PLT-1911: Added graceful_creation_timeout feature for all clusters

docs/data-sources/cluster.md

@@ -50,5 +50,7 @@ resource "local_file" "admin_kube_config" {
 ### Read-Only
 
 - `admin_kube_config` (String) The admin kubeconfig file for accessing the cluster. This is computed automatically.
+- `health` (String) The current health status of the cluster. This is computed automatically.
 - `id` (String) The ID of this resource.
 - `kube_config` (String) The kubeconfig file for accessing the cluster as a non-admin user. This is computed automatically.
+- `state` (String) The current state of the cluster. This is computed automatically.

docs/resources/cluster_aks.md

@@ -129,6 +129,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) The date and time after which to patch the cluster. Prefix the time value with the respective RFC. Ex: `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_aws.md

@@ -160,6 +160,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) Date and time after which to patch cluster `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_azure.md

@@ -142,6 +142,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) Date and time after which to patch cluster `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_custom_cloud.md

@@ -160,6 +160,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `location_config` (Block List) (see [below for nested schema](#nestedblock--location_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) Date and time after which to patch cluster `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_edge_native.md

@@ -99,6 +99,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `location_config` (Block List) (see [below for nested schema](#nestedblock--location_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))

docs/resources/cluster_edge_vsphere.md

@@ -34,6 +34,7 @@ description: |-
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `location_config` (Block List) (see [below for nested schema](#nestedblock--location_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))

docs/resources/cluster_eks.md

@@ -131,6 +131,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `fargate_profile` (Block List) (see [below for nested schema](#nestedblock--fargate_profile))
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) Date and time after which to patch cluster `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_gcp.md

@@ -119,6 +119,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) Date and time after which to patch cluster `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_gke.md

@@ -89,6 +89,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) Date and time after which to patch cluster `RFC3339: 2006-01-02T15:04:05Z07:00`

docs/resources/cluster_maas.md

@@ -142,6 +142,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `location_config` (Block List) (see [below for nested schema](#nestedblock--location_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))

docs/resources/cluster_openstack.md

@@ -101,6 +101,7 @@ resource "spectrocloud_cluster_openstack" "cluster" {
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `location_config` (Block List) (see [below for nested schema](#nestedblock--location_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))

docs/resources/cluster_vsphere.md

@@ -121,6 +121,7 @@ Refer to the [Import section](/docs#import) to learn more.
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_config` (Block List) The host configuration for the cluster. (see [below for nested schema](#nestedblock--host_config))
 - `location_config` (Block List) (see [below for nested schema](#nestedblock--location_config))
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))

docs/resources/virtual_cluster.md

@@ -66,6 +66,7 @@ resource "spectrocloud_virtual_cluster" "cluster" {
 - `description` (String) The description of the cluster. Default value is empty string.
 - `force_delete` (Boolean) If set to `true`, the cluster will be force deleted and user has to manually clean up the provisioned cloud resources.
 - `force_delete_delay` (Number) Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.
+- `graceful_creation_timeout` (Boolean) If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.
 - `host_cluster_uid` (String)
 - `namespaces` (Block List) The namespaces for the cluster. (see [below for nested schema](#nestedblock--namespaces))
 - `os_patch_after` (String) The date and time after which to patch the cluster. Prefix the time value with the respective RFC. Ex: `RFC3339: 2006-01-02T15:04:05Z07:00`

spectrocloud/cluster_common_crud.go

@@ -2,6 +2,8 @@ package spectrocloud
 
 import (
 	"context"
+	"errors"
+	"fmt"
 	"log"
 	"time"
 
@@ -113,6 +115,13 @@ func resourceClusterReadyRefreshFunc(c *client.V1Client, id string) retry.StateR
 func waitForClusterCreation(ctx context.Context, d *schema.ResourceData, uid string, diags diag.Diagnostics, c *client.V1Client, initial bool) (diag.Diagnostics, bool) {
 	d.SetId(uid)
 
+	// Debug logging
+	if gracefulCreationTimeout, ok := d.GetOk("graceful_creation_timeout"); ok {
+		log.Printf("waitForClusterCreation: graceful_creation_timeout = %v", gracefulCreationTimeout)
+	} else {
+		log.Printf("waitForClusterCreation: graceful_creation_timeout not set or false")
+	}
+
 	if initial { // only skip_completion when initally creating a cluster, do not skip when attach addon profile
 		if d.Get("skip_completion") != nil && d.Get("skip_completion").(bool) {
 			return diags, true
@@ -140,6 +149,43 @@ func waitForClusterCreation(ctx context.Context, d *schema.ResourceData, uid str
 	// Wait, catching any errors
 	_, err := stateConf.WaitForStateContext(ctx)
 	if err != nil {
+		// Check if graceful_creation_timeout is enabled and this is a timeout error
+		if gracefulCreationTimeout, ok := d.GetOk("graceful_creation_timeout"); ok && gracefulCreationTimeout.(bool) {
+			// Check if this is a timeout error
+			var timeoutErr *retry.TimeoutError
+			if errors.As(err, &timeoutErr) {
+				log.Printf("waitForClusterCreation (Running-Healthy phase): graceful timeout handling enabled, returning warning instead of error")
+
+				// Get current cluster state for warning message
+				cluster, stateErr := c.GetCluster(d.Id())
+				currentState := timeoutErr.LastState
+				if currentState == "" {
+					currentState = "Unknown"
+				}
+				if stateErr == nil && cluster != nil && cluster.Status != nil {
+					currentState = cluster.Status.State
+					if cluster.Status.State == "Running" {
+						if clusterSummary, _ := c.GetClusterOverview(d.Id()); clusterSummary != nil && clusterSummary.Status.Health != nil {
+							if clusterSummary.Status.Health.State != "" && clusterSummary.Status.Health.State != "Healthy" {
+								currentState += "-" + clusterSummary.Status.Health.State
+							}
+						}
+					}
+				}
+
+				// Return warning instead of error
+				diags = append(diags, diag.Diagnostic{
+					Severity: diag.Warning,
+					Summary:  "Cluster creation timeout",
+					Detail: fmt.Sprintf(
+						"Cluster creation timed out after waiting for %v. Current cluster state is '%s'. "+
+							"The cluster may still be provisioning in the background and could eventually reach the 'Running-Healthy' state.",
+						d.Timeout(schema.TimeoutCreate), currentState),
+				})
+				return diags, false
+			}
+		}
+		// For non-timeout errors or when graceful_creation_timeout is disabled, still return the error
 		return diag.FromErr(err), true
 	}
 	return nil, false

