Terraform Google Cloud Log Sink Module for GKE/Anthos Clusters

This Terraform module is tailored for GKE (Google Kubernetes Engine) and Anthos clusters. It provisions a Google Cloud Logging bucket, creates a sink for collecting logs from a source project related to the cluster, and grants appropriate permissions for the logging bucket in the destination project.

Usage

module "gcp_logging" {
  source              = "telia-oss/gke-logsink-module"
  source_project      = "your-source-project-id"
  destination_project = "your-destination-project-id"
  cluster_name        = ["your-cluster-name"]
}

Requirements

• Terraform v0.14 or newer
• A Google Cloud Platform account and project
• Appropriate IAM permissions to create and manage GCP Logging resources
• A GKE or Anthos cluster setup

Inputs

Name	Description	Type	Default	Required
source_project	The ID of the source GCP project.	string	n/a	yes
destination_project	The ID of the destination GCP project where logs will be stored.	string	n/a	yes
cluster_name	The name of the GKE or Anthos cluster.	list(string)	n/a	yes
custom_log_filter	The custom log filter.	string	n/a	no
retention_days	Log retention on the destination bucket.	number	90	no
multiple_clusters_per_sink	Option to use one sink for multiple cluster.	bool	false	no

Custom Log Filter

While the module is designed to provide a default logging filter for GKE or Anthos clusters, users have the flexibility to specify a custom filter based on their specific requirements.

Usage

To utilize the custom filter option:

1. Specify the custom_log_filter variable when using the module:

module "gcp_logging" {
  source              = "path-to-your-module"
  source_project      = "your-source-project-id"
  destination_project = "your-destination-project-id"
  cluster_name        = "your-cluster-name"
  custom_log_filter   = "YOUR_CUSTOM_FILTER_STRING"
}

If you do not provide a value for custom_log_filter, the module will use the default filter pattern based on your source_project and cluster_name.

2. Custom Filter Format:

Your custom filter string should conform to the Logging query language format. For example:

logName="projects/my-project-id/logs/my-log-id"

By providing a custom filter, you can fine-tune which logs are routed to the specified sink, allowing for greater control and specificity in your logging setup.

Outputs

Name	Description
bucket_id	The ID of the created logging bucket.
sink_name	The name of the created logging sink.
writer_identity	The writer identity of the created logging sink.

License

This module is licensed under the MIT License - see the LICENSE file for details.