Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
Hadoop, mail # general - A question about high availability of NameNode


Copy link to this message
-
Re: A question about high availability of NameNode
Chris Nauroth 2013-11-11, 16:59
Hello,

For more background, I recommend reviewing the documentation on HDFS HA
with Quorum Journal Manager, particularly the Architecture section:

http://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/HDFSHighAvailabilityWithQJM.html

To summarize, edit logs are replicated, but it happens via interaction with
a separate set of daemons called the journal nodes.  The active persists
its edit logs to a majority of journal nodes (deployed in multiple).  The
standby reads the edits from the journal nodes and applies them locally.

The separate journal nodes provide redundancy, so that edits don't need to
be blocked if a single node dies.  By continuously reading the edits, the
standby can stay up to date with changes applied at the active and failover
quickly if needed.

Chris Nauroth
Hortonworks
http://hortonworks.com/

On Sat, Nov 9, 2013 at 7:29 PM, origin <[EMAIL PROTECTED]> wrote:

> Hey guys:
>     I was curious that why Active NameNode do not repilcation edits log to
> Standby NameNode like MySQL dose?

--
CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to
which it is addressed and may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If the reader
of this message is not the intended recipient, you are hereby notified that
any printing, copying, dissemination, distribution, disclosure or
forwarding of this communication is strictly prohibited. If you have
received this communication in error, please contact the sender immediately
and delete it from your system. Thank You.