DevOps Use Cases

2) A telecom in US needed to reduce cycle times for

development, building, and testing. We created a backbone of
automation to get internal customers on board.

 Telco service provider was rolling out new versions of network

cycling through design and deployment. We implemented rapid
design, test, and deploy 10X faster. Today I add security patches
daily rather than every three months.

Jenkins Master and Slave Concept

A Jenkins master comes with the basic installation of Jenkins, and in
this configuration, the master handles all the tasks for your build
system.
You may also enjoy: Getting Started With Jenkins: The Ultimate
Guide

If you are working on multiple projects you may run multiple jobs on
each and every project. Some projects need to run on some particular
nodes, and in this process, we need to configure slaves. Jenkins slaves
connect to the Jenkins master using the Java Network Launch Protocol.
Jenkins Master and Slave Architecture

The Jenkins master acts to schedule the jobs and assign slaves and
send builds to slaves to execute the jobs.
It will also monitor the slave state (offline or online) and getting back
the build result responses from slaves and the display build results on
the console output. The workload of building jobs is delegated to
multiple slaves.

Steps to Configure Jenkins Master and Slave

Nodes
1. Click on Manage Jenkins in the left corner on the Jenkins
dashboard.
2. Click on Manage Nodes.

3. Select New Node and enter the name of the node in the Node
Name field.
 4. Select Permanent Agent and click the OK button. Initially, you
will get only one option, "Permanent Agent." Once you have one
or more slaves you will get the "Copy Existing Node" option.

5. Enter the required information.

Some required fields include:
Name: Name of the Slave. e.g: Test
Description: Description for this slave (optional). e.g: testing slave
# of Executors: Maximum number of Parallel builds Jenkins master
perform on this slave. e.g: #2
Remote root directory: A slave needs to have a directory dedicated to
Jenkins. Specify the path to this directory on the agent. e.g: /home/
Usage: Controls how Jenkins schedules builds on this node. e.g: Only
build jobs with label expressions matching this node.
Launch method: Controls how Jenkins starts this agent. e.g: Launch
agent agents via SSH

6. Enter the Hostname in the Host field.

7. Select the Add button to add credentials. and click Jenkins.
8. Enter Username, Password, ID, and Description.

9. Select the dropdown menu to add credentials in

the Credentials field.
10. Select the next dropdown to add the Host Key Verification Strategy
under Non verifying Verification Strategy.
11. Select Keep this agent online as much as possible in the
Availability field.

12. Click the Save button.

Creating a Freestyle Project and Running on

The Slave Machine
1. Click on Save and it will redirect to job's view page
 2. On the left pane, click the Build Now button to execute your
Pipeline.
3. We can verify the history of the executed build under the Build
History by clicking the build number.
4. Click on the build number and select Console Output. Here you
can see the executed job in the remote host and output.

Creating a Pipeline and Running on The Slave

Machine
1. Click New Item in the top left corner on the dashboard.
2. Enter the name of your project in the Enter an item name field,
and select the Pipeline project, and click OK button.
3. Enter Description (optional).
4. Go to the Pipeline section, make sure the Definition field has
the Pipeline script option selected.
5. Copy and paste the following declarative Pipeline script into a
script field.
node('test'){
stage('stage1') {
sh '''echo stage1 steps'''
}
stage('stage2') {
sh '''echo stage2 steps'''
}
stage('stage3') {
sh '''echo stage3 steps'''
}
}

6. Click on Save, it will redirect to the Pipeline view page.

7. On the left pane, click the Build Now button to execute your
Pipeline.
8. After Pipeline execution is completed, the Pipeline view will be as
shown below.
9. We can verify the history of executed build under the Build
History by clicking the build number.
10. Click on build number and select Console Output. Here you can
see that the pipeline ran on a slave machine.
Basic Questions
1) DevOps ! How can you define it in your words ?
Its highly effective daily collaboration between software developers
and IT operations / web operation engineers to produce a working
system or release software.

A devOps implementation is generally aligned with Agile

methodologies where deploying working software to Production is
generally the highest priority. On Agile implementations, emphasis is
placed on people over processes, so a DevOps engineer must be
willing to work very closely with Agile development teams to ensure
they have an environment necessary to support functions such as
automated testing, continuous Integration and continuous Delivery. On
a traditional implementation, without DevOps, the operations team is
often isolated from developers, often working under a help desk model
under general service level agreements where the system operations
team treats developers as a customer. This is a proven model which
obviously can work very well, but in a DevOps environment,
development and operations are streamlined and barriers between the
two groups should not exist.

