feat: add AMI house keeping lambda (github-aws-runners#3570)

## Description Quite common to build AMI's with packer. Cleaning up old images could be a challenge. This housekeeping lambda can remove old AMI based on a few criteria. The AMI housekeepr is implemented as an extra terraform module with a lambda that runs once a day. The housekeepr can be configured to exclude images based on reference in SSM, and tags. Images. The module is deisabled by default and can used as part of the main module, multi-runner or stand alone. This PR also updates outdated lambda dependencies. And updates terraform workflows to 1.5.6 as main version. ## Tested - [x] default example (with new housekeeper) - [x] multi runner example --------- Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
get-convex · Oct 30, 2023 · 87104e8 · 87104e8
1 parent 76f45dc
commit 87104e8
Show file tree

Hide file tree

Showing 39 changed files with 2,599 additions and 950 deletions.
diff --git a/.github/workflows/terraform.yml b/.github/workflows/terraform.yml
@@ -4,8 +4,7 @@ on:
     branches:
       - main
   pull_request:
-    paths-ignore:
-      - "modules/*/lambdas/**"
+    paths: ["**/*.tf", "**/*.hcl", ".github/workflows/terraform.yml"]
 
 permissions:
   contents: read
@@ -18,7 +17,7 @@ jobs:
     name: Verify module
     strategy:
       matrix:
-        terraform: [1.3.2, "latest"]
+        terraform: [1.5.6, "latest"]
     runs-on: ubuntu-latest
     container:
       image: hashicorp/terraform:${{ matrix.terraform }}
@@ -30,9 +29,10 @@ jobs:
           touch lambdas/functions/webhook/webhook.zip
           touch lambdas/functions/control-plane/runners.zip
           touch lambdas/functions/gh-agent-syncer/runner-binaries-syncer.zip
+          touch lambdas/functions/ami-housekeeper/ami-housekeeper.zip
       - name: terraform init
         run: terraform init -get -backend=false -input=false
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: check terraform formatting
         run: terraform fmt -recursive -check=true -write=false
       - if: contains(matrix.terraform, 'latest') # check formatting for the latest release but avoid failing the build
@@ -41,17 +41,17 @@ jobs:
         continue-on-error: true
       - name: validate terraform
         run: terraform validate
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: Fix for actions/cache on alpine
         run: apk add --no-cache tar
         continue-on-error: true
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         uses: actions/cache@v3.3.1
         name: Cache TFLint plugin dir
         with:
           path: ~/.tflint.d/plugins
           key: tflint-${{ hashFiles('.tflint.hcl') }}
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: Setup TFLint
         uses: terraform-linters/setup-tflint@v4
         with:
@@ -69,7 +69,7 @@ jobs:
       matrix:
         terraform: [1.3.2, "latest"]
         module:
-          ["download-lambda", "multi-runner", "runner-binaries-syncer", "runners", "setup-iam-permissions", "ssm", "webhook"]
+          ["ami-housekeeper", "download-lambda", "multi-runner", "runner-binaries-syncer", "runners", "setup-iam-permissions", "ssm", "webhook"]
     defaults:
       run:
         working-directory: modules/${{ matrix.module }}
@@ -116,7 +116,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        terraform: [1.3.2, "latest"]
+        terraform: [1.5.6, "latest"]
         example:
           ["default", "ubuntu", "prebuilt", "arm64", "ephemeral", "windows", "multi-runner"]
     defaults:
@@ -129,7 +129,7 @@ jobs:
       - uses: actions/checkout@8ade135a41bc03ea155e62e844d188df1ea18608
       - name: terraform init
         run: terraform init -get -backend=false -input=false
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: check terraform formatting
         run: terraform fmt -recursive -check=true -write=false
       - if: contains(matrix.terraform, 'latest') # check formatting for the latest release but avoid failing the build
@@ -138,25 +138,24 @@ jobs:
         continue-on-error: true
       - name: validate terraform
         run: terraform validate
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: Fix for actions/cache on alpine
         run: apk add --no-cache tar
         continue-on-error: true
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         uses: actions/cache@v3.3.1
         name: Cache TFLint plugin dir
         with:
           path: ~/.tflint.d/plugins
           key: tflint-${{ hashFiles('.tflint.hcl') }}
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: Setup TFLint
         uses: terraform-linters/setup-tflint@v3
         with:
           github_token: ${{ secrets.GITHUB_TOKEN }}
-      - if: contains(matrix.terraform, '1.3.')
+      - if: contains(matrix.terraform, '1.5.')
         name: Run TFLint
         working-directory: ${{ github.workspace }}
         run: |
           tflint --init -c ${GITHUB_WORKSPACE}/.tflint.hcl --chdir modules/${{ matrix.module }}
           tflint -f compact -c ${GITHUB_WORKSPACE}/.tflint.hcl --var-file ${GITHUB_WORKSPACE}/.github/lint/tflint.tfvars --chdir examples/${{ matrix.example }}
-
diff --git a/.github/workflows/update-docs.yml b/.github/workflows/update-docs.yml
@@ -2,8 +2,8 @@ name: Update docs
 on:
   push:
     paths:
-      - "*.tf"
-      - "*.md"
+      - "**/*.tf"
+      - "**/*.md"
       - ".github/workflows/update-docs.yml"
 
