Skip to content

Latest commit

 

History

History
executable file
·
376 lines (311 loc) · 28.8 KB

aws-quiz.md

File metadata and controls

executable file
·
376 lines (311 loc) · 28.8 KB
Raw

Amazon Web Services (AWS)

Q1. You have an application using a 100 GB MySQL database that you are migrating into AWS. What should you consider when deciding between whether to host the database on RDS for MySQL or Aurora?

  • cost
  • ease of maintenance vs. granularity of control
  • all of these answers
  • the current storage engine used by the application, such as InnoDB or MyISAM

Q2. Which database is a NoSQL database type that can quickly store and retrieve key-value pairs?

  • Aurora
  • Neptune
  • RDS for MySQL
  • DynamoDB

Q3. Your database is an RDS instance running SQL Server with Multi-AZ replication and you have several older .NET console utilities that perform database operations every 15 seconds. When the cluster has to switch the primary database server to the secondary AZ, the .NET utilities start to report connection failures to the database although other applications are able to access the database. How do you correct this problem?

  • Use the RDS console to force a reboot of the database instance so that the primary server becomes the master server again.
  • The server running the .NET utilities is caching the DNS lookup on the database cluster address. Flush the DNS cache of the server and force the C# utilities to open new connections to the database.
  • A.NET application will retain the IP address of a connection string until the host machine is rebooted.
  • The NET utilities need to change the SQL Server endpoint in the connection strings to read from the secondary database server using a try/catch.

Q4. What AWS services can help you automate your development pipeline for continuous integration and continuous deployment?

  • CodePipeline
  • CodeDeploy
  • all of these answers
  • CodeBuild

Q5. Which AWS service complies with the standards outlined in Payment Card Industry Data Security Standard (PCI DSS) Level 1 for the handling and transmission of credit card data?

  • API Gateway
  • all of these answers
  • Simple Queue Service (SQS)
  • Kinesis Data Streams

Q6. You have a large amount of files on your network-attached storage array that must be archived and maintained for a period of 10 years due to industry regulations. This data will be infrequently accessed but must be kept What is the best AWS service for storing this data?

  • EFS
  • Snowball
  • OEBS
  • S3 Glacier

Q7. For your AWS root account, you have generated a random password of the maximum allowed length and included special characters. Which additional steps should you take to secure your AWS root account?

  • Create an AM role for the account administrator with the highest privileges. Do not store the root password, but when the root account is needed reset the password on the root account via email confirmation and repeat this procedure.
  • Store your randomly generated password in your organizational secrets database using a service such as 1Password or LastPass, and only grant access to this secret to the DevOps team.
  • Create IAM accounts for your administrators and attach the AdministratorAccess policy to their accounts. Disable the root account in the user settings.
  • Create an IAM role for the account administrator with the highest privileges and do not use the root account in day-today operations. Enable two-factor authentication on the root account

Q8. Which Elastic Load Balancing option supports Lambda as a target?

  • Network Load Balancer
  • Lambda can not be called directly by incoming web requests. You must use API Gateway.
  • Classic Load Balancer
  • Application Load Balancer

Q9. How do you architect a solution for an SQL Server database to be replicated across AWS regions in an active-active architecture?

  • Use RDS for SQL Server and create the same instance in two different regions. Use Database Migration Service to keep each database in sync.
  • Use a VPN or VPC peering to establish a connection between the VPCs in each region. Install SQL Server Enterprise Edition on EC2 instances in each region and configure an Always On availability group.
  • Use RDS for SQL Server 2016 or 2017 Enterprise Edition. Enable Multi-AZ support and select the Mirroring/Always On option. Select another region for the mirroring option.
  • You can not set up an active-active architecture for SQL Server that spans geographic regions.

Reference

Q10. What does it cost to launch an EC2 instance from the AWS Marketplace?

