gke-cluster-standard: change logging configuration (#1638)

* Update logging configuration of this module to use object interface in harmony with `gke-cluster-autopilot` module.
* Update blueprints that use this module.
* Add "WORKLOADS" log source to logging configuration of the blueprints where the README files say so.
* Update FAST stage 3 because it uses this module.
This commit is contained in:
Oliver Frolovs
2023-08-31 12:49:15 +01:00
committed by GitHub
parent 80e85ad343
commit 988fd2ee05
9 changed files with 190 additions and 52 deletions

View File

@@ -71,6 +71,58 @@ module "cluster-1" {
# tftest modules=1 resources=1 inventory=dataplane-v2.yaml
```
### Managing GKE logs
This example shows you how to [control which logs are sent from your GKE cluster to Cloud Logging](https://cloud.google.com/stackdriver/docs/solutions/gke/installing).
When you create a new GKE cluster, [Cloud Operations for GKE](https://cloud.google.com/stackdriver/docs/solutions/gke) integration with Cloud Logging is enabled by default and [System logs](https://cloud.google.com/stackdriver/docs/solutions/gke/managing-logs#what_logs) are collected. You can enable collection of several other [types of logs](https://cloud.google.com/stackdriver/docs/solutions/gke/managing-logs#what_logs). The following example enables collection of *all* optional logs.
```hcl
module "cluster-1" {
source = "./fabric/modules/gke-cluster-standard"
project_id = "myproject"
name = "cluster-1"
location = "europe-west1-b"
vpc_config = {
network = var.vpc.self_link
subnetwork = var.subnet.self_link
}
logging_config = {
enable_workloads_logs = true
enable_api_server_logs = true
enable_scheduler_logs = true
enable_controller_manager_logs = true
}
}
# tftest modules=1 resources=1 inventory=logging-config-enable-all.yaml
```
### Disable GKE logs collection
This example shows how to fully disable logs collection on a GKE Standard cluster. This is not recommended.
> **Warning**
> If you've disabled Cloud Logging or Cloud Monitoring, GKE customer support
> is offered on a best-effort basis and might require additional effort
> from your engineering team.
```hcl
module "cluster-1" {
source = "./fabric/modules/gke-cluster-standard"
project_id = "myproject"
name = "cluster-1"
location = "europe-west1-b"
vpc_config = {
network = var.vpc.self_link
subnetwork = var.subnet.self_link
}
logging_config = {
enable_system_logs = false
}
}
# tftest modules=1 resources=1 inventory=logging-config-disable-all.yaml
```
### Cloud DNS
This example shows how to [use Cloud DNS as a Kubernetes DNS provider](https://cloud.google.com/kubernetes-engine/docs/how-to/cloud-dns) for GKE Standard clusters.
@@ -97,7 +149,6 @@ module "cluster-1" {
# tftest modules=1 resources=1 inventory=dns.yaml
```
### Backup for GKE
This example shows how to [enable the Backup for GKE agent and configure a Backup Plan](https://cloud.google.com/kubernetes-engine/docs/add-on/backup-for-gke/concepts/backup-for-gke) for GKE Standard clusters.
@@ -131,9 +182,9 @@ module "cluster-1" {
| name | description | type | required | default |
|---|---|:---:|:---:|:---:|
| [location](variables.tf#L138) | Cluster zone or region. | <code>string</code> | ✓ | |
| [name](variables.tf#L195) | Cluster name. | <code>string</code> | ✓ | |
| [project_id](variables.tf#L221) | Cluster project id. | <code>string</code> | ✓ | |
| [vpc_config](variables.tf#L238) | VPC-level configuration. | <code title="object&#40;&#123;&#10; network &#61; string&#10; subnetwork &#61; string&#10; master_ipv4_cidr_block &#61; optional&#40;string&#41;&#10; secondary_range_blocks &#61; optional&#40;object&#40;&#123;&#10; pods &#61; string&#10; services &#61; string&#10; &#125;&#41;&#41;&#10; secondary_range_names &#61; optional&#40;object&#40;&#123;&#10; pods &#61; string&#10; services &#61; string&#10; &#125;&#41;, &#123; pods &#61; &#34;pods&#34;, services &#61; &#34;services&#34; &#125;&#41;&#10; master_authorized_ranges &#61; optional&#40;map&#40;string&#41;&#41;&#10; stack_type &#61; optional&#40;string&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | ✓ | |
| [name](variables.tf#L210) | Cluster name. | <code>string</code> | ✓ | |
| [project_id](variables.tf#L236) | Cluster project id. | <code>string</code> | ✓ | |
| [vpc_config](variables.tf#L253) | VPC-level configuration. | <code title="object&#40;&#123;&#10; network &#61; string&#10; subnetwork &#61; string&#10; master_ipv4_cidr_block &#61; optional&#40;string&#41;&#10; secondary_range_blocks &#61; optional&#40;object&#40;&#123;&#10; pods &#61; string&#10; services &#61; string&#10; &#125;&#41;&#41;&#10; secondary_range_names &#61; optional&#40;object&#40;&#123;&#10; pods &#61; string&#10; services &#61; string&#10; &#125;&#41;, &#123; pods &#61; &#34;pods&#34;, services &#61; &#34;services&#34; &#125;&#41;&#10; master_authorized_ranges &#61; optional&#40;map&#40;string&#41;&#41;&#10; stack_type &#61; optional&#40;string&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | ✓ | |
