[
https://issues.apache.org/jira/browse/HDFS-16805?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Haiyang Hu updated HDFS-16805:
------------------------------
Description:
Currently, multiple rpc servers are started in namenode, such as client rpc
server, service rpc server, lifeline rpc server, and each rpc server used the
same parameter 'ipc.server.read.threadpool.size' setting for the number of
reader threads for this server .
Consider according to different rpc server requirements set the number of
reader threads for this server，for example:
In client RPC server use parameter 'dfs.namenode.reader.count' ,
In service RPC server use parameter 'dfs.namenode.service.reader.count' ,
In lifeline RPC server use parameter 'dfs.namenode.lifeline.reader.count'
> Support set the number of RPC Readers according to different RPC Servers in
> NameNode
> ------------------------------------------------------------------------------------
>
> Key: HDFS-16805
> URL: https://issues.apache.org/jira/browse/HDFS-16805
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Haiyang Hu
> Assignee: Haiyang Hu
> Priority: Major
>
> Currently, multiple rpc servers are started in namenode, such as client rpc
> server, service rpc server, lifeline rpc server, and each rpc server used the
> same parameter 'ipc.server.read.threadpool.size' setting for the number of
> reader threads for this server .
> Consider according to different rpc server requirements set the number of
> reader threads for this server，for example:
> In client RPC server use parameter 'dfs.namenode.reader.count' ,
> In service RPC server use parameter 'dfs.namenode.service.reader.count' ,
> In lifeline RPC server use parameter 'dfs.namenode.lifeline.reader.count'
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