  • All images in the AWS Marketplace incur additional hourly fees in addition to the charges from the instance size you select.
  • You can only launch images that were created by other users on your AWS account, so you pay only for the instance size you select and the S3 storage costs for the base image.
  • Each image has its own pricing that could either be free, or include charges for software licensing costs. You will also pay for the instance the image runs on
  • All images in the AWS Marketplace contain only open-source software with no additional fees and are created by other AWS users. You will pay only for the instance size you select.

Q11. When using an ECS cluster with EC2 instances, what maintenance tasks should you perform on the EC2s?

  • The instances created by ECS do not have patches that need to be applied; however, you should make sure your containers contain any important security updates.
  • Refresh the cluster with instances built from the latest ECS AMI.
  • ECS clusters do not use EC2 instances.
  • You should not directly manipulate the EC2 instances created by ECS. AWS will automatically update these instances.

Q12. What in-memory caching server is not supported by ElastiCache?

  • Redis 5
  • Memcached
  • Elasticsearch
  • Redis 3

Q13. Which AWS service can be used to help generate the documentation required by various compliance standards, such as Payment Card Industry Data Security Standard (PCI DSS) Level 1 for the handling of credit card data?

  • Artifact
  • DocumentDB
  • Print out the AWS Compliance summary and keep it with your required documentation for an audit.
  • Secrets Manager

Q14. When using AWS for research and development ahead of a planned migration, how do you prevent unexpected increases or spikes in the billing?

  • Use the billing dashboard to create a cost budget. Input the max amount you want to be charged each month. Any charges that occur over this amount will cause AWS to automatically suspend those resources
  • Using the root AWS account, activate IAM access to the billing information for the account. Make sure your IAM users have the Billing FullAccessGroup policy. Then from the billing dashboard, check the accrued charges once a day.
  • If you are using the AWS free tier, you will have to confirm the usage of any service that goes over the AWS free tier limits.
  • Using the root AWS account enable Billing Alerts in the user preferences. Then use CloudWatch to create a billing alarm and set a threshold to a specific dollar amount for your estimated monthly charges.

Q15. You are creating a DynamoDB table to store all movies that have been released since 1938. Your application will allow users to search by movie title and see the details of that film. Given the sample below showing the movie data that you will be importing, what is the best set of keys to apply to this table?

{
    "title":"The Avengers",
    "year": 2012,
    "cast" : ["Mark Ruffalo", "Robert Downey, Jr."],
    "genres" : ["Action"]
}
  • The primary key should be a partition key of the title field.
  • The primary key should be the title field and the partition key should be the genres field.
  • The primary key should be a composite key comprised of a partition key on the title field and a sort key on the year field.
  • The primary key should be created as a completely unique value, such a sequential numerical list of movie IDs. The partition key should be title field for fast lookup.

Q16. What data store provides a simple and quick way of storing basic user attributes in an object-based format?

  • ORDS for Oracle
  • Redshift
  • Neptune
  • DynamoDB

Q17. You need a schemaless database. Which Amazon database service provides that solution?

  • ORDS
  • Aurora
  • Redshift
  • DynamoDB

Q18. Which communication channel does SNS not support natively?

  • OSMS text message
  • push notification
  • email
  • automated phone call

Q19. When designing a serverless web application using Lambda, what key concept must you factor into your design?

  • Serverless web applications run within the web browser of the user, so you will need to store any data the user changes directly in a database.
  • Lambda only allows you to write functions in JavaScript.
  • Lambda does not use servers, so it can only return the same request to every user.
  • Lambda is stateless, so it won't remember who a user is in between requests.

Q20. A principle of DevOps is to view infrastructure as code. Which AWS service allows you to script your AWS infrastructure?

  • CloudTrail
  • CloudFormation
  • AWS Config
  • AWS Service Catalog

Q21. You created a Windows EC2 instance with a public IP address and installed SQL Server. When attempting to connect to SQL Server from SQL Server Enterprise Manager on your local computer, the Windows EC2 instance is unable to establish a connection to the server. What is the first thing you should check?

