Fix cluster status update logic (#911)

Previously, several actors would cancel their context when successful.
This was a relic of the way the reconciliation loop was previously
structured, but now leads to a bug where cluster status is not updated
upon successful action. To address this, this patch deletes the
CancelLoop function entirely.

Author: davidwding

CHANGELOG.md

@@ -7,6 +7,10 @@ to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 # [Unreleased](https://github.com/cockroachdb/cockroach-operator/compare/v2.7.0...master)
 
+## Fixed
+
+* Delete the CancelLoop function, fixing a cluster status update bug
+
 # [v2.7.0](https://github.com/cockroachdb/cockroach-operator/compare/v2.6.0...v2.7.0)
 
 ## Fixed

pkg/actor/BUILD.bazel

@@ -5,7 +5,6 @@ go_library(
     srcs = [
         "actor.go",
         "cluster_restart.go",
-        "context.go",
         "decommission.go",
         "deploy.go",
         "director.go",

pkg/actor/cluster_restart.go

@@ -125,7 +125,6 @@ func (r *clusterRestart) Act(ctx context.Context, cluster *resource.Cluster, log
 		log.Error(err, "failed reseting the restart cluster field")
 	}
 	log.V(DEBUGLEVEL).Info("completed cluster restart")
-	CancelLoop(ctx, log)
 	return nil
 }
 

pkg/actor/context.go

@@ -19,6 +19,7 @@ package actor
 import (
 	"context"
 	"fmt"
+
 	api "github.com/cockroachdb/cockroach-operator/apis/v1alpha1"
 	"github.com/cockroachdb/cockroach-operator/pkg/clustersql"
 	"github.com/cockroachdb/cockroach-operator/pkg/database"
@@ -143,12 +144,10 @@ func (d decommission) Act(ctx context.Context, cluster *resource.Cluster, log lo
 		/// now check if the decommissionStaleErr and update status
 		log.Error(err, "decommission failed")
 		cluster.SetFalse(api.DecommissionCondition)
-		CancelLoop(ctx, log)
 		return err
 	}
 	// TO DO @alina we will need to save the status foreach action
 	cluster.SetTrue(api.DecommissionCondition)
 	log.V(DEBUGLEVEL).Info("decommission completed", "cond", ss.Status.Conditions)
-	CancelLoop(ctx, log)
 	return nil
 }

pkg/actor/decommission.go

@@ -81,7 +81,6 @@ func (d deploy) Act(ctx context.Context, cluster *resource.Cluster, log logr.Log
 
 		if changed {
 			log.Info("created/updated a resource, stopping request processing", "resource", b.ResourceName())
-			CancelLoop(ctx, log)
 			return nil
 		}
 	}

pkg/actor/deploy.go

@@ -81,7 +81,7 @@ func TestDeploysNotInitalizedClusterAfterVersionChecker(t *testing.T) {
 	// 3 is the number of resources we expect to be created. The action should be repeated as it is
 	// restarted on successful creation or update
 	for i := 0; i < 3; i++ {
-		assert.NoError(t, deploy.Act(actor.ContextWithCancelFn(context.TODO(), func() {}), cluster, testLog))
+		assert.NoError(t, deploy.Act(context.Background(), cluster, testLog))
 	}
 
 	assert.Equal(t, expected, actual)

pkg/actor/deploy_test.go

@@ -19,7 +19,6 @@ package actor
 import (
 	"context"
 	"fmt"
-	"github.com/go-logr/logr"
 	"os"
 	"strings"
 	"time"
@@ -31,6 +30,7 @@ import (
 	"github.com/cockroachdb/cockroach-operator/pkg/resource"
 	"github.com/cockroachdb/cockroach-operator/pkg/update"
 	"github.com/cockroachdb/errors"
+	"github.com/go-logr/logr"
 	"go.uber.org/zap/zapcore"
 	appsv1 "k8s.io/api/apps/v1"
 	kubetypes "k8s.io/apimachinery/pkg/types"
@@ -213,7 +213,6 @@ func (up *partitionedUpdate) Act(ctx context.Context, cluster *resource.Cluster,
 
 	// TODO set status that we are completed.
 	log.V(DEBUGLEVEL).Info("update completed with partitioned update", "new version", versionWantedCalFmtStr)
-	CancelLoop(ctx, log)
 	return nil
 }
 

pkg/actor/partitioned_update.go

@@ -19,14 +19,14 @@ package actor
 import (
 	"context"
 	"fmt"
-	"github.com/go-logr/logr"
 	"time"
 
