feat: Long-running operation improvements for mongodbatlas_cloud_backup_snapshot resource (#3536)

* rename

* implement delete_on_create_timeout

* fix delete_on_create_timeout due to lack of update

* refactor test

* rename test

* improve var naming

Author: oarbusi

.changelog/3536.txt

@@ -0,0 +1,3 @@
+```release-note:enhancement
+resource/mongodbatlas_cloud_backup_snapshot: Adds `delete_on_create_timeout` attribute to indicate whether to delete the resource if its creation times out
+``` 

docs/resources/cloud_backup_snapshot.md

@@ -57,6 +57,7 @@ resource "mongodbatlas_cloud_backup_snapshot_restore_job" "test" {
 * `description` - (Required) Description of the on-demand snapshot.
 * `retention_in_days` - (Required) The number of days that Atlas should retain the on-demand snapshot. Must be at least 1.
 * `timeouts`- (Optional) The duration of time to wait for Atlas to create a Cloud Backup Snapshot. The timeout value is defined by a signed sequence of decimal numbers with a time unit suffix such as: `1h45m`, `300s`, `10m`, etc. The valid time units are:  `ns`, `us` (or `µs`), `ms`, `s`, `m`, `h`. Defaults to `1h`. Learn more about timeouts [here](https://www.terraform.io/plugin/sdkv2/resources/retries-and-customizable-timeouts).
+* `delete_on_create_timeout`- (Optional) Flag that indicates whether to delete the resource if creation times out. Default is `true`. When Terraform apply fails, it returns immediately without waiting for cleanup to complete. If you suspect a transient error, wait before retrying to allow resource deletion to finish.
 
 ## Attributes Reference
 

internal/service/cloudbackupsnapshot/data_source_cloud_backup_snapshot.go renamed to internal/service/cloudbackupsnapshot/data_source.go

@@ -10,6 +10,7 @@ import (
 	"github.com/hashicorp/terraform-plugin-sdk/v2/helper/retry"
 	"github.com/hashicorp/terraform-plugin-sdk/v2/helper/schema"
 	"github.com/hashicorp/terraform-plugin-sdk/v2/helper/validation"
+	"github.com/mongodb/terraform-provider-mongodbatlas/internal/common/cleanup"
 	"github.com/mongodb/terraform-provider-mongodbatlas/internal/common/conversion"
 	"github.com/mongodb/terraform-provider-mongodbatlas/internal/common/validate"
 	"github.com/mongodb/terraform-provider-mongodbatlas/internal/config"
@@ -19,9 +20,9 @@ import (
 
 func Resource() *schema.Resource {
 	return &schema.Resource{
-		CreateContext: resourceCreate,
-		ReadContext:   resourceRead,
-		DeleteContext: resourceDelete,
+		CreateWithoutTimeout: resourceCreate,
+		ReadWithoutTimeout:   resourceRead,
+		DeleteWithoutTimeout: resourceDelete,
 		Importer: &schema.ResourceImporter{
 			StateContext: resourceImport,
 		},
@@ -121,10 +122,20 @@ func Resource() *schema.Resource {
 					Type: schema.TypeString,
 				},
 			},
+			"delete_on_create_timeout": { // Don't use Default: true to avoid unplanned changes when upgrading from previous versions.
+				Type:        schema.TypeBool,
+				Optional:    true,
+				ForceNew:    true,
+				Description: "Flag that indicates whether to delete the resource if creation times out. Default is true.",
+			},
 		},
 	}
 }
 
+const (
+	oneMinute = 1 * time.Minute
+)
+
 func resourceCreate(ctx context.Context, d *schema.ResourceData, meta any) diag.Diagnostics {
 	connV2 := meta.(*config.MongoDBClient).AtlasV2
 	groupID := d.Get("project_id").(string)
@@ -155,12 +166,20 @@ func resourceCreate(ctx context.Context, d *schema.ResourceData, meta any) diag.
 		Target:     []string{"completed", "failed"},
 		Refresh:    resourceRefreshFunc(ctx, requestParams, connV2),
 		Timeout:    d.Timeout(schema.TimeoutCreate) - time.Minute,
-		MinTimeout: 60 * time.Second,
-		Delay:      1 * time.Minute,
+		MinTimeout: oneMinute,
+		Delay:      oneMinute,
 	}
-	_, err = stateConf.WaitForStateContext(ctx)
-	if err != nil {
-		return diag.FromErr(err)
+	_, errWait := stateConf.WaitForStateContext(ctx)
+	deleteOnCreateTimeout := true // default value when not set
+	if v, ok := d.GetOkExists("delete_on_create_timeout"); ok {
+		deleteOnCreateTimeout = v.(bool)
+	}
+	errWait = cleanup.HandleCreateTimeout(deleteOnCreateTimeout, errWait, func(ctxCleanup context.Context) error {
+		_, errCleanup := connV2.CloudBackupsApi.DeleteReplicaSetBackup(ctxCleanup, groupID, clusterName, snapshot.GetId()).Execute()
+		return errCleanup
+	})
+	if errWait != nil {
+		return diag.Errorf("error creating a snapshot: %s", errWait)
 	}
 
 	d.SetId(conversion.EncodeStateID(map[string]string{

internal/service/cloudbackupsnapshot/model_cloud_backup_snapshot.go renamed to internal/service/cloudbackupsnapshot/model.go

@@ -3,6 +3,7 @@ package cloudbackupsnapshot_test
 import (
 	"context"
 	"fmt"
+	"regexp"
 	"testing"
 
 	"github.com/hashicorp/terraform-plugin-testing/helper/resource"
@@ -99,6 +100,26 @@ func TestAccBackupRSCloudBackupSnapshot_sharded(t *testing.T) {
 	})
 }
 
+func TestAccBackupRSCloudBackupSnapshot_deleteOnCreateTimeout(t *testing.T) {
+	var (
+		clusterInfo     = acc.GetClusterInfo(t, &acc.ClusterRequest{CloudBackup: true})
+		description     = "Timeout test snapshot"
+		retentionInDays = "1"
+	)
+
+	resource.ParallelTest(t, resource.TestCase{
+		PreCheck:                 acc.PreCheckBasicSleep(t, &clusterInfo, "", ""),
+		ProtoV6ProviderFactories: acc.TestAccProviderV6Factories,
+		CheckDestroy:             checkDestroy,
+		Steps: []resource.TestStep{
+			{
+				Config:      configCreateTimeoutAndDeleteOnCreateTimeout(&clusterInfo, description, retentionInDays),
+				ExpectError: regexp.MustCompile("will run cleanup because delete_on_create_timeout is true"),
+			},
+		},
+	})
+}
+
 func checkExists(resourceName string) resource.TestCheckFunc {
 	return func(s *terraform.State) error {
 		rs, ok := s.RootModule().Resources[resourceName]
@@ -220,3 +241,19 @@ func configSharded(projectID, clusterName, description, retentionInDays string)
 
 	`, projectID, clusterName, description, retentionInDays)
 }
+
+func configCreateTimeoutAndDeleteOnCreateTimeout(info *acc.ClusterInfo, description, retentionInDays string) string {
+	return info.TerraformStr + fmt.Sprintf(`
+		resource "mongodbatlas_cloud_backup_snapshot" "test" {
+			cluster_name     = %[1]s
+			project_id       = %[2]q
+			description       = %[3]q
+			retention_in_days = %[4]q
+			delete_on_create_timeout = true 
+			
+			timeouts {
+				create = "10s"
+			}
+		}
+	`, info.TerraformNameRef, info.ProjectID, description, retentionInDays)
+}