feat: Long-running operation improvements for mongodbatlas_online_archive resource (#3542)

* implement delete_on_create_timeout

* renames

* fix

* use v2 in delete

* date criteria

* config

Author: oarbusi

.changelog/3542.txt

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

docs/resources/online_archive.md

@@ -117,6 +117,7 @@ resource "mongodbatlas_online_archive" "test" {
 * `paused` - (Optional) State of the online archive. This is required for pausing an active online archive or resuming a paused online archive. If the collection has another active online archive, the resume request fails.
 * `sync_creation` - (Optional) Flag that indicates whether the provider will wait for the state of the online archive to reach `IDLE` or `ACTIVE` when creating an online archive. Defaults to `false`.
 * `timeouts`- (Optional) The duration of time to wait for Online Archive to be created. 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`. The default timeout for Online Archive create is `3h`. 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.
 
 ### Criteria
 

internal/service/onlinearchive/data_source_online_archive.go renamed to internal/service/onlinearchive/data_source.go

@@ -12,6 +12,7 @@ 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"
@@ -22,15 +23,16 @@ const (
 	errorOnlineArchivesCreate = "error creating MongoDB Atlas Online Archive:: %s"
 	errorOnlineArchivesDelete = "error deleting MongoDB Atlas Online Archive: %s archive_id (%s)"
 	scheduleTypeDefault       = "DEFAULT"
+	oneMinute                 = 1 * time.Minute
 )
 
 func Resource() *schema.Resource {
 	return &schema.Resource{
-		Schema:        resourceSchema(),
-		CreateContext: resourceCreate,
-		ReadContext:   resourceRead,
-		DeleteContext: resourceDelete,
-		UpdateContext: resourceUpdate,
+		Schema:               resourceSchema(),
+		CreateWithoutTimeout: resourceCreate,
+		ReadWithoutTimeout:   resourceRead,
+		UpdateWithoutTimeout: resourceUpdate,
+		DeleteWithoutTimeout: resourceDelete,
 		Importer: &schema.ResourceImporter{
 			StateContext: resourceImport,
 		},
@@ -210,6 +212,11 @@ func resourceSchema() map[string]*schema.Schema {
 			Type:     schema.TypeString,
 			Computed: true,
 		},
+		"delete_on_create_timeout": { // Don't use Default: true to avoid unplanned changes when upgrading from previous versions.
+			Type:        schema.TypeBool,
+			Optional:    true,
+			Description: "Flag that indicates whether to delete the resource if creation times out. Default is true.",
+		},
 		"sync_creation": {
 			Type:     schema.TypeBool,
 			Optional: true,
@@ -243,15 +250,23 @@ func resourceCreate(ctx context.Context, d *schema.ResourceData, meta any) diag.
 			Pending:    []string{"PENDING", "ARCHIVING", "PAUSING", "PAUSED", "ORPHANED", "REPEATING"},
 			Target:     []string{"IDLE", "ACTIVE"},
 			Refresh:    resourceOnlineRefreshFunc(ctx, projectID, clusterName, archiveID, connV2),
-			Timeout:    d.Timeout(schema.TimeoutCreate) - time.Minute, // When using a CRUD function with a timeout, any StateChangeConf timeouts must be configured below that duration to avoid returning the SDK context: deadline exceeded error instead of the retry logic error.
-			MinTimeout: 1 * time.Minute,
-			Delay:      3 * time.Minute,
+			Timeout:    d.Timeout(schema.TimeoutCreate) - oneMinute, // When using a CRUD function with a timeout, any StateChangeConf timeouts must be configured below that duration to avoid returning the SDK context: deadline exceeded error instead of the retry logic error.
+			MinTimeout: oneMinute,
+			Delay:      oneMinute,
 		}
 
