[
https://issues.apache.org/jira/browse/BEAM-5332?focusedWorklogId=141776&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-141776
]
ASF GitHub Bot logged work on BEAM-5332:
----------------------------------------
Author: ASF GitHub Bot
Created on: 06/Sep/18 14:23
Start Date: 06/Sep/18 14:23
Worklog Time Spent: 10m
Work Description: mxm opened a new pull request #6342: [BEAM-5332] Do not
attempt to evict cache after shutdown
URL: https://github.com/apache/beam/pull/6342
When the job shuts down, the user code classloader is cleared which removes
the
possibility to load new classes. The LoadingCache in JobBundleFactory
attempts
to load the RemovalCause class after job shutdown to evict the cache. This
results in the following exception which prevents cleanup of Docker
containers:
```
2018-09-06 15:37:07,996 ERROR
org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory
- Unable to close.
java.lang.NoClassDefFoundError:
org/apache/beam/repackaged/beam_runners_java_fn_execution/com/google/common/cache/RemovalCause
at
org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.LocalCache$Segment.clear(LocalCache.java:3290)
at
org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.LocalCache.clear(LocalCache.java:4322)
at
org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.LocalCache$LocalManualCache.invalidateAll(LocalCache.java:4937)
at
org.apache.beam.runners.fnexecution.control.JobBundleFactoryBase.close(JobBundleFactoryBase.java:186)
at
org.apache.beam.runners.flink.translation.functions.FlinkBatchExecutableStageContext.close(FlinkBatchExecutableStageContext.java:68)
at
org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory$WrappedContext.closeActual(ReferenceCountingFlinkExecutableStageContextFactory.java:186)
at
org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory$WrappedContext.access$200(ReferenceCountingFlinkExecutableStageContextFactory.java:162)
at
org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory.release(ReferenceCountingFlinkExecutableStageContextFactory.java:150)
at
org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory.lambda$scheduleRelease$1(ReferenceCountingFlinkExecutableStageContextFactory.java:110)
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
at
java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.ClassNotFoundException:
org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.RemovalCause
at java.net.URLClassLoader.findClass(URLClassLoader.java:381)
at java.lang.ClassLoader.loadClass(ClassLoader.java:424)
at
org.apache.flink.runtime.execution.librarycache.FlinkUserCodeClassLoaders$ChildFirstClassLoader.loadClass(FlinkUserCodeClassLoaders.java:129)
at java.lang.ClassLoader.loadClass(ClassLoader.java:357)
... 16 more
```
The solution for now is to attempt to cleanup the cache synchronously when
closing the JobBundleFactory.
CC @angoenka @tweise
Post-Commit Tests Status (on master branch)
------------------------------------------------------------------------------------------------
Lang | SDK | Apex | Dataflow | Flink | Gearpump | Samza | Spark
--- | --- | --- | --- | --- | --- | --- | ---
Go | [](https://builds.apache.org/job/beam_PostCommit_Go_GradleBuild/lastCompletedBuild/)
| --- | --- | --- | --- | --- | ---
Java | [](https://builds.apache.org/job/beam_PostCommit_Java_GradleBuild/lastCompletedBuild/)
| [](https://builds.apache.org/job/beam_PostCommit_Java_ValidatesRunner_Apex_Gradle/lastCompletedBuild/)
| [](https://builds.apache.org/job/beam_PostCommit_Java_ValidatesRunner_Dataflow_Gradle/lastCompletedBuild/)
| [](https://builds.apache.org/job/beam_PostCommit_Java_ValidatesRunner_Flink_Gradle/lastCompletedBuild/)
| [](https://builds.apache.org/job/beam_PostCommit_Java_ValidatesRunner_Gearpump_Gradle/lastCompletedBuild/)
| [](https://builds.apache.org/job/beam_PostCommit_Java_ValidatesRunner_Samza_Gradle/lastCompletedBuild/)
| [](https://builds.apache.org/job/beam_PostCommit_Java_ValidatesRunner_Spark_Gradle/lastCompletedBuild/)
Python | [](https://builds.apache.org/job/beam_PostCommit_Python_Verify/lastCompletedBuild/)
| --- | [](https://builds.apache.org/job/beam_PostCommit_Py_VR_Dataflow/lastCompletedBuild/)
</br> [](https://builds.apache.org/job/beam_PostCommit_Py_ValCont/lastCompletedBuild/)
| --- | --- | --- | ---
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 141776)
Time Spent: 10m
Remaining Estimate: 0h
> SDK harness containers are not eventually shut down after job ends
> ------------------------------------------------------------------
>
> Key: BEAM-5332
> URL: https://issues.apache.org/jira/browse/BEAM-5332
> Project: Beam
> Issue Type: Bug
> Components: runner-flink
> Reporter: Maximilian Michels
> Assignee: Maximilian Michels
> Priority: Major
> Fix For: 2.8.0
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When the job shuts down, the user code classloader is cleared which removes
> the possibility to load new classes. The {{LoadingCache}} attempts to load
> the {{RemovalCause}} class after job shutdown to evict the cache.
> We shouldn't attempt to execute code after the job has been removed. This is
> not safe, at least not with Flink.
> {noformat}
> 2018-09-06 15:37:07,996 ERROR
> org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory
> - Unable to close.
> java.lang.NoClassDefFoundError:
> org/apache/beam/repackaged/beam_runners_java_fn_execution/com/google/common/cache/RemovalCause
> at
> org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.LocalCache$Segment.clear(LocalCache.java:3290)
> at
> org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.LocalCache.clear(LocalCache.java:4322)
> at
> org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.LocalCache$LocalManualCache.invalidateAll(LocalCache.java:4937)
> at
> org.apache.beam.runners.fnexecution.control.JobBundleFactoryBase.close(JobBundleFactoryBase.java:186)
> at
> org.apache.beam.runners.flink.translation.functions.FlinkBatchExecutableStageContext.close(FlinkBatchExecutableStageContext.java:68)
> at
> org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory$WrappedContext.closeActual(ReferenceCountingFlinkExecutableStageContextFactory.java:186)
> at
> org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory$WrappedContext.access$200(ReferenceCountingFlinkExecutableStageContextFactory.java:162)
> at
> org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory.release(ReferenceCountingFlinkExecutableStageContextFactory.java:150)
> at
> org.apache.beam.runners.flink.translation.functions.ReferenceCountingFlinkExecutableStageContextFactory.lambda$scheduleRelease$1(ReferenceCountingFlinkExecutableStageContextFactory.java:110)
> at
> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
> at
> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: java.lang.ClassNotFoundException:
> org.apache.beam.repackaged.beam_runners_java_fn_execution.com.google.common.cache.RemovalCause
> at java.net.URLClassLoader.findClass(URLClassLoader.java:381)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:424)
> at
> org.apache.flink.runtime.execution.librarycache.FlinkUserCodeClassLoaders$ChildFirstClassLoader.loadClass(FlinkUserCodeClassLoaders.java:129)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:357)
> ... 16 more
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