| [backup_configs](variables.tf#L17) | Configuration for Backup for GKE. | <code title="object&#40;&#123;&#10; enable_backup_agent &#61; optional&#40;bool, false&#41;&#10; backup_plans &#61; optional&#40;map&#40;object&#40;&#123;&#10; encryption_key &#61; optional&#40;string&#41;&#10; include_secrets &#61; optional&#40;bool, true&#41;&#10; include_volume_data &#61; optional&#40;bool, true&#41;&#10; namespaces &#61; optional&#40;list&#40;string&#41;&#41;&#10; region &#61; string&#10; schedule &#61; string&#10; retention_policy_days &#61; optional&#40;string&#41;&#10; retention_policy_lock &#61; optional&#40;bool, false&#41;&#10; retention_policy_delete_lock_days &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;, &#123;&#125;&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code>&#123;&#125;</code> |
| [cluster_autoscaling](variables.tf#L37) | Enable and configure limits for Node Auto-Provisioning with Cluster Autoscaler. | <code title="object&#40;&#123;&#10; auto_provisioning_defaults &#61; optional&#40;object&#40;&#123;&#10; boot_disk_kms_key &#61; optional&#40;string&#41;&#10; image_type &#61; optional&#40;string&#41;&#10; oauth_scopes &#61; optional&#40;list&#40;string&#41;&#41;&#10; service_account &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#10; cpu_limits &#61; optional&#40;object&#40;&#123;&#10; min &#61; number&#10; max &#61; number&#10; &#125;&#41;&#41;&#10; mem_limits &#61; optional&#40;object&#40;&#123;&#10; min &#61; number&#10; max &#61; number&#10; &#125;&#41;&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code>null</code> |
| [description](variables.tf#L58) | Cluster description. | <code>string</code> | | <code>null</code> |
@@ -141,15 +192,15 @@ module "cluster-1" {
| [enable_features](variables.tf#L87) | Enable cluster-level features. Certain features allow configuration. | <code title="object&#40;&#123;&#10; binary_authorization &#61; optional&#40;bool, false&#41;&#10; cost_management &#61; optional&#40;bool, false&#41;&#10; dns &#61; optional&#40;object&#40;&#123;&#10; provider &#61; optional&#40;string&#41;&#10; scope &#61; optional&#40;string&#41;&#10; domain &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#10; database_encryption &#61; optional&#40;object&#40;&#123;&#10; state &#61; string&#10; key_name &#61; string&#10; &#125;&#41;&#41;&#10; dataplane_v2 &#61; optional&#40;bool, false&#41;&#10; gateway_api &#61; optional&#40;bool, false&#41;&#10; groups_for_rbac &#61; optional&#40;string&#41;&#10; intranode_visibility &#61; optional&#40;bool, false&#41;&#10; l4_ilb_subsetting &#61; optional&#40;bool, false&#41;&#10; mesh_certificates &#61; optional&#40;bool&#41;&#10; pod_security_policy &#61; optional&#40;bool, false&#41;&#10; resource_usage_export &#61; optional&#40;object&#40;&#123;&#10; dataset &#61; string&#10; enable_network_egress_metering &#61; optional&#40;bool&#41;&#10; enable_resource_consumption_metering &#61; optional&#40;bool&#41;&#10; &#125;&#41;&#41;&#10; shielded_nodes &#61; optional&#40;bool, false&#41;&#10; tpu &#61; optional&#40;bool, false&#41;&#10; upgrade_notifications &#61; optional&#40;object&#40;&#123;&#10; topic_id &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#10; vertical_pod_autoscaling &#61; optional&#40;bool, false&#41;&#10; workload_identity &#61; optional&#40;bool, true&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code title="&#123;&#10; workload_identity &#61; true&#10;&#125;">&#123;&#8230;&#125;</code> |
| [issue_client_certificate](variables.tf#L126) | Enable issuing client certificate. | <code>bool</code> | | <code>false</code> |
| [labels](variables.tf#L132) | Cluster resource labels. | <code>map&#40;string&#41;</code> | | <code>null</code> |
| [logging_config](variables.tf#L143) | Logging configuration. | <code>list&#40;string&#41;</code> | | <code>&#91;&#34;SYSTEM_COMPONENTS&#34;&#93;</code> |
| [maintenance_config](variables.tf#L149) | Maintenance window configuration. | <code title="object&#40;&#123;&#10; daily_window_start_time &#61; optional&#40;string&#41;&#10; recurring_window &#61; optional&#40;object&#40;&#123;&#10; start_time &#61; string&#10; end_time &#61; string&#10; recurrence &#61; string&#10; &#125;&#41;&#41;&#10; maintenance_exclusions &#61; optional&#40;list&#40;object&#40;&#123;&#10; name &#61; string&#10; start_time &#61; string&#10; end_time &#61; string&#10; scope &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code title="&#123;&#10; daily_window_start_time &#61; &#34;03:00&#34;&#10; recurring_window &#61; null&#10; maintenance_exclusion &#61; &#91;&#93;&#10;&#125;">&#123;&#8230;&#125;</code> |
