DEVOPS Archives | CloudIQ Tech

1
2
3

Cypress in Azure DevOps Pipeline for Fast, Easy and Reliable Test Automation

Automation Testing helps complete the entire software testing life cycle (STLC) in less time and improve efficiency of the testing process.

Test Automation enables teams to verify functionality, test for regression and run simultaneous tests efficiently. In this article we will take a detailed look at the Automation Testing Tools available, standards and best practices to be followed during Test Automation.

Following the best practices for Software Testing Life Cycle (Unit testing, Integration Testing & System Testing) ensures that the client gets the software as intended without any bugs. End-to-end testing is the methodology used to test whether the flow of an application is performing as designed from start to finish. Carrying out end-to-end tests helps identify system dependencies and ensure the flow of right information across various system components and the system.

Ultimately Automation Testing increases the speed of test execution and the test coverage.

When to Choose Automation Testing

There is lots of regression work
GUI is same, but you have lot of often functional changes
Requirements do not change frequently
Load and performance testing with many virtual users
Repetitive test cases that tend well to automation & saves time
Huge projects
Projects that need to test the same areas

Steps to Implement Automation Testing

Identify areas within software to automate
Choose the appropriate tool for test automation
Write test scripts
Develop test suits
Execute test scripts
Build result reports
Find possible bugs or performance issues

Choosing your Automation Testing Tool

The strategy to adopt test automation should clearly define when to opt for automation, its scope and selection of the right kind of tools for execution. And when it comes to tools the top ones to go for are

Cypress
Selenium
Protractor
Appium(Mobile)

Why Cypress?

Cypress is a JavaScript based testing framework built for the modern web. Cypress helps to create End-to-end tests, Integration tests and Unit tests. Cypress takes a different approach compared to other testing frameworks, since it’s executed in the same run loop as the application. It also leverages a Node.js server to handle any task that needs to happen outside of the browser. With its ability to understand everything happening inside and outside of the browser, it produces more consistent results.

Key Features of Cypress

Automatic Waiting – No need for adding wait and sleep.
Spies, Stubs, and Clocks – Verify and control the behaviour of functions, server responses, or timers.
Network traffic control and monitoring – Easily control, stub, and test edge cases without involving your server. You can stub network traffic however you like.
Consistent Results – Cypress architecture doesn’t use Selenium or WebDriver. It is fast, consistent and does reliable tests that are flake-free.
Screenshots and Videos – View screenshots taken automatically on failure, or videos of your entire test suite when run from the CLI.

Azure CICD Setup with Cypress

Cypress runs on most of the following CI providers.

Azure DevOps / VSTS CI / TeamFoundation
BitBucket
CircleCI
Docker
GitLab
Jenkins
TravisCI

Azure DevOps – Steps to Integrate Cypress Automation Tests

Pre-Build Testing
Install the Node module and run application in test mode
Run the tests
Publish the test results
Cypress Containerization
Build the docker container of cypress
Push the image to container
Publish the Build

Before we get started here are the basic Cypress installation commands

Clean up the old results
$ rm -rf cypress/reports/
 
Run the cypress application with required spec file.
$ cypress run –spec \”cypress/integration/**/*.spec.ts\” // mention your spec file
 
Configure the mocha reports path for publishing test results.
–reporter junit –reporter-options ‘mochaFile=cypress/reports/test-output-[hash].xml,toConsole=true’
 
Uninstall the application.
$ npm uninstall cypress-multi-reporters; npm uninstall cypress-promise; npm uninstall cypres

Pre-Build Testing

It is critical to test the application before the Build, Deployment or Release. Essentially the process involves regression and smoke testing. And don’t forget the sanity checks before the build is deployed in the staging environment.

Cypress comes in handy for testing angular / JavaScript applications before they are deployed to staging or production environment.

Install the Node module and run application in test mode

Install the required node module of the application then run the application with test mode.

$ npm install –save-dev start-server-and-test

$ start-server-and-test start http://localhost:4200

Publish the test results

The results of the Cypress test execution are stored in specified path and are added to the Azure DevOps test results. Cypress supports JUnit, Mocha, Mochawsome test results reporter formats and provides options to create customised test results and merge all the test results as well.

Cypress Containerization

Cypress supports docker containerization and that makes it easy to set it up in a cluster environment like AKS. The Cypress base images are available at the link below.

https://github.com/cypress-io/cypress-docker-images

Copy the package.json and UI source code to the app folder and run the Cypress test. The following commands are used to run the docker and execute.

  script: |
        docker run -d -it --name cypressName:cypressImageTag bash
        docker commit -p cypressName:cypressImageTag
        docker stop cypressName
        docker rm -f cypressName
    
    - script: docker tag cypressName:cypressImageTag
      displayName: Tag Cypress image 
      
    
    - task: [email protected]
    displayName: Push image To Registry
    inputs:
        command: push
        azureSubscriptionEndpoint: azureSubscriptionEndpoint
        azureContainerRegistry: $(azureContainerRegistry)
        imageName: acrImageName:BuildId
 
    - script: sudo rm -rf /test-results/*
    displayName: Removing Previous Results
 
    - task: [email protected]
    displayName: 'Bash Script - cypress base image post-deployment'
    inputs:
        scriptPath: ./cypress-deployment.sh
        args: $(azureRegistry) $(cypressImageName) $(azureContainerValue) $(CYPRESS_OPTIONS) 
        continueOnError: true
    - task: [email protected]
    displayName: 'Publish Test Results ./test-results-*.xml'
    inputs:
 
    cypress-base-image-post-deployment.sh
 
    docker run -v $systemSourceDirectory:/app/cypress/reports --name vca-arp-ui 
    $cypress_Latestimage npx cypress run $cypressOptions bash

Now the container should be set up on on your local machine and start running your specs.