 		// Wait, catching any errors
-		_, err := stateConf.WaitForStateContext(ctx)
-		if err != nil {
-			return diag.FromErr(fmt.Errorf("error updating the online archive status %s for cluster %s", clusterName, archiveID))
+		_, 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.OnlineArchiveApi.DeleteOnlineArchive(ctxCleanup, projectID, archiveID, clusterName).Execute()
+			return errCleanup
+		})
+		if errWait != nil {
+			return diag.FromErr(fmt.Errorf("error updating the online archive status %s for cluster %s: %s", clusterName, archiveID, errWait))
 		}
 	}
 
@@ -314,21 +329,21 @@ func resourceRead(ctx context.Context, d *schema.ResourceData, meta any) diag.Di
 }
 
 func resourceDelete(ctx context.Context, d *schema.ResourceData, meta any) diag.Diagnostics {
-	conn := meta.(*config.MongoDBClient).Atlas
+	connV2 := meta.(*config.MongoDBClient).AtlasV2
 	ids := conversion.DecodeStateID(d.Id())
-	atlasID := ids["archive_id"]
+	archiveID := ids["archive_id"]
 	projectID := ids["project_id"]
 	clusterName := ids["cluster_name"]
 
-	_, err := conn.OnlineArchives.Delete(ctx, projectID, clusterName, atlasID)
+	_, err := connV2.OnlineArchiveApi.DeleteOnlineArchive(ctx, projectID, archiveID, clusterName).Execute()
 
 	if err != nil {
 		alreadyDeleted := strings.Contains(err.Error(), "404") && !d.IsNewResource()
 		if alreadyDeleted {
 			return nil
 		}
 
-		return diag.FromErr(fmt.Errorf(errorOnlineArchivesDelete, err, atlasID))
+		return diag.FromErr(fmt.Errorf(errorOnlineArchivesDelete, err, archiveID))
 	}
 	return nil
 }

internal/service/onlinearchive/resource_online_archive.go renamed to internal/service/onlinearchive/resource.go

@@ -22,7 +22,7 @@ func TestMigBackupRSOnlineArchiveWithNoChangeBetweenVersions(t *testing.T) {
 	if mig.IsProviderVersionAtLeast("1.12.2") {
 		deleteExpirationDays = 7
 	}
-	config := configWithDailySchedule(clusterTerraformStr, clusterResourceName, 1, deleteExpirationDays)
+	config := configWithDailySchedule(clusterTerraformStr, clusterResourceName, 1, deleteExpirationDays, false)
 
 	resource.ParallelTest(t, resource.TestCase{
 		PreCheck:     mig.PreCheckBasicSleep(t),

internal/service/onlinearchive/resource_online_archive_migration_test.go renamed to internal/service/onlinearchive/resource_migration_test.go

@@ -42,7 +42,7 @@ func TestAccBackupRSOnlineArchive(t *testing.T) {
 				Check:  acc.PopulateWithSampleDataTestCheck(projectID, clusterName),
 			},
 			{
-				Config: configWithDailySchedule(clusterTerraformStr, clusterResourceName, 1, 7),
+				Config: configWithDailySchedule(clusterTerraformStr, clusterResourceName, 1, 7, false),
 				Check: resource.ComposeAggregateTestCheckFunc(
 					resource.TestCheckResourceAttrSet(onlineArchiveResourceName, "state"),
 					resource.TestCheckResourceAttrSet(onlineArchiveResourceName, "archive_id"),
@@ -58,7 +58,7 @@ func TestAccBackupRSOnlineArchive(t *testing.T) {
 				),
 			},
 			{
-				Config: configWithDailySchedule(clusterTerraformStr, clusterResourceName, 2, 8),
+				Config: configWithDailySchedule(clusterTerraformStr, clusterResourceName, 2, 8, false),
 				Check: resource.ComposeAggregateTestCheckFunc(
 					resource.TestCheckResourceAttrSet(onlineArchiveResourceName, "state"),
 					resource.TestCheckResourceAttrSet(onlineArchiveResourceName, "archive_id"),
@@ -146,7 +146,7 @@ func TestAccBackupRSOnlineArchiveBasic(t *testing.T) {
 				),
 			},
 			{
-				Config: configWithDailySchedule(clusterTerraformStr, clusterResourceName, 1, 1),
+				Config: configWithDailySchedule(clusterTerraformStr, clusterResourceName, 1, 1, false),
 				Check: resource.ComposeAggregateTestCheckFunc(
 					resource.TestCheckResourceAttrSet(onlineArchiveResourceName, "state"),
 					resource.TestCheckResourceAttrSet(onlineArchiveResourceName, "archive_id"),
@@ -229,7 +229,7 @@ func TestAccBackupRSOnlineArchiveInvalidProcessRegion(t *testing.T) {
 	})
 }
 