 permissions:

diff --git a/.vscode/gh-runners.code-workspace b/.vscode/gh-runners.code-workspace
@@ -4,6 +4,10 @@
       "name": "✨ root",
       "path": ".."
     },
+    {
+      "name": "🚀 @lambda/functions/ami-housekeeper",
+      "path": "../lambdas/functions/ami-housekeeper"
+    },
     {
       "name": "🚀 @lambda/functions/control-plane",
       "path": "../lambdas/functions/control-plane"

diff --git a/README.md b/README.md
@@ -337,7 +337,7 @@ The example for [ephemeral runners](./examples/ephemeral) is based on the [defau
 
 ### Prebuilt Images
 
-This module also allows you to run agents from a prebuilt AMI to gain faster startup times. You can find more information in [the image README.md](/images/README.md)
+This module also allows you to run agents from a prebuilt AMI to gain faster startup times. The module provides several examples to build your own custom AMI. To remove old images, an [AMI housekeeper module](./modules/ami-housekeeper/README.md) can be used. You can find more information in [the image README.md](/images/README.md) for building custom images.
 
 ### Experimental - Optional queue to publish GitHub workflow job events
 
@@ -468,6 +468,7 @@ We welcome any improvement to the standard module to make the default as secure
 
 | Name | Source | Version |
 |------|--------|---------|
+| <a name="module_ami_housekeeper"></a> [ami\_housekeeper](#module\_ami\_housekeeper) | ./modules/ami-housekeeper | n/a |
 | <a name="module_runner_binaries"></a> [runner\_binaries](#module\_runner\_binaries) | ./modules/runner-binaries-syncer | n/a |
 | <a name="module_runners"></a> [runners](#module\_runners) | ./modules/runners | n/a |
 | <a name="module_ssm"></a> [ssm](#module\_ssm) | ./modules/ssm | n/a |
@@ -491,6 +492,12 @@ We welcome any improvement to the standard module to make the default as secure
 | Name | Description | Type | Default | Required |
 |------|-------------|------|---------|:--------:|
 | <a name="input_ami_filter"></a> [ami\_filter](#input\_ami\_filter) | Map of lists used to create the AMI filter for the action runner AMI. | `map(list(string))` | <pre>{<br>  "state": [<br>    "available"<br>  ]<br>}</pre> | no |
+| <a name="input_ami_housekeeper_cleanup_config"></a> [ami\_housekeeper\_cleanup\_config](#input\_ami\_housekeeper\_cleanup\_config) | Configuration for AMI cleanup.<br><br>    `amiFilters` - Filters to use when searching for AMIs to cleanup. Default filter for images owned by the account and that are available.<br>    `dryRun` - If true, no AMIs will be deregistered. Default false.<br>    `launchTemplateNames` - Launch template names to use when searching for AMIs to cleanup. Default no launch templates.<br>    `maxItems` - The maximum numer of AMI's tha will be queried for cleanup. Default no maximum.<br>    `minimumDaysOld` - Minimum number of days old an AMI must be to be considered for cleanup. Default 30.<br>    `ssmParameterNames` - SSM parameter names to use when searching for AMIs to cleanup. This parameter should be set when using SSM to configure the AMI to use. Default no SSM parameters. | <pre>object({<br>    amiFilters = optional(list(object({<br>      Name   = string<br>      Values = list(string)<br>      })),<br>      [{<br>        Name : "state",<br>        Values : ["available"],<br>        },<br>        {<br>          Name : "image-type",<br>          Values : ["machine"],<br>      }]<br>    )<br>    dryRun              = optional(bool, false)<br>    launchTemplateNames = optional(list(string))<br>    maxItems            = optional(number)<br>    minimumDaysOld      = optional(number, 30)<br>    ssmParameterNames   = optional(list(string))<br>  })</pre> | `{}` | no |
+| <a name="input_ami_housekeeper_lambda_s3_key"></a> [ami\_housekeeper\_lambda\_s3\_key](#input\_ami\_housekeeper\_lambda\_s3\_key) | S3 key for syncer lambda function. Required if using S3 bucket to specify lambdas. | `string` | `null` | no |