  • Check the routing tables for the VPC.
  • Verify that the assigned security groups allow TCP port 1433 traffic from your current IP address.
  • Check the policies within Windows Firewall.
  • Verify that you are connecting to the instance using a user that is not sa.

Q22. You are hosting an application configured to stream media to its clients on TCP ports 3380-3384, 3386-3388, and 3390. The Inbound tab below shows three incoming security group policies attached to this instance. Which policy should you use?

image

  • The rule that exposes TCP ports 3380-3390 would also publicly expose port 3389 (RDP) to the entire internet. Write separate rules to only expose the needed ports.
  • The first security group rule allows all traffic into this instance. Exposing your entire instance to the whole internet leaves the server open to various attacks of the other services running on different port numbers.
  • Verify that the AWS account owners actually control the entire CIDR C block for 12.228.11.0-255 and these are secured IPs for RDP access into this instance.
  • There are no recommendations to make.

Q23. You have four front-end web servers behind a load balancer, which use NFS to access another EC2 instance that resizes and stores images for the front-end application. What security group policies should be assigned to these servers?

  • Assign Elastic IPs to all of the instances and create a group that allows all traffic to pass between each of the five Elastic IP addresses and allow all inbound HTTPS traffic.
  • Front-end web servers should allow HTTPS. Assign another group to all of the instances that allows all traffic to pass between instances using that group.
  • Create a security group that allows inbound NFS, HTTP, and HTTPS traffic from all IP addresses. Apply this group to all of the servers.
  • Create a security group that allows inbound HTTP and HTTPS traffic from all IP addresses and apply this to the web servers. Create a second security group for the NFS filestore that allows outbound NFS traffic to the private IP range of the front-end web servers.

Q24. You have a Linux EC2 web server that suddenly is timing out on all HTTP requests and your SSH connection attempts are timing out. You notice that it is failing the system status check in the EC2 console. What action should you take?

  • Restore the instance from the last AMI image. System status checks indicate that the filesystem on the instance is corrupted.
  • Stop and start the instance. This will move the instance to another host.
  • Contact AWS support. Failing a system status check indicates a failure in the underlying hardware and must be addressed by an AWS representative.
  • Reboot the instance. This will stop and start the instance and move it to another host.

Q25. You have several on-premise servers and would like to store your offsite backups on AWS. What fully managed backup service can you use to ship your backups to AWS?

  • Windows Server 2016 supports S3 as a target when using storage replicas.
  • Use Storage Gateway.
  • Sync files directly to S3 with the AWS CLI.
  • Use the RDS console to force a reboot of the database instance so that the primary server becomes the master server again.

Q26. What is the best practice for creating a highly available PostgreSQL database in RDS that can sustain the loss of a single AWS region?

  • PostgreSQL cannot be replicated across regions. Restore the database backups from an S3 bucket and repoint your database connections to the new instance.
  • Create Read Replicas in other AWS regions. You can designate a new master database from any of the read replicas until the regional failure is resolved.
  • Verify that your instance is configured for Multi-AZ support. Database changes will be automatically synced to another region in the event of a failure and RDS will automatically select a new master until the regional failure is resolved.
  • Create Read Replicas in other AWS regions. Ensure read operations against the database occur on an available Read Replica, and send write operations to another region if you need to promote a Read Replica to a standalone database if the master is down.

Q27. You created a new Linux EC2 instance and installed PostgreSQL but you are not able to establish a connection to the server from your local computer. What steps do you take to resolve this issue?

  • Create a security group rule that allows all traffic from 0.0.0.0/0. This will verify whether or not another rule is denying the traffic.
  • Verify that the assigned security groups allow traffic from your IP address to port 5432. Verify that PostgreSQL is configured to listen to external traffic and is bound to the public interface.
  • Make sure that you are using an Elastic IP and that it is included within the postgresql.conf configuration file.
  • Stop and start the instance. New security group rules will only take effect after a restart.

Q28. What does the statement body of this S3 bucket policy do?

