Learning Path review questions: LP03

Learning Path review questions

01. How many deployment jobs can be run concurrently by a single agent?

A single agent can only run one deployment job at a time. Each agent processes a job sequentially, so if you have multiple deployment jobs that need to run concurrently, you will need multiple agents in your pool to handle them simultaneously.

If you need to run multiple jobs concurrently, you can scale up the number of agents in your deployment pool. This can either be a Microsoft-hosted agent (for ephemeral tasks) or a self-hosted agent (for tasks requiring a custom environment or pre-installed software).

02. What should you create to store values that you want to make available across multiple build and release pipelines?

To store values (such as connection strings, passwords, or configuration settings) that need to be available across multiple build and release pipelines, you should create variable groups in Azure DevOps.

Variable Groups allow you to store a set of variables that can be shared across multiple pipelines, ensuring that you only need to define these values once and can reference them across different stages and jobs.

How to use Variable Groups:

1. Create a Variable Group in the Azure DevOps Library.

2. Add key-value pairs of variables to the group.

3. Reference these variables in your pipelines via $(VariableName).

4. Variable groups also support linking to Azure Key Vault for securely storing sensitive data.

03. How can you provision the agents for deployment groups in each of your VMs?

To provision agents for deployment groups in your Virtual Machines (VMs) for Azure DevOps, follow these steps:

Create a Deployment Group in Azure DevOps:

1. Go to Pipelines > Deployment Groups.

2. Click New deployment group and name it (e.g., WebApp_DeploymentGroup).

Install the Agent on Each VM:

1. On the VMs where you want to run the deployment jobs, you need to install the Azure Pipelines agent.

2. Download and configure the agent by following these steps:

3. From the Deployment Group page, select New agent to download the agent installer.

4. Run the installer and configure the agent on each VM.

5. During the agent setup, provide the URL of your Azure DevOps organization and the authentication token to link the agent to the deployment group.

6. Once set up, the agent will be registered with the deployment group and available to run deployment jobs.

Verify Agent is Registered:

1. After the agent is installed on the VM, it should show up in the Deployment Group.

2. You can then use it to deploy your applications.

04. How can you identify a default release variable?

Default release variables in Azure DevOps are predefined system variables that provide information about the release pipeline, such as build details, deployment environment, etc.

To identify a default release variable, you can:

Check the list of predefined variables:

1. In the Azure DevOps release pipeline, navigate to the Variables tab.

2. There is a section for System Variables that contains default release variables.

3. Examples of default release variables include:

   • Build.BuildId: The ID of the build that triggered the release.

   • Release.ReleaseId: The ID of the release.

   • System.TeamProject: The name of the Azure DevOps project.

Use them in your YAML or Classic pipelines: Default release variables are referenced in the form $(VariableName), e.g., $(Release.ReleaseId).

05. What can you use to prevent a deployment in Azure DevOps when a security testing tool finds a compliance problem?

To prevent a deployment when a security testing tool finds a compliance problem, you can use approval gates and checks in your release pipeline.

Approvals and Checks:

In the pipeline, you can define conditions to require approval or check results before proceeding with the deployment. This could include integrating a security testing tool as part of the pipeline and setting up a gate to fail the deployment if a compliance issue is detected.

For example:

1. You can integrate a SonarQube task or OWASP ZAP (security scanner) in the pipeline to run automated security tests.

2. Configure an Approval Gate that requires a manual or automated security test to pass before proceeding to the next stage. If a compliance problem is detected, the gate will block the deployment from progressing.

3. Steps:

   • In your Release Pipeline, navigate to the stage where you want the security check to occur.

   • Under Pre-deployment conditions, configure a Quality gate or Approvals based on the security testing tool's results.

06. Even if you create exactly what a user requested at the start of the project, the solution will often be unsuitable for the same user. Why?

This happens because user needs evolve over time. Here are some reasons why the initial solution may become unsuitable:

1. Changing Requirements: As the project progresses, user requirements may change due to evolving business goals, market conditions, or new technology.

2. Lack of Communication: Sometimes, the initial requirements provided by the user were not fully comprehensive or were misinterpreted. As a result, the delivered solution might not align well with the actual needs.

3. Scalability Issues: The initial solution might not have been designed to scale as the user’s needs grow, leading to performance bottlenecks or inefficiencies.

4. User Experience: Over time, users may realize that the solution does not provide the expected usability, ease of use, or features they need to perform tasks more effectively.

5. Technological Advancements: New technologies or tools might emerge that offer better solutions than the one initially delivered.

6. Feedback and Iteration: As users interact with the solution, they may provide feedback that leads to refinements or major changes in how the solution should function to better meet their needs.

Summary

In software development, this is often addressed through agile methodologies, where the solution is delivered iteratively, allowing for ongoing feedback and improvements throughout the project lifecycle.