AWS Certified DevOps Engineer – Professional Practice Questions Part 2

When the status of Trusted Advisor checks are changing, you are to provide a solution to notify the team properly. For example, when Trusted Advisor’s cost optimization checks have just identified an Amazon EC2 instance with a low use system, the operating team should be informed by the Slack Channel to respond to the changes in status and potentially reduce cost. In order to satisfy this requirement, what two combinations should the company’s DevOps Engineer use? (Choose two)

Juliet’s company needs to reduce costs and all teams must develop an operational cost reduction plan, where possible. You work within the DevOps team and your team manager asked the team to think of how to optimize the cost of usage of AWS resources. Which tool from the following DOES NOT help in cost optimization?

An organization has its application on AWS EC2 instance, and they give secure access to AWS Service APIs by defining IAM roles. Now the organization wants to fetch the API keys for using with AWS SDK. What should you as developer professional engineer do to configure the application on the instance?

Austin wants to deploy his new application in AWS with the Blue/Green Deployment technique. For Blue/Green Deployment he wants to use Java AWS SDK. The application requires services like ELB and Auto-scaling group for both current and new releases. But it may be possible that after the new ASG is attached to ELB, the tests fail. From the following, which is an important consideration in failed condition?

A company has video games, and for that, they are creating new APIs. The number of reads is 100 times more than writes. The top 1% of scores are 100 times more frequently hit than the rest of the scores. How will you as developer professional engineer design the use of DynamoDB?

A large number of Lambda functions with an AWS Code Deploy service are deployed by your DevOps team. In CodeDeploy, the team has already created new apps. References to the deployment groups that include configurations used during deployment are then set up. Which methods are supported for the deployment configurations? (Choose 3)

It takes more than four hours for your current log analysis application to report the top 10 users of your web application. You were asked to implement a system that would allow you to report this information in real time, make sure that the report is always up to date and that the number of requests for your web application has increased so it will be able to handle it. Choose the cost-effective option that can meet the requirements.

Management observed the increase in billing cost after reviewing the last quarter monthly bills from Amazon. On researching for the increase in cost, you found that one of the new services is doing a lot of GET bucket API calls to S3 to build a metadata cache of all objects in the application bucket. You need to provide a solution, which reduces the amount of these GET bucket API calls. What would be your strategy?

You have a complex system, involving multiple three-tier applications, networking, and IAM policies. The requirements for the new system remain so you still do not know how many AWS components there will be in the final design. The AWS CloudFormation is used to define these AWS resources so that your infrastructure can be automated and version controlled. How will you use AWS CloudFormation to provide your customers with agile new environments in an economical and reliable way?

William’s team has several pipelines for AWS CodePipeline service but, even when some pipelines have failed, you have found your team is unable to get any notification. You want to set up a feature in order to receive an email and an SMS from your on-call team if the whole or a certain phase of the pipeline changes its status to FAILED. To enforce this, what combinations of services should you use? (Choose 2)

HTTP health check has been enabled for Elastic Load Balancing. You see that all instances take health checks after looking at the AWS management console, but your customers are informed that your site is not responding. What is the reason for this?

You decided that deployments in blue/green technique would prove beneficial for your company, after a regular review with the development team. What should you do to incorporate this technique?

You have been assigned to use AWS OpsWorks to implement the scalable distributed system. You have to scale up your distributed system on demand. Every node must have a configuration file that includes the hostnames of the other instances in the layer, as it is distributed. How should AWS OpsWorks be configured to manage to scale dynamically?

A company saves its code for web applications in the Git Repository, and now needs to deploy this application in AWS, which is Node.js. How will this be done? (Choose 2)

You use CloudFormation to organize the resources of your application. During your test phase, your Amazon RDS type instance was changed and the instance was re-created, which resulted in the loss of test data. How are you to avoid this in the future?

The use of AWS CodeStar for the central control and monitoring of a new application on an AWS system is in consideration. It is a web application, which is Node.js and deployed in a Lambda function. The project was designed smoothly in 10 minutes with a template given by CodeStar. What part of the information should you view and track in the AWS CodeStar dashboard for this project?

You are working for a start-up that has developed a new mobile app for photo sharing. Your application has grown in popularity in recent months, resulting in a decrease in application performance due to the increase in load. Your application features a two- tier architecture consisting of an Auto-scaling PHP application tier and MySQL RDS instance initially deployed with CloudFormation. The auto-scaling group has a min value of 4 and a max value of 8. Due to the high CPU usage of the instances, the desired capacity is now 8. Once analyzed, you are confident that performance problems stem from a CPU capacity restriction, while memory use is low. You thus decide to move from M3 instances to C3 instances that are computer-optimized.
How would you deploy this change to reduce interruptions for your end-users as minimum as possible?

