[ 
https://issues.apache.org/jira/browse/HDFS-9199?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14944104#comment-14944104
 ] 

Mingliang Liu commented on HDFS-9199:
-------------------------------------

Thanks for reporting this. I agree that we need to make the replication config 
keys' name consistent.

I see
{code}
  public static final String  DFS_REPLICATION_MAX_KEY = "dfs.replication.max";
  public static final int     DFS_REPLICATION_MAX_DEFAULT = 512;
{code}

and
{code}
  public static final String  DFS_NAMENODE_REPLICATION_CONSIDERLOAD_KEY = 
"dfs.namenode.replication.considerLoad";
  public static final boolean DFS_NAMENODE_REPLICATION_CONSIDERLOAD_DEFAULT = 
true;
  public static final String  DFS_NAMENODE_REPLICATION_INTERVAL_KEY = 
"dfs.namenode.replication.interval";
  public static final int     DFS_NAMENODE_REPLICATION_INTERVAL_DEFAULT = 3;
  public static final String  DFS_NAMENODE_REPLICATION_MIN_KEY = 
"dfs.namenode.replication.min";
  public static final int     DFS_NAMENODE_REPLICATION_MIN_DEFAULT = 1;
{code}

Do we need to consider all of them as well? Which clause of renaming is 
preferred?

I think these keys are all used in namenode.

> rename dfs.namenode.replication.min to dfs.replication.min
> ----------------------------------------------------------
>
>                 Key: HDFS-9199
>                 URL: https://issues.apache.org/jira/browse/HDFS-9199
>             Project: Hadoop HDFS
>          Issue Type: Bug
>            Reporter: Allen Wittenauer
>            Assignee: Mingliang Liu
>
> dfs.namenode.replication.min should be dfs.replication.min to match the other 
> dfs.replication config knobs.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to