ambari support more than one standby namenode or not

[#1]

As we know, Hadoop 3.x support more than 2 namenode or more than 1 standby namenode now. https://issues.apache.org/jira/browse/HDFS-6440

So if ambari which is 2.7.1.0 now support this feature or not?

[#2]

@owen chaos

Ambari 2.7 provides support for NameNode Federation:

An HDFS federation allows you to scale a cluster horizontally by configuring multiple namespaces and NameNodes. The DataNodes in the cluster are available as a common block of storage for every NameNode in the federation.

https://docs.hortonworks.com/HDPDocuments/Ambari-2.7.0.0/managing-and-monitoring-ambari/content/amb_...

.

Also Ambari 2.7.0 Release Notes "New Features" (HDFS: NameNode Federation) section talks about this: https://docs.hortonworks.com/HDPDocuments/Ambari-2.7.0.0/bk_ambari-release-notes/content/ambari_reln...

[#3]

Thank you for your reply.

More than one standby NameNode means we can tolerant more than two namenode failure.

As we have only two namenode in Hadoop 2.x, the cluster will go down once the active and standby namenode went down.

And we can have one Active NameNode and 2 or 4 Standby NameNode in Hadoop3 which can provide a better high availability.

So I'am asking if Ambari could support this feature.

[#4 · Accepted solution]

Hi @owen chaos

As Jay mentioned, this is supported in Ambari 2.7 with HDP 3

Have a look at the following document, from page 34 onwards,

The HDFS NameNode High Availability feature enables you to run redundant NameNodes in the same cluster in an Active/Passive configuration with a hot standby. This eliminates the NameNode as a potential single point of failure (SPOF) in an HDFS cluster. As of Hadoop 3.0, you can configure more than one backup NameNode.
...
This guide provides an overview of the HDFS NameNode High Availability (HA) feature, instructions on how to deploy Hue with an HA cluster, and instructions on how to enable HA on top of an existing HDP cluster using the Quorum Journal Manager (QJM) and ZooKeeper Failover Controller for configuration and management. Using the QJM and ZooKeeper Failover Controller enables the sharing of edit logs between the Active and Standby NameNodes.

mainly, this is where you setup multiple NN's in ambari 2.7 with hdp3;

<property>
 <name>dfs.ha.namenodes.mycluster</name>
 <value>nn1,nn2,nn3</value>
 <description>Unique identifiers for each NameNode in the
 nameservice</description>
</property><br>

Also, note; the minimum number of NameNodes for HA is two, but you can configure more. You should not exceed five NameNodes due to communication overhead. Three NameNodes are recommended.

[#5]

Thank you very much.

But something still not clear, which is how I setup the third Namenode in Ambari 2.7.

I already update hdfs-site.xml with a nn3 config, and restart HDFS which functionally normal without the third Namenode.

If we should create the the third Namenode by Ambari API, as I found no place or button in Ambari GUI.

[#6]

Were you able to add another third namenode. I can't find a way doing this via ambari. Can anyone share guide on adding this via REST api.

[#7]

@saqie,  as this thread was marked 'Solved' in 2018, you would have a better chance of receiving a useful response by starting a new thread. This will also provide you with the opportunity to provide details specific to your issue that could aid others in providing a more tailored answer to your question.

[#8]

I also wanna know how to make third namenode

I need help