[
https://issues.apache.org/jira/browse/FLINK-8856?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16391298#comment-16391298
]
ASF GitHub Bot commented on FLINK-8856:
---------------------------------------
Github user aljoscha commented on a diff in the pull request:
https://github.com/apache/flink/pull/5658#discussion_r173170798
--- Diff:
flink-core/src/main/java/org/apache/flink/util/ExceptionUtils.java ---
@@ -81,12 +81,15 @@ public static String stringifyException(final Throwable
e) {
* <p>Currently considered fatal exceptions are Virtual Machine errors
indicating
* that the JVM is corrupted, like {@link InternalError}, {@link
UnknownError},
* and {@link java.util.zip.ZipError} (a special case of InternalError).
+ * The {@link ThreadDeath} exception is also treated as a fatal error,
because when
+ * a thread is forcefully stopped, there is a high chance that parts of
the system
+ * is in an inconsistent state.
--- End diff --
nit/typo: "are in an inconsistent state"?
> Move all interrupt() calls to TaskCanceler
> ------------------------------------------
>
> Key: FLINK-8856
> URL: https://issues.apache.org/jira/browse/FLINK-8856
> Project: Flink
> Issue Type: Bug
> Components: TaskManager
> Reporter: Stephan Ewen
> Assignee: Stephan Ewen
> Priority: Blocker
> Fix For: 1.5.0, 1.6.0
>
>
> We need this to work around the following JVM bug:
> https://bugs.java.com/bugdatabase/view_bug.do?bug_id=8138622
> To circumvent this problem, the {{TaskCancelerWatchDog}} must not call
> {{interrupt()}} at all, but only join on the executing thread (with timeout)
> and cause a hard exit once cancellation takes to long.
> A user affected by this problem reported this in FLINK-8834
> Personal note: The Thread.join(...) method unfortunately is not 100% reliable
> as well, because it uses {{System.currentTimeMillis()}} rather than
> {{System.nanoTime()}}. Because of that, sleeps can take overly long when the
> clock is adjusted. I wonder why the JDK authors do not follow their own
> recommendations and use {{System.nanoTime()}} for all relative time
> measures...
> EDIT: I am not the only one wondering why:
> https://stackoverflow.com/questions/42544387/why-does-thread-join-use-currenttimemillis
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