2) Why we need DevOps ?

Companies are now facing the need to delivering more and faster and
better applications to meet the ever more pressing demands of
conscious users to reduce the " Time To Market ". Devops often helps
deployment to happen very fast.

3) What is agile development and Scrum ?

Agile development used as an alternative to Waterfall development
practice. In Agile, the development process is more iterative and
incremental, there is more testing and feedback at every stage of
development as opposed to only the last stage in Waterfall.

Scrum is used to manage complex software and product

development, using iterative and incremental practices. Scrum has
three roles ie product owner, scrum master, and team.

4) Can we consider DevOps as an agile

methodology ?
Of course! DevOps is a movement to reconcile and synchronize
development and production start through a set of good practices . Its
emergence is motivated by a deep changing demands of business,
who want to speed up the changes to stick closer to the requirements
of business and the customer.
5) What is DevOps engineer's duty with regards to
Agile development ?
DevOps engineer work very closely with Agile development teams to
ensure they have an environment necessary to support functions such
as automated testing, continuous Integration and continuous Delivery.
DevOps engineer must be in constant contact with the developers and
make all required parts of environment work seamlessly.

Technical Questions
6) Have you worked on containers ?
Containers are form of lightweight virtualization, more heavy than
chroot but lighter than hypervisors. They provide isolation among
processes while using same kernel as the host machine, and cgroups
functionality within kernel. But container formats differ among
themselves in a way that some provide more VM-like experience while
other containerize only application.

LXC containers are most VM-like and most heavy weight, while
Docker used to be more light weight and was initially designed for
single application container. But in more recent releases Docker
introduced whole machine containerization features so now Docker
can be used both ways. There is also rkt from CoreOS and LXD from
Canonical, which builds upon LXC.

7) What is Kubernetes? Explain

It is massively scalable tool for managing containers.

It is used internally on huge deployments and because of that it is

maybe the best option for production use of containers.

It supports self healing by restating non responsive containers, it pack

containers in a way that they take less resources and has many other
great features.
8) What is the function of CI (Continuous Integration)
server ?
CI server function is to continuously integrate all changes being made
and committed to repository by different developers and check for
compile errors.

It needs to build code several times a day, preferably after every

commit so it can detect which commit made the breakage if the
breakage happens.

9) What is Continuous Delivery ?

Is it practice of delivering the software for testing as soon as it is build
by CI (Continuous Integration) server's.

It requires heavy use of Versioning Control System for so always

available to developers and testers alike.

10) What is Vagrant and what is it used for ?

Vagrant is a tool that can create and manage virtualized (or
containerized) environments for testing and developing software. At
first, Vagrant used virtualbox as the hypervisor for virtual
environments, but now it supports also KVM.

11) Do you ever used any scripting language ?

As far as scripting languages go, the simpler the better. In fact, the
language itself isn’t as important as understanding design patterns
and development paradigms such as procedural, object-oriented, or
functional programming.

Currently, several scripting languages are available so the question

arises : what is the most appropriate language for DevOps approach?
Simply everything , it depends on the context of the project and tools
used for example if Ansible used its good have knowledge in Python
and if its for Chef its on Ruby.
12) What is the role of a configuration management
tool in devops ?
Automation plays an essential role in server configuration
management. For that purpose we use CM tools , they store
information about versions and builds of the software and testware
and provide the traceability between software and testware.

13) What is the purpose of CM tools and which one

you have used ?
Configuration Management tools' purpose is to automatize
deployment and configuration of software on big number of servers.
Most CM tools usually use agent architecture which means that every
machine being manged needs to have agent installed. My favorite tool
is one that uses agentless architecture - Ansible. It only requires SSH
and Python. And if raw module is being used, not even Python is
required because it can run raw bash commands. Other available and
popular CM tools are Puppet, Chef, SaltStack.

14) What is OpenStack ?

