Merge pull request #828 from gdbranco/fix/ocm-12048

OCM-12048 | feat: include indicator for bypassing machine pool deletion errors

docs/data-sources/hcp_machine_pool.md

@@ -39,6 +39,7 @@ data "rhcs_hcp_machine_pool" "machine_pool" {
 - `aws_node_pool` (Attributes) AWS settings for node pool (see [below for nested schema](#nestedatt--aws_node_pool))
 - `current_version` (String) The currently running version of OpenShift on the machine pool, for example '4.11.0'.
 - `id` (String) Unique identifier of the machine pool.
+- `ignore_deletion_error` (Boolean) Indicates to the provider to disregard API errors when deleting the machine pool. This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors. Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file. This is not recommended to be set in other use cases
 - `kubelet_configs` (String) Name of the kubelet config applied to the machine pool.
 - `labels` (Map of String) Labels for the machine pool. Format should be a comma-separated list of 'key = value'. This list will overwrite any modifications made to node labels on an ongoing basis.
 - `replicas` (Number) The number of machines of the pool

docs/data-sources/machine_pool.md

@@ -34,6 +34,7 @@ data "rhcs_machine_pool" "machine_pool" {
 - `availability_zones` (List of String) A list of Availability Zones. Relevant only for multiple availability zones machine pool. For single availability zone check "availability_zone" attribute.
 - `aws_additional_security_group_ids` (List of String) AWS additional security group ids.
 - `disk_size` (Number) The root disk size, in GiB.
+- `ignore_deletion_error` (Boolean) Indicates to the provider to disregard API errors when deleting the machine pool. This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors. Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file. This is not recommended to be set in other use cases
 - `labels` (Map of String) The list of the Labels of this machine pool.
 - `machine_type` (String) Identifier of the machine type used by the nodes, for example `m5.xlarge`.
 - `max_replicas` (Number) The maximum number of replicas for auto-scaling functionality. relevant only in case of 'autoscaling_enabled = true'

docs/resources/hcp_machine_pool.md

@@ -42,6 +42,7 @@ resource "rhcs_hcp_machine_pool" "machine_pool" {
 
 ### Optional
 
+- `ignore_deletion_error` (Boolean) Indicates to the provider to disregard API errors when deleting the machine pool. This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors. Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file. This is not recommended to be set in other use cases
 - `kubelet_configs` (String) Name of the kubelet config applied to the machine pool. A single kubelet config is allowed. Kubelet config must already exist.
 - `labels` (Map of String) Labels for the machine pool. Format should be a comma-separated list of 'key = value'. This list will overwrite any modifications made to node labels on an ongoing basis.
 - `replicas` (Number) The number of machines of the pool

docs/resources/machine_pool.md

@@ -38,6 +38,7 @@ resource "rhcs_machine_pool" "machine_pool" {
 - `aws_additional_security_group_ids` (List of String) AWS additional security group ids. After the creation of the resource, it is not possible to update the attribute value.
 - `aws_tags` (Map of String) Apply user defined tags to all machine pool resources created in AWS. After the creation of the resource, it is not possible to update the attribute value.
 - `disk_size` (Number) Root disk size, in GiB. After the creation of the resource, it is not possible to update the attribute value.
+- `ignore_deletion_error` (Boolean) Indicates to the provider to disregard API errors when deleting the machine pool. This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors. Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file. This is not recommended to be set in other use cases
 - `labels` (Map of String) Labels for the machine pool. Format should be a comma-separated list of 'key = value'. This list will overwrite any modifications made to node labels on an ongoing basis.
 - `max_replicas` (Number) The maximum number of replicas for autoscaling functionality.
 - `max_spot_price` (Number) Max Spot price. After the creation of the resource, it is not possible to update the attribute value.

provider/machinepool/classic/machine_pool_datasource.go

@@ -165,6 +165,13 @@ func (r *MachinePoolDatasource) Schema(ctx context.Context, req datasource.Schem
 				ElementType: types.StringType,
 				Computed:    true,
 			},
+			"ignore_deletion_error": schema.BoolAttribute{
+				Description: "Indicates to the provider to disregard API errors when deleting the machine pool." +
+					" This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors." +
+					" Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file." +
+					" This is not recommended to be set in other use cases",
+				Computed: true,
+			},
 		},
 	}
 }
@@ -195,5 +202,7 @@ func (r *MachinePoolDatasource) Read(ctx context.Context, req datasource.ReadReq
 		return
 	}
 