+| <a name="input_ami_housekeeper_lambda_s3_object_version"></a> [ami\_housekeeper\_lambda\_s3\_object\_version](#input\_ami\_housekeeper\_lambda\_s3\_object\_version) | S3 object version for syncer lambda function. Useful if S3 versioning is enabled on source bucket. | `string` | `null` | no |
+| <a name="input_ami_housekeeper_lambda_schedule_expression"></a> [ami\_housekeeper\_lambda\_schedule\_expression](#input\_ami\_housekeeper\_lambda\_schedule\_expression) | Scheduler expression for action runner binary syncer. | `string` | `"rate(1 day)"` | no |
+| <a name="input_ami_housekeeper_lambda_timeout"></a> [ami\_housekeeper\_lambda\_timeout](#input\_ami\_housekeeper\_lambda\_timeout) | Time out of the lambda in seconds. | `number` | `300` | no |
+| <a name="input_ami_housekeeper_lambda_zip"></a> [ami\_housekeeper\_lambda\_zip](#input\_ami\_housekeeper\_lambda\_zip) | File location of the lambda zip file. | `string` | `null` | no |
 | <a name="input_ami_id_ssm_parameter_name"></a> [ami\_id\_ssm\_parameter\_name](#input\_ami\_id\_ssm\_parameter\_name) | Externally managed SSM parameter (of data type aws:ec2:image) that contains the AMI ID to launch runner instances from. Overrides ami\_filter | `string` | `null` | no |
 | <a name="input_ami_kms_key_arn"></a> [ami\_kms\_key\_arn](#input\_ami\_kms\_key\_arn) | Optional CMK Key ARN to be used to launch an instance from a shared encrypted AMI | `string` | `null` | no |
 | <a name="input_ami_owners"></a> [ami\_owners](#input\_ami\_owners) | The list of owners used to select the AMI of action runner instances. | `list(string)` | <pre>[<br>  "amazon"<br>]</pre> | no |
@@ -501,6 +508,7 @@ We welcome any improvement to the standard module to make the default as secure
 | <a name="input_create_service_linked_role_spot"></a> [create\_service\_linked\_role\_spot](#input\_create\_service\_linked\_role\_spot) | (optional) create the service linked role for spot instances that is required by the scale-up lambda. | `bool` | `false` | no |
 | <a name="input_delay_webhook_event"></a> [delay\_webhook\_event](#input\_delay\_webhook\_event) | The number of seconds the event accepted by the webhook is invisible on the queue before the scale up lambda will receive the event. | `number` | `30` | no |
 | <a name="input_disable_runner_autoupdate"></a> [disable\_runner\_autoupdate](#input\_disable\_runner\_autoupdate) | Disable the auto update of the github runner agent. Be aware there is a grace period of 30 days, see also the [GitHub article](https://github.blog/changelog/2022-02-01-github-actions-self-hosted-runners-can-now-disable-automatic-updates/) | `bool` | `false` | no |
+| <a name="input_enable_ami_housekeeper"></a> [enable\_ami\_housekeeper](#input\_enable\_ami\_housekeeper) | Option to disable the lambda to clean up old AMIs. | `bool` | `false` | no |
 | <a name="input_enable_cloudwatch_agent"></a> [enable\_cloudwatch\_agent](#input\_enable\_cloudwatch\_agent) | Enables the cloudwatch agent on the ec2 runner instances. The runner uses a default config that can be overridden via `cloudwatch_config`. | `bool` | `true` | no |
 | <a name="input_enable_ephemeral_runners"></a> [enable\_ephemeral\_runners](#input\_enable\_ephemeral\_runners) | Enable ephemeral runners, runners will only be used once. | `bool` | `false` | no |
 | <a name="input_enable_event_rule_binaries_syncer"></a> [enable\_event\_rule\_binaries\_syncer](#input\_enable\_event\_rule\_binaries\_syncer) | Option to disable EventBridge Lambda trigger for the binary syncer, useful to stop automatic updates of binary distribution. | `bool` | `true` | no |

diff --git a/examples/default/main.tf b/examples/default/main.tf
@@ -96,6 +96,19 @@ module "runners" {
 
