You are viewing a plain text version of this content. The canonical link for it is here.

Posted to common-user@hadoop.apache.org by Himanshu Sharma <hi...@yahoo-inc.com> on 2008/07/30 20:00:49 UTC

NameNode HA

The "dfs.name.dir" attribute, if given a comma-delimited list of directories,
replicates the name table in all such directories for redundancy.

My question is : whether it does so synchronously or asynchronously i.e does
it make sure that for each entry in EditLog file, all such replicated files
get an entry before that operation is termed as completed ?
-- 
View this message in context: http://www.nabble.com/NameNode-HA-tp18739140p18739140.html
Sent from the Hadoop core-user mailing list archive at Nabble.com.

Re: NameNode HA

Posted by Raghu Angadi <ra...@yahoo-inc.com>.

Himanshu Sharma wrote:
> The "dfs.name.dir" attribute, if given a comma-delimited list of directories,
> replicates the name table in all such directories for redundancy.
> 
> My question is : whether it does so synchronously or asynchronously i.e does
> it make sure that for each entry in EditLog file, all such replicated files
> get an entry before that operation is termed as completed ?

synchronously. All the files get the entry before the operation is 
considered complete.

Raghu.