.cpp:6276: (((0 != _theca) && hasWriteMutex(currentThread) && (getTotalSize() >= softMaxBytes) && (softMaxBytes >= getUsedBytes()))) 
3XEHSTTYPE     23:43:37:474851307 GMT j9prt.58 -  <j9shmem_stat exit - contents of controlFile C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\caches\javasharedresources\\C290M4F1A64_DefinedLocationGroupAccessJavaWL3_G35 is corrupted 
3XEHSTTYPE     23:43:37:474838657 GMT j9prt.56 -  >j9shmem_stat entered, name = C290M4F1A64_DefinedLocationGroupAccessJavaWL3_G35 
3XEHSTTYPE     23:43:37:474830133 GMT j9shr.2197 - * CC setCacheAreaBoundaries: The softmx bytes 0 is smaller than the already used bytes 3980759048 
3XEHSTTYPE     23:43:37:474393980 GMT j9prt.58 -  <j9shmem_stat exit - contents of controlFile C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\caches\javasharedresources\\C290M4F1A64_DefinedLocationGroupAccessJavaWL3_G35 is corrupted 
3XEHSTTYPE     23:43:37:474383497 GMT j9prt.56 -  >j9shmem_stat entered, name = C290M4F1A64_DefinedLocationGroupAccessJavaWL3_G35 
3XEHSTTYPE     23:43:37:474374591 GMT j9shr.1028 -   CC setCacheAreaBoundaries: finalSegmentStart=000000000FA99000, newCacheEnd=0000000022640000, _theca->totalBytes=314572608 
3XEHSTTYPE     23:43:37:474367741 GMT j9shr.1027 -   CC setCacheAreaBoundaries: finalSegmentStart=000000000FA981E0, origCacheEnd=0000000022640000, _theca->totalBytes=314572608 
3XEHSTTYPE     23:43:37:474357480 GMT j9shr.1026 -  >CC setCacheAreaBoundaries: Entering 
3XEHSTTYPE     23:43:37:474347666 GMT j9shr.2269 -   CC startup: initializing a new cache 

@Mesbah-Alam
Copy link
Contributor Author

Mesbah-Alam commented Jan 18, 2019

Interesting why the crash did not occur in the 5x Grinder on Linux and also in the 1x Grinder on Windows during testing.

JVMSHRC777W The softmx limit for shared cache usage is smaller than the minimum feasible value 3980759048 bytes.

  • Is the 'minimum feasible' value for softmx universal or platform specific?

The Workload that crashes, WL3 runs the following command:

C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/openjdkbinary/j2sdk-image/jre/bin/../../bin/java -Xshareclasses:name=DefinedLocationGroupAccessJavaWL3,cacheDir=C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results\caches,nonpersistent,groupAccess -Xcompressedrefs -classpath C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\jvmtest\systemtest\stf\stf.load\bin;C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\jvmtest\systemtest\stf\stf.core\bin;C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/jvmtest/systemtest/systemtest_prereqs/log4j-2.3/log4j-api-2.3.jar;C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/jvmtest/systemtest/systemtest_prereqs/log4j-2.3/log4j-core-2.3.jar;C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/jvmtest/systemtest/systemtest_prereqs/junit-4.12/junit-4.12.jar;C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/jvmtest/systemtest/systemtest_prereqs/junit-4.12/hamcrest-core-1.3.jar;C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\jvmtest\systemtest\openjdk-systemtest\openjdk.test.lang\bin;C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\jvmtest\systemtest\openjdk-systemtest\openjdk.test.util\bin;C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\jvmtest\systemtest\openjdk-systemtest\openjdk.test.math\bin net.adoptopenjdk.loadTest.LoadTest -resultsDir C:\Users\jenkins\workspace\Test-extended.system-JDK8-win_x86-64_cmprssptrs\openjdk-tests\TestConfig\test_output_15477678633841\SharedClassesAPI_0\20190117-174310-SharedClassesAPI\results -resultsPrefix 29.WL3. -dumpRequested false -reportFailureLimit 1 -abortAtFailureLimit 10 -maxTotalLogFileSpace 200M -maxSingleLogSize 1/25 -suite.mini-mix.inventoryFile C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/openjdk-tests/TestConfig/test_output_15477678633841/SharedClassesAPI_0/20190117-174310-SharedClassesAPI/results/29.WL3.inventory/openjdk.test.load/config/inventories/mix/mini-mix.xml -suite.mini-mix.inventoryExcludeFile C:/Users/jenkins/workspace/Test-extended.system-JDK8-win_x86-64_cmprssptrs/openjdk-tests/TestConfig/test_output_15477678633841/SharedClassesAPI_0/20190117-174310-SharedClassesAPI/results/29.WL3.inventory/openjdk.test.load/config/inventories/mix/mini-mix_exclude.xml -suite.mini-mix.threadCount 3 -suite.mini-mix.totalNumberTests 78 -suite.mini-mix.selection random -suite.mini-mix.seed -1 -suite.mini-mix.repeatCount 1 -suite.mini-mix.thinkingTime 0ms..0ms

@Mesbah-Alam
Copy link
Contributor Author

Also, this test does not set any softmx. It does not seem to be a test bug.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants