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

Posted to user@hbase.apache.org by Saurabh Sehgal <sa...@gmail.com> on 2011/10/04 03:25:44 UTC

namenode failures

Hi,

I am looking into HBase and would like to know if there are any best
practices for recovering from namenode failures.

I found this while doing some research online:

http://wiki.apache.org/hadoop/NameNodeFailover

, but I would also like to hear opinions from the HBase community.

Is configuring an NFS mount the only option to reliably recover from a
namenode failure ?

Thank you,

Saurabh

Re: namenode failures

Posted by Suraj Varma <sv...@gmail.com>.

It's a little bit of a moving target.

See http://www.cloudera.com/blog/2011/02/hadoop-availability/ for a
summary of efforts that are in motion

See http://www.cloudera.com/blog/2009/07/hadoop-ha-configuration/
for one approach that uses DRDB.

--Suraj


On Mon, Oct 3, 2011 at 6:25 PM, Saurabh Sehgal <sa...@gmail.com> wrote:
> Hi,
>
> I am looking into HBase and would like to know if there are any best
> practices for recovering from namenode failures.
>
> I found this while doing some research online:
>
> http://wiki.apache.org/hadoop/NameNodeFailover
>
> , but I would also like to hear opinions from the HBase community.
>
> Is configuring an NFS mount the only option to reliably recover from a
> namenode failure ?
>
> Thank you,
>
> Saurabh
>