+	state.IgnoreDeletionError = types.BoolNull()
+
 	resp.Diagnostics.Append(resp.State.Set(ctx, state)...)
 }

provider/machinepool/classic/machine_pool_resource.go

@@ -34,6 +34,7 @@ import (
 	"github.com/hashicorp/terraform-plugin-framework/path"
 	"github.com/hashicorp/terraform-plugin-framework/resource"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema"
+	"github.com/hashicorp/terraform-plugin-framework/resource/schema/booldefault"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema/boolplanmodifier"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema/int64planmodifier"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema/planmodifier"
@@ -215,6 +216,15 @@ func (r *MachinePoolResource) Schema(ctx context.Context, req resource.SchemaReq
 				ElementType: types.StringType,
 				Optional:    true,
 			},
+			"ignore_deletion_error": schema.BoolAttribute{
+				Description: "Indicates to the provider to disregard API errors when deleting the machine pool." +
+					" This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors." +
+					" Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file." +
+					" This is not recommended to be set in other use cases",
+				Optional: true,
+				Computed: true,
+				Default:  booldefault.StaticBool(false),
+			},
 		},
 	}
 }
@@ -803,6 +813,8 @@ func adjustInitialStateToPlan(state, plan *MachinePoolState) {
 	// update the Replicas with the plan value (important for nil and zero value cases)
 	state.Replicas = plan.Replicas
 
+	state.IgnoreDeletionError = plan.IgnoreDeletionError
+
 	if common.HasValue(plan.AwsTags) {
 		state.AwsTags = plan.AwsTags
 	}
@@ -926,6 +938,19 @@ func (r *MachinePoolResource) Delete(ctx context.Context, req resource.DeleteReq
 		MachinePool(state.ID.ValueString())
 	_, err := resource.Delete().SendContext(ctx)
 	if err != nil {
+		if common.BoolWithFalseDefault(state.IgnoreDeletionError) {
+			resp.Diagnostics.AddWarning(
+				"Cannot delete machine pool",
+				fmt.Sprintf(
+					"An error occurred deleting the pool,"+
+						" because ignore deletion error is set it will still be removed from the terraform state. Reason: %v",
+					err,
+				),
+			)
+			// Remove the state:
+			resp.State.RemoveResource(ctx)
+			return
+		}
 		// We can't delete the pool, see if it's the last one:
 		numPools, err2 := r.countPools(ctx, state.Cluster.ValueString())
 		if numPools == 1 && err2 == nil {
@@ -941,7 +966,9 @@ func (r *MachinePoolResource) Delete(ctx context.Context, req resource.DeleteReq
 					state.Cluster.ValueString(),
 				),
 			)
-			// No return, we want to remove the state
+			// Remove the state:
+			resp.State.RemoveResource(ctx)
+			return
 		} else {
 			// Wasn't the last one, return error
 			resp.Diagnostics.AddError(
@@ -956,6 +983,7 @@ func (r *MachinePoolResource) Delete(ctx context.Context, req resource.DeleteReq
 		}
 	}
 
+	// Everything went fine deleting the resource in OCM
 	// Remove the state:
 	resp.State.RemoveResource(ctx)
 }

provider/machinepool/classic/machine_pool_state.go

@@ -41,6 +41,7 @@ type MachinePoolState struct {
 	DiskSize                   types.Int64   `tfsdk:"disk_size"`
 	AdditionalSecurityGroupIds types.List    `tfsdk:"aws_additional_security_group_ids"`
 	AwsTags                    types.Map     `tfsdk:"aws_tags"`
+	IgnoreDeletionError        types.Bool    `tfsdk:"ignore_deletion_error"`
 }
 
 type Taints struct {

provider/machinepool/hcp/machine_pool_datasource.go

@@ -172,6 +172,13 @@ func (r *HcpMachinePoolDatasource) Schema(ctx context.Context, req datasource.Sc
 					"upgrade to OpenShift 4.12.z from 4.11 or before).",
 				Computed: true,
 			},
+			"ignore_deletion_error": schema.BoolAttribute{
+				Description: "Indicates to the provider to disregard API errors when deleting the machine pool." +
+					" This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors." +
+					" Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file." +
+					" This is not recommended to be set in other use cases",
+				Computed: true,
+			},
 		},
 	}
 }
@@ -204,6 +211,7 @@ func (r *HcpMachinePoolDatasource) Read(ctx context.Context, req datasource.Read
 