{
  "Sid": "bucketpolicy1",
  "Effect": "Allow",
  "Principal": "*",
  "Action": "s3:GetObject",
  "Resource": "arn:aws:s3:::userreports/*",
  "Condition": {
    "IpAddress": {"aws:SourceIp": "68.249.108.0/24"},
    "NotIpAddress": {"aws:SourceIp": "68.249.108.128/32"}
  }
}
  • bucketpolicy1 allows any user to perform any action on the objects in the userreports bucket, but limits the objects to read-only permissions for anyone coming from 68.249.108.0 to 68.249.108.255 - except 68.249.108.128.
  • bucketpolicy1 allows any user coming from the IP range of 68.249.108.0 to access objects in the userreports bucket and denies access to 68.249.108.128.
  • bucketpolicy1 allows any user to perform any action on the objects in the userreports bucket - except anyone coming from the IP of 68.249.108.128.
  • bucketpolicy1 allows any user coming from the IP range of 68.249.108.0 to 68.249.108.255 to access objects in the userreports bucket-except anyone coming from the IP of 68.249.108.128.

Q29. A new developer has been added to the team and you have been asked to provide access to the organization's AWS account. What is the best practice for granting access?

  • Give the new developer the IAM login that is assigned to the development team. This IAM user should already include all of the policies that a developer would need.
  • Create a IAM user for the new developer. Manually assign policies to the new IAM user account.
  • Do not give the new developer access to the AWS console. Using the IAM user that is assigned to the development group, generate a new set of access keys and label these with the name of the developer.
  • Create a IAM user for the new developer. Assign the new developer the a developer group you already created for the other developers.

Q30. When launching an EC2 instance with an instance type that supports instance storage, what use case is best for instance storage?

  • Use the instance storage to serve temporary files that require low I/O latency.
  • Use the instance storage to handle files uploaded by your users. Since it is more secure than an EBS volume, you can isolate any malicious files from infecting your server.
  • Instance storage is faster than EBS volumes, so install the root of the operating system on this volume to speed up server performance.
  • Instance storage is a deprecated option for storage and should not be used.

Q31. What's the best practice for horizontally scaling a legacy ASP.NET web application that relies on Active Directory and is currently deployed to a single Windows EC2 instance?

  • Use Sysprep to shut down the instance during a maintenance window. Create an AMI image and place both servers behind Application Load Balancer with sticky sessions.
  • Launch a new EC2 with the latest version of Windows Server and install the application again. Use Application Load Balancer and sticky sessions to balance between both servers.
  • Create a clone of the server using an AMI image and user Application Load Balancer to balance the traffic between both instances using sticky sessions.
  • Horizontal scaling is not the best practice in this situation. Increase the size of the existing EC2 instance and vertically scale the application.

Q32. What does this small section of a CloudFormation template do?

FlowLog:
        Type: AWS::EC2::FlowLog
        Properties:
                DeliverLogsPermissionArn: !GetAtt IamRole.Arn
                LogGroupName: FlowLogsGroup
                ResourceId: !Ref LogVpcId
                ResourceType: VPC
                TrafficType: ALL
  • It writes the VPC network flow logs to the CloudWatch FlowLogsGroup log group. You could use this to inspect the network connections of your VPC.
  • It logs all of the network traffic within a VPC except Instance IDs defined by LogVpcID and logs it to the CloudWatch FlowLogsGroup log group.
  • It logs all the network traffic going to and from a single EC2 instance into the CloudWatch FlowLogsGroup log group. You could use this to inspect suspicious network traffic coming into an EC2 instance.
  • It logs all of the DNS requests made by resources within a VPC and logs them to the CloudWatch FlowLogsGroup. Use this to diagnose DNS lookup errors within your environment.

Q33. You are running Docker containers on ECS. What is the most important metric to monitor?

  • The running container count for each service from within CloudWatch.
  • The instance health of each EC2 instance in your cluster from within CloudWatch.
  • Monitor the EC2 service dashboard. Watch for posted outages to the ECS service.
  • The memory consumption of each EC2 instance in your cluster from within CloudWatch.

