resource/aws_emr_cluster: Fix aws_emr_security_configuration destroy issues

.changelog/12578.txt

@@ -0,0 +1,3 @@
+```release-note:bug
+resource/aws_emr_cluster: Wait for the cluster to reach a terminated state on deletion
+```

internal/service/emr/cluster.go

@@ -972,7 +972,7 @@ func resourceClusterCreate(d *schema.ResourceData, meta interface{}) error {
 	cluster, err := waitClusterCreated(conn, d.Id())
 
 	if err != nil {
-		return fmt.Errorf("error waiting for EMR Cluster (%s) to be created: %w", d.Id(), err)
+		return fmt.Errorf("error waiting for EMR Cluster (%s) to create: %w", d.Id(), err)
 	}
 
 	// For multiple master nodes, EMR automatically enables
@@ -1377,90 +1377,27 @@ func resourceClusterUpdate(d *schema.ResourceData, meta interface{}) error {
 func resourceClusterDelete(d *schema.ResourceData, meta interface{}) error {
 	conn := meta.(*conns.AWSClient).EMRConn
 
-	req := &emr.TerminateJobFlowsInput{
+	log.Printf("[DEBUG] Deleting EMR Cluster: (%s)", d.Id())
+	_, err := conn.TerminateJobFlows(&emr.TerminateJobFlowsInput{
 		JobFlowIds: []*string{
 			aws.String(d.Id()),
 		},
-	}
-
-	_, err := conn.TerminateJobFlows(req)
-	if err != nil {
-		log.Printf("[ERROR], %s", err)
-		return err
-	}
-
-	input := &emr.ListInstancesInput{
-		ClusterId: aws.String(d.Id()),
-	}
-	var resp *emr.ListInstancesOutput
-	var count int
-	err = resource.Retry(20*time.Minute, func() *resource.RetryError {
-		var err error
-		resp, err = conn.ListInstances(input)
-
-		if err != nil {
-			return resource.NonRetryableError(err)
-		}
-
-		count = CountRemainingInstances(resp, d.Id())
-		if count != 0 {
-			return resource.RetryableError(fmt.Errorf("EMR Cluster (%s) has (%d) Instances remaining", d.Id(), count))
-		}
-		return nil
 	})
 
-	if tfresource.TimedOut(err) {
-		resp, err = conn.ListInstances(input)
-
-		if err == nil {
-			count = CountRemainingInstances(resp, d.Id())
-		}
+	if err != nil {
+		return fmt.Errorf("error terminating EMR Cluster (%s): %w", d.Id(), err)
 	}
 
-	if count != 0 {
-		return fmt.Errorf("EMR Cluster (%s) has (%d) Instances remaining", d.Id(), count)
-	}
+	log.Println("[INFO] Waiting for EMR Cluster to be terminated")
+	_, err = waitClusterDeleted(conn, d.Id())
 
 	if err != nil {
-		return fmt.Errorf("error waiting for EMR Cluster (%s) Instances to drain: %s", d.Id(), err)
+		return fmt.Errorf("error waiting for EMR Cluster (%s) to delete: %w", d.Id(), err)
 	}
 
 	return nil
 }
 
-func CountRemainingInstances(resp *emr.ListInstancesOutput, emrClusterId string) int {
-	if resp == nil {
-		log.Printf("[ERROR] response is nil")
-		return 0
-	}
-
-	instanceCount := len(resp.Instances)
-
-	if instanceCount == 0 {
-		log.Printf("[DEBUG] No instances found for EMR Cluster (%s)", emrClusterId)
-		return 0
-	}
-
-	// Collect instance status states, wait for all instances to be terminated
-	// before moving on
-	var terminated []string
-	for j, i := range resp.Instances {
-		instanceId := aws.StringValue(i.Ec2InstanceId)
-		if i.Status != nil {
-			if aws.StringValue(i.Status.State) == emr.InstanceStateTerminated {
-				terminated = append(terminated, instanceId)
-			}
-		} else {
-			log.Printf("[DEBUG] Cluster instance (%d : %s) has no status", j, instanceId)
-		}
-	}
-	if len(terminated) == instanceCount {
-		log.Printf("[DEBUG] All (%d) EMR Cluster (%s) Instances terminated", instanceCount, emrClusterId)
-		return 0
-	}
-	return len(resp.Instances)
-}
-
 func expandApplications(apps []interface{}) []*emr.Application {
 	appOut := make([]*emr.Application, 0, len(apps))
 

internal/service/emr/cluster_test.go

@@ -15,6 +15,7 @@ import (
 	"github.com/hashicorp/terraform-provider-aws/internal/conns"
 	tfec2 "github.com/hashicorp/terraform-provider-aws/internal/service/ec2"
 	tfemr "github.com/hashicorp/terraform-provider-aws/internal/service/emr"
+	"github.com/hashicorp/terraform-provider-aws/internal/tfresource"
 )
 
 func TestAccEMRCluster_basic(t *testing.T) {
@@ -1545,24 +1546,17 @@ func testAccCheckDestroy(s *terraform.State) error {
 			continue
 		}
 
-		input := &emr.DescribeClusterInput{
-			ClusterId: aws.String(rs.Primary.ID),
+		_, err := tfemr.FindClusterByID(conn, rs.Primary.ID)
+
+		if tfresource.NotFound(err) {
+			continue
 		}
 