 	state.UpgradeAcksFor = types.StringNull()
 	state.Version = types.StringNull()
+	state.IgnoreDeletionError = types.BoolNull()
 
 	resp.Diagnostics.Append(resp.State.Set(ctx, state)...)
 }

provider/machinepool/hcp/machine_pool_resource.go

@@ -35,6 +35,7 @@ import (
 	"github.com/hashicorp/terraform-plugin-framework/path"
 	"github.com/hashicorp/terraform-plugin-framework/resource"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema"
+	"github.com/hashicorp/terraform-plugin-framework/resource/schema/booldefault"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema/planmodifier"
 	"github.com/hashicorp/terraform-plugin-framework/resource/schema/stringplanmodifier"
 	"github.com/hashicorp/terraform-plugin-framework/schema/validator"
@@ -206,6 +207,15 @@ func (r *HcpMachinePoolResource) Schema(ctx context.Context, req resource.Schema
 					"upgrade to OpenShift 4.12.z from 4.11 or before).",
 				Optional: true,
 			},
+			"ignore_deletion_error": schema.BoolAttribute{
+				Description: "Indicates to the provider to disregard API errors when deleting the machine pool." +
+					" This will remove the resource from the management file, but not necessirely delete the underlying pool in case it errors." +
+					" Setting this to true can bypass issues when destroying the cluster resource alongside the pool resource in the same management file." +
+					" This is not recommended to be set in other use cases",
+				Optional: true,
+				Computed: true,
+				Default:  booldefault.StaticBool(false),
+			},
 		},
 	}
 }
@@ -826,6 +836,7 @@ func adjustInitialStateToPlan(state, plan *HcpMachinePoolState) {
 	state.Replicas = plan.Replicas
 	state.NodePoolStatus = plan.NodePoolStatus
 	state.Version = plan.Version
+	state.IgnoreDeletionError = plan.IgnoreDeletionError
 
 	if state.AWSNodePool == nil {
 		state.AWSNodePool = new(AWSNodePool)
@@ -1068,6 +1079,19 @@ func (r *HcpMachinePoolResource) Delete(ctx context.Context, req resource.Delete
 		NodePool(state.ID.ValueString())
 	_, err := resource.Delete().SendContext(ctx)
 	if err != nil {
+		if common.BoolWithFalseDefault(state.IgnoreDeletionError) {
+			resp.Diagnostics.AddWarning(
+				"Cannot delete machine pool",
+				fmt.Sprintf(
+					"An error occurred deleting the pool,"+
+						" because ignore deletion error is set it will still be removed from the terraform state. Reason: %v",
+					err,
+				),
+			)
+			// Remove the state:
+			resp.State.RemoveResource(ctx)
+			return
+		}
 		// We can't delete the pool, see if it's the last one:
 		numPools, err2 := r.countPools(ctx, state.Cluster.ValueString())
 		if numPools == 1 && err2 == nil {
@@ -1083,7 +1107,9 @@ func (r *HcpMachinePoolResource) Delete(ctx context.Context, req resource.Delete
 					state.Cluster.ValueString(),
 				),
 			)
-			// No return, we want to remove the state
+			// Remove the state:
+			resp.State.RemoveResource(ctx)
+			return
 		} else {
 			// Wasn't the last one, return error
 			resp.Diagnostics.AddError(
@@ -1098,6 +1124,7 @@ func (r *HcpMachinePoolResource) Delete(ctx context.Context, req resource.Delete
 		}
 	}
 
+	// Everything went fine deleting the resource in OCM
 	// Remove the state:
 	resp.State.RemoveResource(ctx)
 }

provider/machinepool/hcp/machine_pool_state.go

@@ -42,6 +42,8 @@ type HcpMachinePoolState struct {
 	TuningConfigs  types.List   `tfsdk:"tuning_configs"`
 	KubeletConfigs types.String `tfsdk:"kubelet_configs"`
 	AutoRepair     types.Bool   `tfsdk:"auto_repair"`
+
+	IgnoreDeletionError types.Bool `tfsdk:"ignore_deletion_error"`
 }
 
 type Taints struct {