 	"github.com/cenkalti/backoff"
 	api "github.com/cockroachdb/cockroach-operator/apis/v1alpha1"
 	"github.com/cockroachdb/cockroach-operator/pkg/kube"
 	"github.com/cockroachdb/cockroach-operator/pkg/resource"
 	"github.com/cockroachdb/errors"
+	"github.com/go-logr/logr"
 	appsv1 "k8s.io/api/apps/v1"
 	v1 "k8s.io/api/core/v1"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
@@ -140,8 +140,6 @@ func (rp *resizePVC) Act(ctx context.Context, cluster *resource.Cluster, log log
 		}*/
 
 	log.Info("PVC resize completed")
-	CancelLoop(ctx, log)
-
 	return nil
 }
 

pkg/actor/resize_pvc.go

@@ -136,7 +136,6 @@ func (v *versionChecker) Act(ctx context.Context, cluster *resource.Cluster, log
 
 	if changed {
 		log.V(int(zapcore.DebugLevel)).Info("created/updated job, stopping request processing")
-		CancelLoop(ctx, log)
 		return nil
 	}
 
@@ -349,7 +348,6 @@ func (v *versionChecker) completeVersionChecker(
 	}
 	log.V(int(zapcore.DebugLevel)).Info("completed version checker", "calVersion", version,
 		"containerImage", imageName)
-	CancelLoop(ctx, log)
 	return nil
 }
 

pkg/actor/validate_version.go

@@ -49,7 +49,6 @@ go_test(
         "@io_k8s_apimachinery//pkg/types:go_default_library",
         "@io_k8s_sigs_controller_runtime//:go_default_library",
         "@io_k8s_sigs_controller_runtime//pkg/client/fake:go_default_library",
-        "@io_k8s_sigs_controller_runtime//pkg/log:go_default_library",
         "@org_uber_go_zap//zaptest:go_default_library",
     ],
 )

pkg/controller/BUILD.bazel

@@ -147,8 +147,7 @@ func (r *ClusterReconciler) Reconcile(ctx context.Context, req reconcile.Request
 		return noRequeue()
 	}
 
-	// Save context cancellation function for actors to call if needed
-	ctx = actor.ContextWithCancelFn(ctx, cancel)
+	ctx = context.Background()
 
 	log.Info(fmt.Sprintf("Running action with name: %s", actorToExecute.GetActionType()))
 	if err := actorToExecute.Act(ctx, &cluster, log); err != nil {
@@ -187,13 +186,6 @@ func (r *ClusterReconciler) Reconcile(ctx context.Context, req reconcile.Request
 		cluster.SetActionFinished(actorToExecute.GetActionType())
 	}
 
-	// Stop processing and wait for Kubernetes scheduler to call us again as the actor
-	// modified actorToExecute resource owned by the controller
-	if cancelled(ctx) {
-		log.V(int(zapcore.InfoLevel)).Info("request was interrupted")
-		return noRequeue()
-	}
-
 	// Check if the resource has been updated while the controller worked on it
 	fresh, err := cluster.IsFresh(fetcher)
 	if err != nil {
@@ -254,12 +246,3 @@ func InitClusterReconcilerWithLogger(l logr.Logger) func(ctrl.Manager) error {
 		}).SetupWithManager(mgr)
 	}
 }
-
-func cancelled(ctx context.Context) bool {
-	select {
-	case <-ctx.Done():
-		return true
-	default:
-		return false
-	}
-}

pkg/controller/cluster_controller.go

@@ -23,7 +23,6 @@ import (
 	"time"
 
 	"github.com/go-logr/logr"
-	"sigs.k8s.io/controller-runtime/pkg/log"
 
 	api "github.com/cockroachdb/cockroach-operator/apis/v1alpha1"
 	"github.com/cockroachdb/cockroach-operator/pkg/actor"
@@ -43,14 +42,10 @@ import (
 )
 
 type fakeActor struct {
-	cancelCtx bool
-	err       error
+	err error
 }
 
 func (a *fakeActor) Act(ctx context.Context, _ *resource.Cluster, logger logr.Logger) error {
-	if a.cancelCtx {
-		actor.CancelLoop(ctx, log.NullLogger{})
-	}
 	return a.err
 }
 func (a *fakeActor) GetActionType() api.ActionType {
@@ -115,14 +110,6 @@ func TestReconcile(t *testing.T) {
 			want:    ctrl.Result{Requeue: true},
 			wantErr: "",
 		},
-		{
-			name: "reconcile action cancels the context",
-			action: fakeActor{
-				cancelCtx: true,
-			},
-			want:    ctrl.Result{Requeue: false},
-			wantErr: "",
-		},
 		{
 			name: "reconcile action fails to probe expected condition",
 			action: fakeActor{