-		output, err := conn.DescribeCluster(input)
 		if err != nil {
 			return err
 		}
 
-		// if output.Cluster != nil &&
-		// 	*output.Cluster.Status.State == "WAITING" {
-		// 	return fmt.Errorf("EMR Cluster still exists")
-		// }
-		if output.Cluster == nil || output.Cluster.Status == nil || aws.StringValue(output.Cluster.Status.State) == emr.ClusterStateTerminated {
-			continue
-		}
-
-		return fmt.Errorf("EMR Cluster still exists")
+		return fmt.Errorf("EMR Cluster %s still exists", rs.Primary.ID)
 	}
 
 	return nil
@@ -1574,30 +1568,21 @@ func testAccCheckClusterExists(n string, v *emr.Cluster) resource.TestCheckFunc
 		if !ok {
 			return fmt.Errorf("Not found: %s", n)
 		}
+
 		if rs.Primary.ID == "" {
-			return fmt.Errorf("No cluster id set")
-		}
-		conn := acctest.Provider.Meta().(*conns.AWSClient).EMRConn
-		describe, err := conn.DescribeCluster(&emr.DescribeClusterInput{
-			ClusterId: aws.String(rs.Primary.ID),
-		})
-		if err != nil {
-			return fmt.Errorf("EMR error: %w", err)
+			return fmt.Errorf("No EMR Cluster ID is set")
 		}
 
-		if describe.Cluster == nil || *describe.Cluster.Id != rs.Primary.ID {
-			return fmt.Errorf("EMR cluster %q not found", rs.Primary.ID)
-		}
+		conn := acctest.Provider.Meta().(*conns.AWSClient).EMRConn
 
-		*v = *describe.Cluster
+		output, err := tfemr.FindClusterByID(conn, rs.Primary.ID)
 
-		if describe.Cluster.Status != nil {
-			state := aws.StringValue(describe.Cluster.Status.State)
-			if state != emr.ClusterStateRunning && state != emr.ClusterStateWaiting {
-				return fmt.Errorf("EMR cluster %q is not RUNNING or WAITING, currently: %s", rs.Primary.ID, state)
-			}
+		if err != nil {
+			return err
 		}
 
+		*v = *output
+
 		return nil
 	}
 }

internal/service/emr/find.go

@@ -40,6 +40,13 @@ func FindClusterByID(conn *emr.EMR, id string) (*emr.Cluster, error) {
 		return nil, err
 	}
 
+	// Eventual consistency check.
+	if aws.StringValue(output.Id) != id {
+		return nil, &resource.NotFoundError{
+			LastRequest: input,
+		}
+	}
+
 	if state := aws.StringValue(output.Status.State); state == emr.ClusterStateTerminated || state == emr.ClusterStateTerminatedWithErrors {
 		return nil, &resource.NotFoundError{
 			Message:     state,

internal/service/emr/wait.go

@@ -14,18 +14,16 @@ const (
 	ClusterCreatedTimeout    = 75 * time.Minute
 	ClusterCreatedMinTimeout = 10 * time.Second
 	ClusterCreatedDelay      = 30 * time.Second
+
+	ClusterDeletedTimeout    = 20 * time.Minute
+	ClusterDeletedMinTimeout = 10 * time.Second
+	ClusterDeletedDelay      = 30 * time.Second
 )
 
 func waitClusterCreated(conn *emr.EMR, id string) (*emr.Cluster, error) {
 	stateConf := &resource.StateChangeConf{
-		Pending: []string{
-			emr.ClusterStateBootstrapping,
-			emr.ClusterStateStarting,
-		},
-		Target: []string{
-			emr.ClusterStateRunning,
-			emr.ClusterStateWaiting,
-		},
+		Pending:    []string{emr.ClusterStateBootstrapping, emr.ClusterStateStarting},
+		Target:     []string{emr.ClusterStateRunning, emr.ClusterStateWaiting},
 		Refresh:    statusCluster(conn, id),
 		Timeout:    ClusterCreatedTimeout,
 		MinTimeout: ClusterCreatedMinTimeout,
@@ -44,3 +42,26 @@ func waitClusterCreated(conn *emr.EMR, id string) (*emr.Cluster, error) {
 
 	return nil, err
 }
+
+func waitClusterDeleted(conn *emr.EMR, id string) (*emr.Cluster, error) {
+	stateConf := &resource.StateChangeConf{
+		Pending:    []string{emr.ClusterStateTerminating},
+		Target:     []string{emr.ClusterStateTerminated, emr.ClusterStateTerminatedWithErrors},
+		Refresh:    statusCluster(conn, id),
+		Timeout:    ClusterDeletedTimeout,
+		MinTimeout: ClusterDeletedMinTimeout,
+		Delay:      ClusterDeletedDelay,
+	}
+
+	outputRaw, err := stateConf.WaitForState()
+
+	if output, ok := outputRaw.(*emr.Cluster); ok {
+		if stateChangeReason := output.Status.StateChangeReason; stateChangeReason != nil {
+			tfresource.SetLastError(err, fmt.Errorf("%s: %s", aws.StringValue(stateChangeReason.Code), aws.StringValue(stateChangeReason.Message)))
+		}
+
+		return output, err
+	}
+
+	return nil, err
+}