feat(logging): add log_class parameter to runner log files configuration

[Brend-Smits]

This pull request updates the logging configuration by introducing support for the log_class property, allowing log groups to be created with either the STANDARD or INFREQUENT_ACCESS class. The change is applied throughout the configuration to ensure log groups and log files can specify their class, defaulting to STANDARD if not set.

Logging configuration enhancements:

• Added a log_class property (defaulting to "STANDARD") to the runner_log_files and multi_runner_config variables in variables.tf, modules/runners/variables.tf, and modules/multi-runner/variables.tf to allow specifying the log group class. [1] [2] [3]
• Updated the local log file definitions in modules/runners/logging.tf to include the log_class property for each log file, defaulting to "STANDARD".
• Modified the CloudWatch log group resource in modules/runners/logging.tf to use the specified log_class when creating log groups, and refactored the logic to group log files by both name and class.

Documentation improvements:

• Enhanced the description of the runner_log_files variable to document the new log_class property and its valid values.

[github-actions]

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None

[npalm]

@Brend-Smits did you run a quick check on the multi-runner as well default example?

Please can you also update all other log groups managed by the module? See https://github.com/github-aws-runners/terraform-aws-github-runner/search?q=repo%3Agithub-aws-runners%2Fterraform-aws-github-runner%20path%3A*.tf%20resource%20%22aws_cloudwatch_log_group%22&type=code

[Brend-Smits]

@Brend-Smits did you run a quick check on the multi-runner as well default example?

Please can you also update all other log groups managed by the module? See github-aws-runners/terraform-aws-github-runner/search?type=code (repo:github-aws-runners/terraform-aws-github-runner path:*.tf resource "aws_cloudwatch_log_group")

Yes I tested this with the multi runner example and it works as advertised. As expected, when you change the log class for existing log groups, it forces a replacement on the whole group.

I also updated it for all the other modules.

[Copilot]

Pull request overview

This PR adds a log_class parameter throughout the Terraform module, enabling users to configure CloudWatch log groups to use either the STANDARD or INFREQUENT_ACCESS class (defaulting to STANDARD). This affects lambda function log groups, runner EC2 instance log groups, and per-log-file CloudWatch log groups configured via runner_log_files.

Changes:

• New log_class variable added to all relevant modules (root, runners, multi-runner, webhook, ami-housekeeper, runner-binaries-syncer) with validation, and propagated through all module call chains
• Runner EC2 instance log groups refactored from count-based to for_each-based resource indexing (to support per-log-group class configuration), and each default log file entry now hardcodes log_class = "STANDARD"
• runner_log_files and multi_runner_config.runner_log_files type objects updated with an optional log_class field, and all CloudWatch log group resources updated with log_group_class = var.log_class (or equivalent)

Reviewed changes

Copilot reviewed 41 out of 41 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
variables.tf	Adds root-level log_class variable with validation; adds log_class to runner_log_files object type
main.tf	Propagates log_class to all child module calls
modules/runners/variables.tf	Adds log_class variable with validation; updates runner_log_files description and type
modules/runners/logging.tf	Adds log_class to default log files; refactors count → for_each for log groups; uses per-entry class
modules/runners/pool.tf / pool/main.tf / pool/variables.tf	Passes log_class through to the pool lambda log group
modules/runners/scale-up.tf / scale-down.tf / ssm-housekeeper.tf	Applies log_group_class = var.log_class to lambda log groups
modules/lambda/variables.tf / main.tf	Adds optional log_class to lambda config object; applies to log group resource
modules/webhook/ (webhook, direct, eventbridge)	Adds log_class to config objects and applies to CloudWatch log groups
modules/runner-binaries-syncer/	Adds log_class variable and applies to syncer log group
modules/ami-housekeeper/	Adds log_class variable and applies to housekeeper log group
modules/termination-watcher/variables.tf	Adds log_class to the config object (flows to lambda module)
modules/multi-runner/	Adds log_class variable; propagates to all child modules including runner_log_files in multi_runner_config
README.md files	Documentation updates reflecting new log_class variable in all modules

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[npalm]

@Brend-Smits thx LGTM. Problem with destroying log groups seems solved.