   # Enable debug logging for the lambda functions
   # log_level = "debug"
+
+  enable_ami_housekeeper = true
+  ami_housekeeper_cleanup_config = {
+    ssmParameterNames = ["*/ami-id"]
+    minimumDaysOld    = 10
+    amiFilters = [
+      {
+        Name   = "name"
+        Values = ["*al2023*"]
+      }
+    ]
+  }
+
 }
 
 module "webhook_github_app" {

diff --git a/images/README.md b/images/README.md
@@ -6,6 +6,8 @@ These images share the same scripting as used in the user-data mechanism in `/mo
 
 The examples in `linux-al2023` and `windows-core-2019` also upload a `start-runner` script that uses the exact same startup process as used in the user-data mechanism. This means that the image created here does not need any extra scripts injected or changes to boot up and connect to GH.
 
+To remove old images the [AMI house keeper module](./modules/ami-housekeeper/README.md) can be used.
+
 ## Building your own
 
 To build these images you first need to install packer.

diff --git a/lambdas/functions/ami-housekeeper/jest.config.ts b/lambdas/functions/ami-housekeeper/jest.config.ts
@@ -0,0 +1,17 @@
+import type { Config } from 'jest';
+
+import defaultConfig from '../../jest.base.config';
+
+const config: Config = {
+  ...defaultConfig,
+  coverageThreshold: {
+    global: {
+      statements: 100,
+      branches: 100,
+      functions: 100,
+      lines: 100,
+    },
+  },
+};
+
+export default config;
diff --git a/lambdas/functions/ami-housekeeper/package.json b/lambdas/functions/ami-housekeeper/package.json
@@ -0,0 +1,48 @@
+{
+  "name": "@terraform-aws-github-runner/ami-housekeeper",
+  "version": "1.0.0",
+  "main": "lambda.ts",
+  "license": "MIT",
+  "scripts": {
+    "start": "ts-node-dev src/local.ts",
+    "test": "NODE_ENV=test jest",
+    "test:watch": "NODE_ENV=test jest --watch",
+    "lint": "yarn eslint src",
+    "watch": "ts-node-dev --respawn --exit-child src/local.ts",
+    "build": "ncc build src/lambda.ts -o dist",
+    "dist": "yarn build && cd dist && zip ../ami-housekeeper.zip index.js",
+    "format": "prettier --write \"**/*.ts\"",
+    "format-check": "prettier --check \"**/*.ts\"",
+    "all": "yarn build && yarn format && yarn lint && yarn test"
+  },
+  "devDependencies": {
+    "@trivago/prettier-plugin-sort-imports": "^4.2.1",
+    "@types/aws-lambda": "^8.10.125",
+    "@types/jest": "^29.5.6",
+    "@types/node": "^20.8.9",
+    "@typescript-eslint/eslint-plugin": "^6.9.0",
+    "@typescript-eslint/parser": "^6.9.0",
+    "@vercel/ncc": "^0.38.1",
+    "aws-sdk-client-mock": "^3.0.0",
+    "aws-sdk-client-mock-jest": "^3.0.0",
+    "eslint": "^8.52.0",
+    "eslint-plugin-prettier": "5.0.1",
+    "jest": "^29.5",
+    "jest-mock": "^29.5.0",
+    "jest-mock-extended": "^3.0.4",
+    "nock": "^13.3.6",
+    "prettier": "3.0.3",
+    "ts-jest": "^29.1.0",
+    "ts-node": "^10.9.1",
+    "ts-node-dev": "^2.0.0"
+  },
+  "dependencies": {
+    "@aws-sdk/client-ec2": "^3.436.0",
+    "@aws-sdk/client-ssm": "^3.436.0",
+    "@aws-sdk/types": "^3.433.0",
+    "@terraform-aws-github-runner/aws-powertools-util": "*",
+    "@terraform-aws-github-runner/aws-ssm-util": "*",
+    "cron-parser": "^4.8.1",
+    "typescript": "^5.0.4"
+  }
+}