-func configWithDailySchedule(clusterTerraformStr, clusterResourceName string, startHour, deleteExpirationDays int) string {
+func configWithDailySchedule(clusterTerraformStr, clusterResourceName string, startHour, deleteExpirationDays int, deleteOnTimeout bool) string {
 	var dataExpirationRuleBlock string
 	if deleteExpirationDays > 0 {
 		dataExpirationRuleBlock = fmt.Sprintf(`
@@ -238,6 +238,15 @@ func configWithDailySchedule(clusterTerraformStr, clusterResourceName string, st
 		}
 		`, deleteExpirationDays)
 	}
+	deleteOnCreateTimeoutStr := ""
+	if deleteOnTimeout {
+		deleteOnCreateTimeoutStr = `
+		delete_on_create_timeout = true
+		timeouts {
+			create = "1s"
+		}
+		`
+	}
 
 	return fmt.Sprintf(`
 	%[1]s
@@ -281,6 +290,8 @@ func configWithDailySchedule(clusterTerraformStr, clusterResourceName string, st
 		}
 
 		sync_creation = true
+
+		%[5]s
 	}
 	
 	data "mongodbatlas_online_archive" "read_archive" {
@@ -293,7 +304,7 @@ func configWithDailySchedule(clusterTerraformStr, clusterResourceName string, st
 		project_id =  mongodbatlas_online_archive.users_archive.project_id
 		cluster_name = mongodbatlas_online_archive.users_archive.cluster_name
 	}
-	`, clusterTerraformStr, startHour, dataExpirationRuleBlock, clusterResourceName)
+	`, clusterTerraformStr, startHour, dataExpirationRuleBlock, clusterResourceName, deleteOnCreateTimeoutStr)
 }
 
 func configWithoutSchedule(clusterTerraformStr, clusterResourceName string) string {
@@ -512,3 +523,25 @@ func testAccBackupRSOnlineArchiveConfigWithMonthlySchedule(clusterTerraformStr,
 	}
 	`, clusterTerraformStr, startHour, clusterResourceName)
 }
+
+func TestAccOnlineArchive_deleteOnCreateTimeout(t *testing.T) {
+	var (
+		clusterInfo = acc.GetClusterInfo(t, clusterRequest())
+	)
+
+	resource.ParallelTest(t, resource.TestCase{
+		PreCheck:                 acc.PreCheckBasicSleep(t, &clusterInfo, "", ""),
+		ProtoV6ProviderFactories: acc.TestAccProviderV6Factories,
+		CheckDestroy:             acc.CheckDestroyCluster,
+		Steps: []resource.TestStep{
+			{
+				Config: clusterInfo.TerraformStr,
+				Check:  acc.PopulateWithSampleDataTestCheck(clusterInfo.ProjectID, clusterInfo.Name),
+			},
+			{
+				Config:      configWithDailySchedule(clusterInfo.TerraformStr, clusterInfo.ResourceName, 1, 7, true),
+				ExpectError: regexp.MustCompile("will run cleanup because delete_on_create_timeout is true"),
+			},
+		},
+	})
+}