OpenStack is often called Cloud Operating System, and that is not far
from the truth. It is the complete environment for deploying IaaS which
gives you possibility of making your own cloud similar to AWS. It is
highly modular and consists of many sub-projects so you can pick and
chose which functionality you need. OpenStack distribution are
available from Red Hat, Mirantis, HPE, Oracle, Canonical and many
others. It is completely open source project but some vendors make
proprietary distributions.

15) Classify Cloud Platforms anategory ?

Cloud Computing software can be classified as Software as a Service
or SaaS, Infrastructure as a Service or IaaS and Platform as a Service
or PaaS.
SaaS is peace of software that runs over network on remote server
and has only user interface exposed to users, usually in web browser.
For example salesforce.com.

Infrastructure as a service is a cloud environment that exposes VM to

user to use as entire OS or container where you could install anything
you would install on your server. Example for this would be
OpenStack, AWS, Eucalyptus.
PaaS allows users to deploy their own application on the preinstalled
platform, usually framework of application server and suite of
developer tools. Examples for this would be OpenShHeroku.

16) What are easiest ways to build a small cloud ?

VMfest is one one of the options for making IaaS cloud from
VirtualBox VMs in no time. If you want a lightweight PaaS there is
Dokku which is basically a bash script that makes PaaS out of Dokku
containers.

17) What is AWS (Amazon Web Services)? Did got

chance to work on Amazon tools ?
AWS provides a set of flexible services designed to enable companies
to create and deliver products with greater speed and reliability using
AWS and DevOps practices . These services simplify commissioning
and infrastructure management , application code deployment ,
automated software release process and monitoring of the application
and infrastructure performance. Amazon used tools like AWS
CodeCommit, AWS CodeDeploy, AWS CodePipeline etc, that helps to
make devops easier.

18) What is EC2 ?

Amazon EC2 Container Service (ECS) is a highly scalable container
management service and high performance that supports the Docker
containers and allows you to easily run applications on a cluster
managed by Amazon EC2 instances.
The EC2 service is inseparable from the concept of Amazon Machine
Image - AMI . The May is Indeed the image of a virtual machine That
Will Be Executed . EC2 based on XEN virtualization , that's why it is
quite easy to move XEN servers to EC2 .

19) Do you find any advantage of using NoSQL

database over RDBMS ?
Typical web applications are built with a three-tier architecture. To
carry the load, more Web servers are simply added behind a load
balancer to support more users. The ability to scale out is a key
principle in the world of cloud computing, more and more important in
which VM instances can be easily added or removed to meet demand.

However, when it comes to the data layer, relational databases

(RDBMS) does not allow a passage to the simple scale and do not
provide a flexible data model. Manage more users means adding
more servers and large servers are very complex, owners and
disproportionately expensive, in contrast to low-cost hardware, the
"commodity hardware", architectures in the cloud. Organizations are
beginning to see performance issues with their relational databases
for existing or new applications. Especially as the number of users
increases, they realize the need for a faster and more flexible basis.
This is the time to begin to assess and adopt NoSQL database like in
their Web applications.

20) What are the main SQL migration difficulties

NoSQL ?
Each record in a relational database according to a schema - with a
fixed number of fields (columns) each having a specified object and a
data type. Each record is the same. The data is denormalized in
several tables. The advantage is that there is less of duplicate data in
the database. The downside is that a change in the pattern means
performing several "alter table" that require expensive to lock multiple
tables simultaneously to ensure that change does not leave the
database in an inconsistent state.
With databases data, on the other hand, each document can have a
completely different structure from other documents. No additional
management is required on the database to manage changes in the
schemes.

21) What are the benefits of NoSQL databases

Documents ?
The main advantages of document databases are the following :

 flexible data model data can be inserted without a defined schema

and format of the data that is inserted can change at any time ,
providing extreme flexibility , which ultimately allows a significant
agility to business
 Consistent , high-performance Advanced NoSQL database
technologies are putting cache data , transparently, in system
memory ; a behavior that is completely transparent to the developer
and the team in charge of operations .
 Some easy scalability NoSQL databases automatically propagate
data between servers , requiring no participation applications. Servers
can be added and removed without disruption to applications , with
data and I/O spread across multiple servers.

22 ) What are the main advantages of Git over CVS

?
The biggest advantage is that Git is distributed while CVS is
centralised. Changes in CVS are per file, while changes (commits) in
Git they always refer to the whole project. Git offers much more tools
than CVS.

