Continuous Infrastructure Delivery

Transcript

1. Continuous Infrastructure Delivery
2. Continuous delivery is an established software practice.
3. What goes into continuous delivery?
4. Automated testing.
5. Code integration.
6. Deployment.
7. Could we apply this workflow to infrastructure?
8. Xander Grzywinski Developer Advocate @
10. ⁄ Continuous delivery.
11. ⁄ Build quality in.
12. ⁄ Work in small batches.
13. ⁄ Pursue continuous improvement.
14. ⁄ Everyone is responsible.
15. ⚙ Automated testing.
17. Infrastructure as code.
18. CODE EDITOR # configure the aws provider provider "aws" { profile = "default" region = "us-east-1" } # create an instance resource "aws_instance" "example" { ami = “ami-2757f631" instance_type = "t2.micro" }
19. https://github.com/gruntwork-io/terratest
20. CODE EDITOR func TestTerraformAwsExample(t *testing.T) { t.Parallel() // Give this EC2 Instance a unique ID expectedName := fmt.Sprintf("terratest-aws-example-%s", random.UniqueId()) // At the end of the test, run `terraform destroy` to clean up any resources that were created defer terraform.Destroy(t, terraformOptions) // This will run `terraform init` and `terraform apply` and fail the test if there are any errors terraform.InitAndApply(t, terraformOptions) aws.AddTagsToResource(t, awsRegion, instanceID, map[string]string{"testing": "testing-tag-value"}) // Look up the tags for the given Instance ID instanceTags := aws.GetTagsForEc2Instance(t, awsRegion, instanceID) testingTag, containsTestingTag := instanceTags["testing"] assert.True(t, containsTestingTag) }
21. TERMINAL $ go test $
22. A note on testing infrastructure code …
23. Modularize your infrastructure.
24. CODE EDITOR # call the servers module with a value of 5 servers module "servers" { source = "./app-cluster" servers = 5 }
25. Continuous integration.
26. TERMINAL $ tflint $ go test $ terraform plan $ git push $
27. ☁ Deployment.
28. TERMINAL $ terraform apply $
29. What are some modern practices for software deployment?
30. ⚙ Testing in production. please don’t @ me …
31. ☠ Feature flags.
32. CODE EDITOR # feature flag variable to enable new ami variable "enable_new_ami" { default = false type = bool } # aws instance enabled by feature flag resource "aws_instance" "example_bionic" { count = var.enable_new_ami ? 1 : 0 instance_type = "t2.micro" ami = data.aws_ami.ubuntu_bionic.id vpc_security_group_ids = [aws_security_group.instances.id] subnet_id = aws_subnet.public.id tags = { has_toggle = var.enable_new_ami } }
33. Blue-Green deployments.
34. https://hashicorp.com/blog
35. Deployment pipeline.
37. CODE EDITOR version: 2 jobs: test: docker: - image: salaxander/terratest:latest steps: - checkout - run : | go test -v -timeout 30m workflows: version: 2 ci: jobs: - test filters: branches: only: master
38. CODE EDITOR version: 2 jobs: plan: docker: - image: hashicorp/terraform:0.12.20 steps: - checkout - run: | terraform init && terraform plan apply: docker: - image: hashicorp/terraform:0.12.20 steps: - checkout - run: | terraform init && terraform apply --auto-approve workflows: version: 2 ci: jobs: - plan - approve: type: approval requires: - plan - apply: requires: - approve filters: branches: only: master
39. ⁄ Demo time.
40. Why would we want to apply this workflow to infrastructure?
41. Thank You xander@hashicorp.com community.hashicorp.com