You are responsible for an insurance company for the daily operation of the online quota system used in your company to offer insurance quotes to the public. Your company wishes to use the application logs created by the system to better understand customer behavior. You have developed a log management system that meets the following requirement: All log entries must be kept by the system even during an unplanned instance failure. The consumer monitoring teams need direct access to logs from the last seven days, and eventually, all past logs must be accessible to the fraud investigation teams, but they will wait until 24 hours before the logs are usable. What steps would you take to achieve these requirements? (Choose 3)

A company wants to make easier deployment and reduce the time taken by the deployment for the improvement of deployment as they deploy five times a week at max. They hired you as the DevOps engineer to create a CI pipeline that can build AMIs. How will you do this?

John has an ELB with an Auto-scaling group and he needs to phase-out all old instances and replace them with new types of instance. What steps should he take to fulfill his task? (Choose 2)

Alan has created a web app and stores it for static website hosting in an Amazon S3 bin. This application can access the data in Amazon DynamoDB table using the AWS SDK for JavaScript in the browser. How can you ensure that API keys are kept secure to access DynamoDB data?

An educational institution uses AWS to host its application. It uses CloudFormation templates and Auto-scaling. Now, the institute observes that the number of students using applications is increasing and it is facing many performance issues. If the EC2 instance type is changed to C3, the performance will improve. What is the proper way to change the instance type?

Louis deployed his application in AWS. The application has two ASGs attach to an ELB. From these two ASGs, ASG1 is live and ASG2 is idle. The new version is deployed on ASG2 and fully reviewed once new releases are published. If no error is found, ASG2 comes alive and ASG1 gets idle. What kind of deployment / delivery method is possible to use in this case? (Choose 2)

A company has an application, in which the changes occur, but if the changes fail, then they rollback the updates, but it takes 5-6 hrs for rolling back. How will you as a DevOps engineer provide a solution to reduce the rolling back duration?

You have just joined an IT company to take responsibility for the management of AWS resources with another DevOps engineer. After a while, you find that an Elastic Load Balancer health check timer needs to be changed. The ELB was set up a year ago with Auto-scaling settings and you can find the CloudFormation template for this setup, which is in a CodeCommit repository. During the year, such configurations can be highly altered by somebody. However, the modification details for the CloudFormation stack are not recorded. What are you supposed to do to update the ELB health check timer?

You will be working for a SaaS organization as the new head of operations. Your CTO asked you to simplify and speed up debugging every part of the operation. She complains she does not know what is happening in a complex, service-based architecture because the developers only log onto the disk and with so many services, it is very difficult to find mistakes in logs. How can you best fulfill your CTO’s requirement?

In order to meet an ongoing compliance audit in the business, Wayne needs to test and set up CloudWatch alarms for the AWS resources in production like EC2, DynamoDB, Lambda, etc. To save time, he plans to use existing CloudWatch Metrics and create CloudWatch alarms accordingly. Which scenarios are suitable for this? (Choose 2)

An organization has an application of which it released a new feature, making it a highly available application. For testing the new feature, A/B testing is used. The logs from each updated instance need to analyze real-time to observe the working. If the behavior of the log is anomalous, the instances are changes to a more stable one. What would be your strategy in this scenario?

Jordan needs to find a solution to cost optimization without impacting the services of AWS. What should he do? (Choose 2)

The design of a CI / CD pipeline is your responsibility for the current request. One of the main requirements is that the AWS applications should be highly available and they should auto-heal when errors occur. Which AWS resources will be helpful for you? (Choose 2)

Joel plans to use the Amazon RDS facility for fault tolerance of the application and needs its features. What is one of the benefits of Amazon RDS Multi-Availability Zone?

A company has an application in which whenever a user uploads a photo in the app, a message is sent immediately to their friends. To save the photo information per user, an AWS DynamoDB table has been created and the DynamoDB stream is enabled to capture the item changes. Which service can be integrated with DynamoDB in order to process and notify events?

The firm has a DevOps department that handles all other departments of the company’s AWS accounts. A number of CloudWatch Event Rules are defined in the master AWS account to allow CloudWatch Events to be sent in other AWS accounts. In order to do this, permissions are set up in the master account, which is in region ap-southeast-2 with default CloudWatch event bus. Which entities can be added to allow submitting events on the master account in CloudWatch Event Bus? (Choose 2)

For the integration of thousands of Microservices, AWS ECS was used by a major IT company. There is a new management requirement that every state change must be reported to a group channel of Microsoft teams either by ECS container instances or ECS tasks. When a change occurs, the alert notifications should be readable and sent directly. Which two methods in combinations meet this requirement? (Choose 2)

An enterprise has a hybrid network consisting of multiple EC2 instances and Raspberry Pi devices. In order to manage all servers like patching strategy on both cloud and an on-premises environment, the enterprise plans to use AWS System Manager. The agents are also installed on all servers. Now, it wants to identify which server is the Raspberry Pi server from the Managed Instances Console. Which feature of the Raspberry Pi distinguishes it from others?