| [max_pods_per_node](variables.tf#L172) | Maximum number of pods per node in this cluster. | <code>number</code> | | <code>110</code> |
| [min_master_version](variables.tf#L178) | Minimum version of the master, defaults to the version of the most recent official release. | <code>string</code> | | <code>null</code> |
| [monitoring_config](variables.tf#L184) | Monitoring components. | <code title="object&#40;&#123;&#10; enable_components &#61; optional&#40;list&#40;string&#41;&#41;&#10; managed_prometheus &#61; optional&#40;bool&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code title="&#123;&#10; enable_components &#61; &#91;&#34;SYSTEM_COMPONENTS&#34;&#93;&#10;&#125;">&#123;&#8230;&#125;</code> |
| [node_locations](variables.tf#L200) | Zones in which the cluster's nodes are located. | <code>list&#40;string&#41;</code> | | <code>&#91;&#93;</code> |
| [private_cluster_config](variables.tf#L207) | Private cluster configuration. | <code title="object&#40;&#123;&#10; enable_private_endpoint &#61; optional&#40;bool&#41;&#10; master_global_access &#61; optional&#40;bool&#41;&#10; peering_config &#61; optional&#40;object&#40;&#123;&#10; export_routes &#61; optional&#40;bool&#41;&#10; import_routes &#61; optional&#40;bool&#41;&#10; project_id &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code>null</code> |
| [release_channel](variables.tf#L226) | Release channel for GKE upgrades. | <code>string</code> | | <code>null</code> |
| [tags](variables.tf#L232) | Network tags applied to nodes. | <code>list&#40;string&#41;</code> | | <code>null</code> |
| [logging_config](variables.tf#L143) | Logging configuration. | <code title="object&#40;&#123;&#10; enable_system_logs &#61; optional&#40;bool, true&#41;&#10; enable_workloads_logs &#61; optional&#40;bool, false&#41;&#10; enable_api_server_logs &#61; optional&#40;bool, false&#41;&#10; enable_scheduler_logs &#61; optional&#40;bool, false&#41;&#10; enable_controller_manager_logs &#61; optional&#40;bool, false&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code>&#123;&#125;</code> |
| [maintenance_config](variables.tf#L164) | Maintenance window configuration. | <code title="object&#40;&#123;&#10; daily_window_start_time &#61; optional&#40;string&#41;&#10; recurring_window &#61; optional&#40;object&#40;&#123;&#10; start_time &#61; string&#10; end_time &#61; string&#10; recurrence &#61; string&#10; &#125;&#41;&#41;&#10; maintenance_exclusions &#61; optional&#40;list&#40;object&#40;&#123;&#10; name &#61; string&#10; start_time &#61; string&#10; end_time &#61; string&#10; scope &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code title="&#123;&#10; daily_window_start_time &#61; &#34;03:00&#34;&#10; recurring_window &#61; null&#10; maintenance_exclusion &#61; &#91;&#93;&#10;&#125;">&#123;&#8230;&#125;</code> |
| [max_pods_per_node](variables.tf#L187) | Maximum number of pods per node in this cluster. | <code>number</code> | | <code>110</code> |
| [min_master_version](variables.tf#L193) | Minimum version of the master, defaults to the version of the most recent official release. | <code>string</code> | | <code>null</code> |
| [monitoring_config](variables.tf#L199) | Monitoring components. | <code title="object&#40;&#123;&#10; enable_components &#61; optional&#40;list&#40;string&#41;&#41;&#10; managed_prometheus &#61; optional&#40;bool&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code title="&#123;&#10; enable_components &#61; &#91;&#34;SYSTEM_COMPONENTS&#34;&#93;&#10;&#125;">&#123;&#8230;&#125;</code> |
| [node_locations](variables.tf#L215) | Zones in which the cluster's nodes are located. | <code>list&#40;string&#41;</code> | | <code>&#91;&#93;</code> |
| [private_cluster_config](variables.tf#L222) | Private cluster configuration. | <code title="object&#40;&#123;&#10; enable_private_endpoint &#61; optional&#40;bool&#41;&#10; master_global_access &#61; optional&#40;bool&#41;&#10; peering_config &#61; optional&#40;object&#40;&#123;&#10; export_routes &#61; optional&#40;bool&#41;&#10; import_routes &#61; optional&#40;bool&#41;&#10; project_id &#61; optional&#40;string&#41;&#10; &#125;&#41;&#41;&#10;&#125;&#41;">object&#40;&#123;&#8230;&#125;&#41;</code> | | <code>null</code> |
| [release_channel](variables.tf#L241) | Release channel for GKE upgrades. | <code>string</code> | | <code>null</code> |
| [tags](variables.tf#L247) | Network tags applied to nodes. | <code>list&#40;string&#41;</code> | | <code>null</code> |
## Outputs