spectrocloud/data_source_cluster.go

@@ -43,6 +43,16 @@ func dataSourceCluster() *schema.Resource {
 				Default:     false,
 				Description: "If set to true, the cluster will treated as a virtual cluster. Defaults to `false`.",
 			},
+			"state": {
+				Type:        schema.TypeString,
+				Computed:    true,
+				Description: "The current state of the cluster. This is computed automatically.",
+			},
+			"health": {
+				Type:        schema.TypeString,
+				Computed:    true,
+				Description: "The current health status of the cluster. This is computed automatically.",
+			},
 		},
 	}
 }
@@ -73,6 +83,21 @@ func dataSourceClusterRead(_ context.Context, d *schema.ResourceData, m interfac
 			if err := d.Set("name", cluster.Metadata.Name); err != nil {
 				return diag.FromErr(err)
 			}
+
+			// Set cluster state
+			if cluster.Status != nil && cluster.Status.State != "" {
+				if err := d.Set("state", cluster.Status.State); err != nil {
+					return diag.FromErr(err)
+				}
+			}
+
+			// Set cluster health
+			clusterSummary, summaryErr := c.GetClusterOverview(cluster.Metadata.UID)
+			if summaryErr == nil && clusterSummary.Status.Health != nil && clusterSummary.Status.Health.State != "" {
+				if err := d.Set("health", clusterSummary.Status.Health.State); err != nil {
+					return diag.FromErr(err)
+				}
+			}
 		}
 	}
 	return diags

spectrocloud/data_source_cluster_test.go

@@ -2,10 +2,11 @@ package spectrocloud
 
 import (
 	"context"
+	"testing"
+
 	"github.com/hashicorp/terraform-plugin-sdk/v2/diag"
 	"github.com/hashicorp/terraform-plugin-sdk/v2/helper/schema"
 	"github.com/stretchr/testify/assert"
-	"testing"
 )
 
 func TestDataSourceClusterRead(t *testing.T) {
@@ -23,6 +24,8 @@ func TestDataSourceClusterRead(t *testing.T) {
 				"virtual":           {Type: schema.TypeBool, Optional: true},
 				"kube_config":       {Type: schema.TypeString, Computed: true},
 				"admin_kube_config": {Type: schema.TypeString, Computed: true},
+				"state":             {Type: schema.TypeString, Computed: true},
+				"health":            {Type: schema.TypeString, Computed: true},
 			}, map[string]interface{}{
 				"name":    "test-cluster",
 				"context": "some-context",
@@ -39,6 +42,8 @@ func TestDataSourceClusterRead(t *testing.T) {
 				"virtual":           {Type: schema.TypeBool, Optional: true},
 				"kube_config":       {Type: schema.TypeString, Computed: true},
 				"admin_kube_config": {Type: schema.TypeString, Computed: true},
+				"state":             {Type: schema.TypeString, Computed: true},
+				"health":            {Type: schema.TypeString, Computed: true},
 			}, map[string]interface{}{
 				"name":              "test-cluster",
 				"context":           "some-context",

spectrocloud/resource_cluster_aks.go

@@ -306,6 +306,12 @@ func resourceClusterAks() *schema.Resource {
 				Description:      "Delay duration in minutes to before invoking cluster force delete. Default and minimum is 20.",
 				ValidateDiagFunc: validation.ToDiagFunc(validation.IntAtLeast(20)),
 			},
+			"graceful_creation_timeout": {
+				Type:        schema.TypeBool,
+				Optional:    true,
+				Default:     false,
+				Description: "If set to `true`, cluster creation timeouts will be handled gracefully with warnings instead of errors, preserving the resource in state for subsequent applies to reconcile. Default is `false`.",
+			},
 		},
 	}
 }
@@ -331,6 +337,9 @@ func resourceClusterAksCreate(ctx context.Context, d *schema.ResourceData, m int
 	if isError {
 		return diagnostics
 	}
+	if len(diagnostics) > 0 {
+		diags = append(diags, diagnostics...)
+	}
 
 	resourceClusterAksRead(ctx, d, m)