An Auto-scaling group is configured to launch EC2 instances for the application, but you observed that the Auto-scaling group is not launching the instances in the right proportion. Instances are launched too fast. How will you resolve this issue? (Choose 2)

In a very large organization, there are multiple applications, which are constructed in different programming languages. How can you deploy these applications as fast as possible?

Aidan has a company in which he helps his IT team for setting up the AWS CodeCommit repositories. He and his team choose to pull or push code with their current SSH keys, however, two team members told him that they could NOT run successfully on their own Linux machines on CodeCommit repositories. How would he solve this problem? (Choose 2)

A number of instances are running on your OpsWork stacks. If you want to install security updates, what is the AWS recommendation in accordance with this task? (Choose 2)

AWS CodeDeploy is used to manage several deployment phases, including feature testing, system integration testing, and manufacturing. Every phase has a CodeDeploy deployment group. Each group is supplemented with tags with relevant targets. Like servers containing the product, tag is deployed only if the Production Deployment Group creates a new deployment. Which targets CANNOT be added to Deployment Group by Tags in AWS CodeDeploy? (Choose 2)

The Docker Containers for a Java Application is controlled by AWS ECS within your business. Several important features have recently been developed to satisfy the market requirements and the team is under pressure to deliver new releases. When the ECS role comes into stopped status, the manager asks you to alert the team both on the slack channel and email so that the relevant staff can take immediate action. What is the best way for you to do this?

Phillip is using AWS Systems Manager to maintain EC2 instances. For example, for instances with a tag of “QA”, you can run command to execute a shellscript. However, you want to limit the utilization of the “Run Command” feature for certain IAM users on security issues. For these specific users, you need an IAM policy to only allow them to run command for instances that have the “department” tag of “dev1” or “dev2”. Which IAM policy can help you to achieve this requirement?

There is a number of CloudFormation templates. Most of them are used to manage AMI IDs for various regions and instance types with mappings. The AMI IDs can, however, change regularly; for example, when there are software updates. In that case, all related CloudFormation templates must be updated, which takes time. You plan to have an automated way to search and get the correct CloudFormation AMI IDs. Which two strategies will help you to accomplish this together? (Choose 2)

A new employee has joined AWS codecommit in several AWS region for the management of code repositories. To link to SSH’s CodeCommit, the newest version of a GIT has been installed in the Linux machine, an ssh-keygen key pair has been created and a public key has been added to its IAM users’ security identifiers. But still, no repositories using SSH can be cloned. What ~/.ssh/ config file does the Linux computer need to create?

Arthur has a large number of AWS accounts that are managed by him. Some AWS accounts belong to AWS Organization and some do not. When there is a new EC2 termination event for any other account, it is necessary to set up the CloudWatch rule in the master account. In each account, you plan to set a CloudWatch Events rule, and send the event to the master account in the default Event Bus. What is NOT the solution to work with in this requirement?

A big financial company runs a Hybrid system. Separate deployment mechanisms for the on-site and AWS servers have previously been used. The deployment management must be increased in all instances with the same method. For this necessity, you suggested using AWS CodeDeploy because all on-site machines run under Ubuntu LTS 16.04. To configure the instances on site with AWS CodeDeploy, what pre-conditions must be met? (Choose 2)

All instances in an organization have certain latest security patches that need to be installed in time. Nevertheless, it is necessary for patches to be installed in dev and test environment for one week before deployment in production instances. It is possible to distinguish all EC2 instances by tags. How can the AWS System Manager be implemented in the best way possible?

A CloudFormation stack has included several AWS resources including EC2 instances. In the past, AMI ID management for different regions and instance types was done via a mapping table. The AMI ID is obtained in the function “Fn::FindInMap” during the creation of the EC2 resource. In this method, however, if a new AMI ID is available, the mapping table must be updated. You look for better ways to get your updated AMIs automatically. You already have a Lambda feature, which can obtain the latest AMI with the input as a region and instance type. How can the Lambda function best be used in order to achieve this?

The Jenkins server was maintained by your team in an EC2 instance. The Jenkins server is used mainly to build Java-based artefacts. Currently, this Jenkins server has no Disaster Recovery Strategy. A new disaster recovery plan must be drawn up for your team, both RTO and RPO, within 24 hours. What strategy should the team choose to recover from disasters?

A Service is designed that aggregates the clickstream data in batch and deliver reports to the subscriber through the emails. The data is extremely spikey, high-scaled, geographically distributed and unpredictable. How will you design this system?