Apache Ambari Interview Questions and Answers

Apache Ambari is nothing but a project which is solely focused to make the life simple while using the Hadoop management system. This software helps to manage and providing the comfort zone in terms of the following aspect like:

• Provisioning easily
• Managing to our convenience
• Monitoring Hadoop cluster
• Provides intuitive interface
• Support to backed up RESTful API
• It provides to use the easy Hadoop management web UI.

Tools that are needed to build the Ambari are:

• Pyton 2.6 or later
• Apache Maven 3.3.9 or later
• JDK 7±
• Node JS
• Xcode in case of Mac
• G++

The latest version of the ambary is 2.6.2.

• It will be protected Zeppelin notebook SSL credentials
• And we can set the appropriate HTTP headers to use the cloud object stores with HDP.
• In Ambari 2.6.1 version added the following features:
• The conditional installation of LZO packages through Ambari.
• In Ambari 2.6.0 version added the following features:
• Host recovery improvements for restart
• The moving masters with minimum impact and scale testing
• Distributed mode of AMS(Ambari Mertics System’s) along with the multiple collectors.
• Improvements in the Data Archival and Purging in Ambari Infra.

Apache ambary is a gift to the individuals who use the hadoop in their daily work life.

By using the Ambari, hadoop users will get the core benefits like:

• Simplified the installing process
• Simplified the configuration and overall management
• It has the centralized setup security process
• And it gives the out full visibility in terms of cluster health
• It has extensively extendable and has an option to customize if it needed.

Repository is a space where it hosts the software packages which can be used for download and the installing process. Apache provides different versions of the repositories that are in OS specific. Based on the internet access there are two formats of repositories.

• Tarball (.tar format,, if we don’t have internet access)
• Repo file(.repo format, for the temporary internet access)

It is used only there is no active internet connection is available. So we have to setup a local repository to, follow the below steps:

• First and foremost, set to host wit Apache httpd
• Next to download Tarball copy for every reposity entire contents
• Once it’s downloaded, one has to extract the content.

A local repository is nothing but hosted space in local environment. Usually the local repository is set up when the machines don’t have an active internet connection or restricted or very limited network access. With the use of this setup, the user will be able to obtain Ambari and HDP software packages.

In Ambari the different life cycle commands are:

• Status
• Install
• Configure
• Start
• Stop

Apache ambari is compatible with the 64-bit version and the following operating systems are used to implement the ambary:

• Ubuntu 12 & 14
• CentOS 6 and 7
• Debian 7
• SLES (Suse Linex Enterprise Server) 11
• OEL (Oracle Enterprise Linex) 6 and 7
• RHEL (Redhat Enterprise linux) 6 and 7

There are four types of Ambari respositories are available. They are:

• Ambari: These is used for Ambari server, Ambari agent and the other monitoing software packages.
• HDP: These is used to host hadoop stack packages.
• HDO – UTILS: It is used for all the utility packages for Ambari and HDP are available.
• EPEL: It is used to stands for “Extra packages for Enterprise Linux. It has a set of additional packages for the Enterprise linux.

There are 3 major function of Apache Ambari is:

• Provision of Hadoop Cluster
• Manage Hadoop Cluster
• Monitor Hadoop Cluster

No, In present Ambari can manage only one cluster. However, we can do remotely view the “views” of other clusters in the same instance.

The components of Ambari are imported for the integration and automation. Those are actually divided into three pieces of information. They are:

• Ambari stacks
• Ambari Blueprints
• Ambari API

Ambari built from scratch and it is deals with the Automation and the Integration problems carefully.

The Ambari shell is developed in the “JAVA” language and it is actually based on the spring shell framework and Ambari REST.

Ambari Shell can be provided an interactive and handy command line tool that actually supports the following:

• All the functionality available in Ambari Web-app
• All the availability context-aware
• Tab completion
• If any needed required parameters to be supported.

Yum is package manager which actually fetches the software packages from the repository.

On RHEL/CentOS, typically “yum”,

On SLES, typically “zipper”.

By using services tab, we can perform the following tasks:

• To start and sop of all services
• To Display the service operating summary
• Adding a service
• To change the configuration settings
• Performing service actions
• Rolling restarts
• Background operations monitoring
• By using the quick links
• Service removal
• YARN capacity refresh scheduler
• Auditing operations
• HDFS management
• In a storm environment Atlas management.

By using the host tab, we can perform following tasks:

• Managing the host components
• Performing the host related actions
• Decommissioning a Slave node or Master node
• Host status analysing
• Hosts page searching
• Establishing the rack awareness
• Removing or adding hosts to clusters
• Maintenance mode setting up
• Deleting the components

Yes, Ambari Python client can be used to make good use of the Ambari API’s.

The components of Apache ambary are:

• Ambari server: A server which handles all the metadata, and it contains the instance of Postgres database is what we call Ambari Server. In the cluster, each host attains one copy of Ambari agent by which Ambari server control the each host.
• Ambari agent: Agent is an active member of the host. Its main purpose is send heartbeats from the nodes to the Ambari server along with multiple operational metrics. Moreover, it helps to decide health status of the nodes.
• Ambari web: A client-side Java Script application that periodically accesses Ambari RESTful API in order to perform cluster operations is what we call Ambari Web UI. By using the RESTful API, it helps to asynchronous communication between the applications as well as the server.

There are two ways to setup the local repositories. It can be depends upon how we can execute and based on our active internet connection.

• All mirror the packages to the local repository.
• If in case first method doesn’t work out then to download all the Repository Tarball and start building the local repository.

The independent extensions that contributed to Ambari Codebase are as follows:

• Ambari SCOM Management Pack.
• Apache Slider view.

In the Ambari monitoring purpose there are two open sources. They are:

• Ganglia
• nagios

In Ambari the functionalities of ganglia are:

• cluster monitoring
• identified trending patens
• collecting metrics in the clusters
• To support the detailed heatmaps.

In ambari the functionalities of the Nagios are:

• The health checking of nodes and the sending alerts.
• To send the alert notifications of email and the server type.

The following code will be helps to create the Ambari client::

from ambari_client.ambari_api import  AmbariClient

headers_dict={‘X-Requested-By’:’mycompany’} #Ambari needs X-Requested-By header

client = AmbariClient(“localhost”, 8080, “admin”, “admin”, version=1,http_header=headers_dict)

print client.version

print client.host_url

print”n”

all_hosts = client.get_all_hosts()

print all_hosts

print all_hosts.to_json_dict()

print”n”

all_clusters = client.get_all_clusters()

print all_clusters.to_json_dict()

print all_clusters