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 Paul Smith <ps...@aconex.com> on 2009/10/02 00:02:30 UTC
Re: local node Quotas (for an R&D cluster)
On 23/09/2009, at 10:47 AM, Ravi Phulari wrote:
> Hello Paul here is quick answer to your question -
> You can use dfs.datanode.du.pct and dfs.datanode.du.reserved
> property in hdfs-site.xml config file to configure
> maximum local disk space used by hdfs and mapreduce.
>
> <property>
> <name>dfs.datanode.du.pct</name>
> <value>0.85f</value>
> <description>When calculating remaining space, only use this
> percentage of the real available space
> </description>
> </property>
>
> <property>
> <name>dfs.datanode.du.reserved</name>
> <value>1070000</value>
> <description>Reserved space in bytes per volume. Always leave
> this much space free for non dfs use.
> </description>
> </property>
Sorry for taking so long to reply here and actually use this
suggestion, but I'm not sure if this is working or not, I have placed
the above snippet in hdfs-site.xml and replicated that config around
the cluster. When I start up the cluster though, and go to the
management console page I don't see the overall DFS available size
decrease, nor does each node appear to report any difference in
available space.
I was working on the assumption that by adding these properties, these
free space metrics would change (go down) after a restart, but I don't
see that. Should I?
I'm using 0.20.1, r810220
cheers,
Paul