Cypress is simple and easily integrates with your CI environment. Apart from the browser support, Cypress reduces the efforts of manual testing and is relatively faster when compared to other automation testing tools.

Requirements:-

First step in getting up and running is to install VirtualBox. You can get appropriate version from the www.virtualbox.org
Need to install vagrant. The same procedure is applies; grab the installer from ww.vagrantup.com.

We can start the cluster setup, so we need the vagrant file for cluster setup using that only we can set it out.

Or Else clone the below git repository for getting sample vagrant file

https://github.com/coreos/coreos-vagrant

Now that every thing is downloaded, we can look at how to configure vagrant for your CoreOS development environment

Make copies and rename the configuration files: copy-user-data to user-data, and copy and rename config.rb.sample to config.rb
Open confi.rb so that you can change the a few parameters to get vagrant up and running properly.
```
 # Size of the CoreOS cluster created by Vagrant
        $num_instances=2
         
```

You may also want to tweak some other settings in config.rb. CPU, Memory settings can be modified as per your need.

 #Customize VMs
        $vm_gui = false
        $vm_memory = 1024
        $vm_cpus = 1
        $vb_cpuexecutioncap = 100

And then open the git shell to interact with vagrant

Go to your current working directory in your shell and issue this command

 vagrant up

You will see the things happening, which will look like this ,

Once the operation is completed you can verify that everything is up and running properly by logging in to one of the machines and using fleetctl to check the cluster

 vagrant ssh core-01
        fleetctl list-machines

If you see list of machines you created then you are finished, you now have a local cluster of CoreOS machines.

AWS, DEVOPS

Microservices – Aligning business and technology for closer collaboration, agility & flexibility

A microservices-based architecture introduces agility, flexibility and supports a sustainable DEVOPS culture ensuring closer collaboration within businesses and the news is that it’s actually happening for those who embraced it.

True, monolith apps architectures have enabled businesses to benefit from IT all along as it is single coded, simple to develop, test and run. As they are also based on a logical modular hexagonal or layered architectures (Presentation Layer responsible for handling HTTP requests and responding with either HTML or JSON/XML, Business logic layer, Database access and Apps integration) they cover and tie all processes, functions and gaps to an extent.

Despite these ground level facts, monolith software, which is instrumental for businesses embrace IT in their initial stages and which even exists today, is seeing problems. The growing complex business operation conditions are purely to be blamed.

So, how do businesses today address new pressures caused by digitization, continuous technology disruptions, increased customer awareness & interceptions and sudden regulatory interventions? The answer lies in agility, flexibility and scalability of the underlying IT infrastructure- the pillars of rapid adaptability to changes.

Monolith Apps, even though it is based on a well-designed 3 tier architecture, in the long run, loses fluidity and turns rigid. Irrespective of its modularity, modules are still dependent on each other and any minimal change in one module needs generation and deployment of all artifacts in each server pool, touched across the distributed environment.

Besides whenever there is a critical problem, the blame game starts amongst the UI developers, business logic experts, backend developers, database programmers, etc as they are predominantly experts in their domains, but have little knowledge about other processes. As the complexity of business operations sets in, the agility, flexibility and scalability part of your software is highly tested in a monolithic environment.

Here’s where Microservices plays a huge role as the underlying architecture helps you break your software applications into independent loosely coupled services that can be deployed and managed solely at that level and needn’t have to depend on other services.

For example, if your project needs you to design and manage inventory, sales, shipping, and billing and UI shopping cart modules, you can break each service down as an independently deployable module. Each has its own database, where monitoring and maintenance of application servers are done independently as the architecture allows you to decentralize the database, reducing complexity. Besides it enables continuous delivery/deployment of large, complex applications which means technology also evolves along with the business.

The other important aspect is that microservices promotes a culture wherein whoever develops the service is also responsible to manage it. This avoids the handover concept and the following misunderstandings and conflicts whenever there is a crisis.

In line with the DevOps concept, Microservices enables easy collaboration between the development and operations team as they embrace and work on a common toolset that establishes common terminology, as well as processes for requirements, dependencies, and problems. There is no denying the fact that DevOps and microservices work better when applied together.

Perhaps that’s the reason companies like Netflix, Amazon, etc are embracing the concept of microservices in their products. And for other new businesses embracing it, a new environment where agility, flexibility and closer collaboration between business and technology becomes a reality providing the much-needed edge in these challenging times.