View File

@@ -181,10 +181,27 @@ resource "google_container_cluster" "cluster" {
}
}
# Send GKE cluster logs from chosen sources to Cloud Logging.
# System logs must be enabled if any other source is enabled.
# This is validated by input variable validation rules.
dynamic "logging_config" {
for_each = var.logging_config != null ? [""] : []
for_each = var.logging_config.enable_system_logs ? [""] : []
content {
enable_components = var.logging_config
enable_components = toset(compact([
var.logging_config.enable_api_server_logs ? "APISERVER" : null,
var.logging_config.enable_controller_manager_logs ? "CONTROLLER_MANAGER" : null,
var.logging_config.enable_scheduler_logs ? "SCHEDULER" : null,
"SYSTEM_COMPONENTS",
var.logging_config.enable_workloads_logs ? "WORKLOADS" : null,
]))
}
}
# Don't send any GKE cluster logs to Cloud Logging. Input variable validation
# makes sure every other log source is false when enable_system_logs is false.
dynamic "logging_config" {
for_each = var.logging_config.enable_system_logs == false ? [""] : []
content {
enable_components = []
}
}

View File

@@ -142,8 +142,23 @@ variable "location" {
variable "logging_config" {
description = "Logging configuration."
type = list(string)
default = ["SYSTEM_COMPONENTS"]
type = object({
enable_system_logs = optional(bool, true)
enable_workloads_logs = optional(bool, false)
enable_api_server_logs = optional(bool, false)
enable_scheduler_logs = optional(bool, false)
enable_controller_manager_logs = optional(bool, false)
})
default = {}
nullable = false
# System logs are the minimum required component for enabling log collection.
# So either everything is off (false), or enable_system_logs must be true.
validation {
condition = (
!anytrue(values(var.logging_config)) || var.logging_config.enable_system_logs
)
error_message = "System logs are the minimum required component for enabling log collection."
}
}
variable "maintenance_config" {