Q34. Application Load Balancer can route traffic to several different target groups based upon several conditions. Which of these use cases is not supported by Application Load Balancer?

  • A request with a HTTP header of X-Requested-With: staging can be routed to a target group for an ECS service in your staging environment.
  • Source IPs matching 192.0.2.0/24 on a listener port of 1433 can be routed to a target group for an RDS for SQL Server cluster.
  • A path of /signup* can be routed to a target group for a Lambda function that processes new user registrations.
  • An Http POST query string of ? action=createuser can be routed to a target group for an ECS service.

Q35. What does a VPC do?

  • creates a cloud-based network to interconnect a set of virtual servers and appliances
  • creates a secure tunnel between two networks
  • creates a shared storage plane for application data to be shared across multiple instances.
  • creates a private network that is completely isolated from the public internet.

Q36. Can you lose the public IP address associated with your EC2 instance?

  • Yes, you can lose it if you reboot the instance.
  • Yes, you can lose it if you stop and start the instance.
  • No, you will never lose the public IP address for your instance.
  • Yes, you can lose it when you edit the instance properties and release the IP address.

Q37. Where is the best place to store database backups on an EC2 instance that is configured as a database server?

  • an S3 bucket, synced with the database backups via a script that calls the AWS CLI
  • EBS volume attached to the instance
  • instance attached to the instance
  • instance storage, with a script that replicates the database backups to another instance in a different availability zone.

Q38. Which of these is a valid restriction on the properties of a VPC?

  • You can have only 10 internet gateways per region on a new AWS account.
  • You can have only 10 VPCs per region on a new AWS account
  • You cannot create a CIDR block with a netmask larger than /16
  • You can have only 10 subnets within a VPC

Q39. You have a Linux EC2 instance that is not responding to requests and you can not connect to it via SSH. Using the EC2 console, you issued a command to stop the instance, but for the past 10 minutes the instance has been in the "stopping" state. What is the next step you should take?

  • Issue another stop action via the EC2 console, and choose the option to forcefully stop the instance.
  • Create an AMI image of the instance, and choose the option to take the image without restarting the instance.
  • Edit the instance properties and increase the instance size.
  • Contact AWS support. Any further actions could corrupt the file system.

Q40. You have 14 on-premise web servers, 4 database servers, 6 servers using GIS software, 3 file servers, and 4 development servers. What considerations should you take into account when migrating these servers into AWS?

  • AWS does not have a way to separate billing for compute costs, so you will need to design a way to split the budget between departments.
  • New AWS accounts are limited to 20 on-demand EC2 instances. Submit a request to increase your rate limits before starting a migration.

Q41. As your web application grows and your application monitoring needs become more complex, which additional log monitoring service should you NOT consider?

  • ELK stack: Elasticsearch, Loggly, and Kibana
  • PRTG
  • New Relic
  • Datadog

Q42. You have a T2 EC2 instance that is critical to your infrastructure. How would you monitor the most important metric for this instance?

  • Turn on CloudWatch Auto Recovery and put monitors on the System Status and Instance Status checks for the instance to notify you when either is in alarm.
  • Use CloudWatch to put monitors on the remaining CPU credits. If you run out of CPU credit the instance will be stopped.

Q43. Which feature can be used to respond to a sudden increase in web traffic?

  • EC2 Auto Scaling groups
  • AWS Shield Advanced
  • RDS Read Replicas
  • all of these answers

Q44. If a set of servers are located within a private subnet of your VPC, how can you connect those servers to on-premise servers?

  • Establish a connection with AWS Direct Connect.
  • Use the AWS Client VPN.
  • Install a OpenVPN server on an instance that is located within the subnet with an elastic IP.
  • All of these options can establish a connection to a private subnet.

Q45. You have a UDP load balancer that is created by an instance that is running an NGINX proxy. Your application performance management (APM) solution can detect failures in your load balancer instance and transfer the Elastic IP to a passive standby instance. Using the AWS CLI, which script do you program into your APM to move the Elastic IP?

