[
https://issues.apache.org/jira/browse/HDFS-15249?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17071456#comment-17071456
]
Toshihiro Suzuki commented on HDFS-15249:
-----------------------------------------
Currently, we already have a synchronized block for *checksInProgress* and
*completedChecks* in addResultCachingCallback(), but don't have in schedule():
https://github.com/apache/hadoop/blob/80b877a72f52ef0f4acafe15db55b8ed61fbe6d2/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/checker/ThrottledAsyncChecker.java#L170-L174
https://github.com/apache/hadoop/blob/80b877a72f52ef0f4acafe15db55b8ed61fbe6d2/hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/checker/ThrottledAsyncChecker.java#L179-L183
I thought we also needed a synchronized block in schedule() and filed this Jira.
> ThrottledAsyncChecker is not thread-safe.
> -----------------------------------------
>
> Key: HDFS-15249
> URL: https://issues.apache.org/jira/browse/HDFS-15249
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: federation
> Reporter: Toshihiro Suzuki
> Assignee: Toshihiro Suzuki
> Priority: Major
>
> ThrottledAsyncChecker should be thread-safe because it can be used by
> multiple threads when we have multiple namespaces.
> *checksInProgress* and *completedChecks* are respectively HashMap and
> WeakHashMap which are not thread-safe. So we need to put them in synchronized
> block whenever we access them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