23) Difference between containers and virtual

machines ?
Each VM instantiation requires starting a full OS. VMs take up a lot of
system resources. This quickly adds up to a lot of RAM and CPU
cycles. Container host uses the process and file system isolation
features of the linux kernel.
24) What is CoreOS, and what are alternatives ?
CoreOS is striped down linux distribution meant for running
containters, mainly with its own rkt format but others are also
supported. It was initially based on ChromeOS and supported Docker.
The alternatives to this are canonical's ubuntu snappy or red hat
enterprise linux atomic host. Of course, Containers can also be ran on
regular Linux system.

25) What is Kickstart ?

It is a way to install Red Hat based systems by automated way. During
manual install process, Anaconda installer creates file anaconda-
ks.cfg which then can be used with system-config-kickstart tool to
install same configuration automatically on multiple systems.

26) What are tools for network monitoring? List few

For example, Nagios, Icinga 2, OpenNMS, Splunk and Wireshark.
Those tools are used to monitor network traffic, network quality and
detect network problems even before they arise. Of those listed, only
Splunk is proprietary other are open source.

27) What is Juju ?

Juju is orchestration tool primarily for ubuntu for management,
provision and configuration on Ubuntu systems. It is was initially
written in Python and since have been rewritten in Go.

28) Give me an examples of how you would handle

projects ?
As a DevOps engineer, I would demonstrate a clear understanding of
DevOps project management tactics and also work with teams to set
objectives, streamline workflow, maintain scope, research and
introduce new tools or frameworks, translate requirements into
workflow and follow up. I would resort to CI, release management and
other tools to keep interdisciplinary projects on track.
29) What is post mortem meetings ?
It is a meeting where we discuss what went wrong and what steps
should be taken so that failure doesn't happen again. Post mortem
meetings are not about finding the one to be blamed, they are for
preventing outages from reoccurring and planing redesign of the
infrastructure so that downtime can be minimised. It is about learning
from mistakes.

30) What you know about serverless model ?

Serverless refers to a model where the existence of servers is hidden
from developers. It means you no longer have to deal with capacity,
deployments, scaling and fault tolerance and OS. It will essentially
reducing maintenance efforts and allow developers to quickly focus on
developing codes.

Examples are Amazon AWS Lambda and Auth0 serveless platform.

Devops Example : Deploying

Applications with Ansible
Ansible is a lightweight, extensible solution for automating your
application provisioning. Ansible has no dependencies other than
Python and SSH. It doesn’t require any agents to be set up on the
remote hosts and it doesn’t leave any traces after it runs either. It
allows you to significantly simplify our operations by creating easy
YAML based playbooks. It’s good for configuration automation,
deployments and orchestration.

Components of Ansible
Playbooks : Ansible playbooks are a way to send commands to
remote computers in a scripted way. Instead of using Ansible
commands individually to remotely configure computers from the
command line, you can configure entire complex environments by
passing a script to one or more systems.
Ansible playbooks are written in the YAML data serialization format. If
you don't know what a data serialization format is, think of it as a way
to translate a programmatic data structure (lists, arrays, dictionaries,
etc) into a format that can be easily stored to disk. The file can then be
used to recreate the structure at a later point. JSON is another popular
data serialization format, but YAML is much easier to read.

Let's look at a basic playbook that allow us to install a web application

(nginx) in a multiple hosts :

hosts: webservers
tasks:
- name: Installs nginx web server
apt: pkg=nginx state=installed update_cache=true
notify:
- start nginx

handlers:
- name: start nginx
service: name=nginx state=started
The hosts file : (by default under /etc/ansible/hosts) this is the
Ansible Inventory file, and it stores the hosts, and their mappings to
the host groups (webservers ,databases etc)

[webservers] 10.0.15.22
# example of setting a host inventory by IP address.
# also demonstrates how to set per-host variables.

[repository_servers] example-repository
#example of setting a host by hostname. Requires local lookup
in /etc/hosts
# or DNS.
[dbservers] db01

The SSH key : For the first run, we'll need to tell ansible the SSH and
Sudo passwords, because one of the thing that the common role does
is to configure passwordless sudo, and deploy a SSH key. So in this
case ansible can execute the playbook’s commands in the remote
nodes (hosts ) and deploy the web application nginx.