  • A
aws ec2 disassociate-address --association-id eipassoc-2bebb712
aws ec2 associate-address --instance-id i-8b953 --allocation-id eipalloc-02d021a
  • B
aws ec2 release-address --association-id eipassoc-2bebb712
aws ec2 assign-address --instance-id i-8b953 --allocation-id eipalloc-02d021a
  • C
aws ec2 stop-instances --instance-ids i-8b953
wait 30
aws ec2 disassociate-address --association-id eipassoc-2bebb712
aws ec2 associate-address --instance-id i-8b953 --allocation-id eipalloc-02d021a
aws ec2 start-instances --instance-ids i-8b953
  • D
aws ec2 release-address --association-id eipassoc-2bebb712
aws ec2 associate-address --instance-id i-8b953 --allocation-id eipalloc-02d021a

Q46. What service can host your Docker containers?

  • Lightsail
  • Elastic Container Service (ECS)
  • Elastic Compute Cloud (EC2)
  • All of these services can host a Docker container.

Q47. In the S3 console, underneath the Access column, what does the public badge next to the bucket name indicate?

image

  • All objects within this bucket are assigned public access and could be readable or writable by anyone on the internet. Ensure no sensitive data is being publicly shared within this bucket.
  • All objects within this bucket are writable, which means that the public internet has the ability to upload any file directly to your S3 bucket. Your S3 bucket could be used to serve malware.
  • Some objects within this bucket are assigned public access. Verify that any publicly shared objects within this bucket contain no sensitive data.
  • Objects within this bucket can be made public, if the ACL on that object is set to allow everyone access. Private buckets do not allow you to set public permissions on any object.

Q48. What privilege is specific to the AWS root account, and cannot be granted to another IAM user on the account?

  • Revoke the AdministratorAccess role or grant it to another IAM user.
  • Create a new hosted zone in Route 53.
  • Delete the AWS account.
  • Modify the billing details.

Q49. Your application is sending 50,000 emails through SES each day. Since you must maintain a low bounce rate to avoid being put on probation, what simple system do you architect to automatically process hard bounces?

  • Configure SES to send all bounce events to an SNS topic. Create a Lambda function that processes each hard bounce event and automatically flags that account as a bounce in your application to prevent further sending attempts.
  • Configure SES to no longer send to email addresses that are on your bounce list.
  • Configure SES to send the logs of all delivery attempts through Kinesis Firehose. Process each event and look for bounce types and remove these emails from your list.
  • Send all emails through SES with a custom reply-to header. Configure SES to listen for events on this email address and flag any email address that replies to this account as a bounced message and remove it from your email list.

Q50. Your web application is getting a suspicious amount of bad requests from foreign IP addresses. Your business is operating in only a few countries and you would like to block any other traffic. What is the best practice for limiting access to your web application by country?

  • Use Web Application Firewall and create a geo match condition to drop all requests from countries that aren't on your allow list.
  • Use Application Load Balancer to create a new routing rule that looks at source IP address. Add an IP block for the countries that have access.
  • Host the front end of your website in CloudFront and configure a geo restriction on the distribution.
  • Use CloudTrail to monitor the IP addresses of the bad requests. Use Lambda to add these IP addresses to an Application Load Balancer rule that blocks the IPs.

Q51. What is the best practice for maintaining Windows EC2 instances and applying updates?

  • Turn on auto update in Windows Update on each EC2 that is launched, or create your own AMI with this feature enabled and launch all of your EC2 instances from this AMI.
  • Create a maintenance schedule that an employee must fill out each week confirming a visual inspection of each instance was conducted and which patches were applied.
  • Use AWS Systems Manager Patch Manager to find an patch instances that require updates during a set maintenance window.
  • Install Window Server Update Services on your primary Active Directory controller.

Q52. In addition to CloudFormation, you can use other orchestration tools to automate server formation and maintenance. Which tool is not an efficient choice for the orchestration of a large infrastructure?

  • Chef
  • Ansible
  • Puppet
  • Vagrant