Resilient Architectures - CloudFormation

Question 1

Characteristics 1

Answer 1

• Lets you model, provision, and manage AWS and third-party resources by treating the infrastructure as code
• Change sets allows to check how any change might impact running resources
• Supports source control
• It’s extensible by using AWS CloudFormation Registry: where APN Partners publish third-party resources and modules (i.e. MongoDB, JFrog, Splunk, Gremlin)

Question 2

Characteristics 2

Answer 2

• Don’t need to specify the order of which resources are created, updated, or deleted. CloudFormation determines the correct sequence of actions to take for each resource
• CloudFormation Drift Detection allows to detect changes made to resources outside of AWS
• CloudFormation Templates only checks property values set by stack templates or template parameters, doesn’t check for default property values

Question 3

Stack

Answer 3

• It’s a collection of resources that you create, update, and delete as a single unit
• Can manually manage individual resources in a stack
• Stack output: key-value pair that CloudFormation makes available to other stacks and via the “aws cloudformation describe-stacks” CLI command

Question 4

Template sections

Answer 4

• Format version (optional): format of version
• Description (optional): documentation about the template
• Parameters (optional): to provide values at runtime, including their types
• Resources (mandatory): AWS resources to be created (name, type, properties, reference to parameters, and other configurations)
• Output (optional): output of the execution of the template (name, value and description of each output parameter)

Question 5

Multiple templates advantages

Answer 5

• Different teams manage different resources
• Resources have different lifecycles
• Distributing resources across different stacks makes them easier to manage

Question 6

Auto-scaling tasks

Answer 6

• Launch a certain number of instances into the Auto Scaling group
• Add the instances to a specific target group
• Terminate and recreate unhealthy instances
• Scale in or out based on average resource usage