[
https://issues.apache.org/jira/browse/HDFS-17254?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17796514#comment-17796514
]
ASF GitHub Bot commented on HDFS-17254:
---------------------------------------
tomscut commented on code in PR #6307:
URL: https://github.com/apache/hadoop/pull/6307#discussion_r1426092305
##########
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/server/datanode/web/DatanodeHttpServer.java:
##########
@@ -144,7 +146,9 @@ public DatanodeHttpServer(final Configuration conf,
confForCreate.set(FsPermission.UMASK_LABEL, "000");
this.bossGroup = new NioEventLoopGroup();
- this.workerGroup = new NioEventLoopGroup();
+ int workerThreads = conf.getInt(DFS_DATANODE_NETTY_WORKER_NUM_THREADS_KEY,
+ DFS_DATANODE_NETTY_WORKER_NUM_THREADS_DEFAULT);
+ this.workerGroup = new NioEventLoopGroup(workerThreads);
Review Comment:
Better to use a cached thread pool here. `this.workerGroup = new
NioEventLoopGroup(workerCount, Executors.newCachedThreadPool())`.
> DataNode httpServer has too many worker threads
> -----------------------------------------------
>
> Key: HDFS-17254
> URL: https://issues.apache.org/jira/browse/HDFS-17254
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode
> Reporter: Liangjun He
> Assignee: Liangjun He
> Priority: Minor
> Labels: pull-request-available
>
> When optimizing the thread number of high-density storage DN, we found the
> number of worker threads for the DataNode httpServer is twice the number of
> available cores on node , resulting in too many threads. We can change this
> to be configurable.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
