Merge pull request #208 from darkowlzz/add-status-to-all-crds

Add status to all crds

README.md

@@ -111,7 +111,7 @@ Parameter | Description | Default
 `secretRefName` | Reference name of storageos secret |
 `secretRefNamespace` | Namespace of storageos secret |
 `namespace` | Namespace where storageos cluster resources are created | `storageos`
-`images.nodeContainer` | StorageOS node container image | `storageos/node:1.5.0`
+`images.nodeContainer` | StorageOS node container image | `storageos/node:1.5.1`
 `images.initContainer` | StorageOS init container image | `storageos/init:1.0.0`
 `images.csiNodeDriverRegistrarContainer` | CSI Node Driver Registrar Container image | `quay.io/k8scsi/csi-node-driver-registrar:v1.0.1`
 `images.csiClusterDriverRegistrarContainer` | CSI Cluster Driver Registrar Container image | `quay.io/k8scsi/csi-cluster-driver-registrar:v1.0.1`

pkg/controller/job/job_controller.go

@@ -8,7 +8,6 @@ import (
 	"strings"
 	"time"
 
-	storageosv1 "github.com/storageos/cluster-operator/pkg/apis/storageos/v1"
 	appsv1 "k8s.io/api/apps/v1"
 	corev1 "k8s.io/api/core/v1"
 	"k8s.io/apimachinery/pkg/api/errors"
@@ -27,6 +26,8 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/reconcile"
 	logf "sigs.k8s.io/controller-runtime/pkg/runtime/log"
 	"sigs.k8s.io/controller-runtime/pkg/source"
+
+	storageosv1 "github.com/storageos/cluster-operator/pkg/apis/storageos/v1"
 )
 
 var log = logf.Log.WithName("storageos.job")
@@ -96,6 +97,10 @@ func (r *ReconcileJob) Reconcile(request reconcile.Request) (reconcile.Result, e
 	reconcilePeriod := 10 * time.Second
 	reconcileResult := reconcile.Result{RequeueAfter: reconcilePeriod}
 
+	// Return this for a immediate retry of the reconciliation loop with the
+	// same request object.
+	immediateRetryResult := reconcile.Result{Requeue: true}
+
 	// Fetch the Job instance
 	instance := &storageosv1.Job{}
 	err := r.client.Get(context.TODO(), request.NamespacedName, instance)
@@ -107,26 +112,31 @@ func (r *ReconcileJob) Reconcile(request reconcile.Request) (reconcile.Result, e
 			return reconcile.Result{}, nil
 		}
 		// Error reading the object - requeue the request.
-		return reconcileResult, err
+		return immediateRetryResult, err
 	}
 
 	// Set Spec attribute values.
 	instance.Spec.LabelSelector = instance.Spec.GetLabelSelector()
 
 	// Update the object.
 	if err := r.client.Update(context.Background(), instance); err != nil {
-		return reconcileResult, err
+		return immediateRetryResult, err
 	}
 
 	// Define a new DaemonSet object
 	daemonset, err := newDaemonSetForCR(instance)
 	if err != nil {
-		return reconcileResult, err
+		log.Info("Failed to create DaemonSet for Job", "error", err)
+		instance.Status.Completed = false
+		if updateErr := r.client.Status().Update(context.Background(), instance); updateErr != nil {
+			err = fmt.Errorf("%v, %v", updateErr, err)
+		}
+		return immediateRetryResult, err
 	}
 
 	// Set Job instance as the owner and controller
 	if err := controllerutil.SetControllerReference(instance, daemonset, r.scheme); err != nil {
-		return reconcileResult, err
+		return immediateRetryResult, err
 	}
 
 	// Check if this DaemonSet already exists
@@ -136,30 +146,34 @@ func (r *ReconcileJob) Reconcile(request reconcile.Request) (reconcile.Result, e
 		log.Info("Creating a new DaemonSet")
 		err = r.client.Create(context.TODO(), daemonset)
 		if err != nil {
-			return reconcileResult, err
+			log.Info("Failed to create DaemonSet", "error", err)
+			// Update status.
+			instance.Status.Completed = false
+			if updateErr := r.client.Status().Update(context.Background(), instance); updateErr != nil {
+				err = fmt.Errorf("%v, %v", updateErr, err)
+			}
+			return immediateRetryResult, err
 		}
 
 		// DaemonSet created successfully - don't requeue
 		return reconcile.Result{}, nil
 	} else if err != nil {
-		return reconcileResult, err
+		return immediateRetryResult, err
 	}
 
 	if instance.Status.Completed {
 		// Job completed. Do nothing - don't requeue
 		return reconcile.Result{}, nil
 	}
 
-	// Check DaemonSet Pod status.
+	// Check DaemonSet Pod status and update it.
 	completed, err := checkPods(r.clientset, instance, r.recorder)
 	if err != nil {
-		return reconcileResult, err
+		log.Info("Failed to check pod status", "error", err)
 	}
-
-	// Update the Completed status of the Job (status subresource).
 	instance.Status.Completed = completed
 	if err := r.client.Status().Update(context.Background(), instance); err != nil {
-		return reconcileResult, err
+		return immediateRetryResult, err
 	}
 
 	return reconcileResult, nil

pkg/controller/storageosupgrade/storageosupgrade_controller.go

@@ -6,7 +6,7 @@ import (
 	"strings"
 	"time"
 
-	storageosv1 "github.com/storageos/cluster-operator/pkg/apis/storageos/v1"
+	storageosapi "github.com/storageos/go-api"
 	batchv1 "k8s.io/api/batch/v1"
 	corev1 "k8s.io/api/core/v1"
 	rbacv1 "k8s.io/api/rbac/v1"
@@ -24,7 +24,7 @@ import (
 	logf "sigs.k8s.io/controller-runtime/pkg/runtime/log"
 	"sigs.k8s.io/controller-runtime/pkg/source"
 
-	storageosapi "github.com/storageos/go-api"
+	storageosv1 "github.com/storageos/cluster-operator/pkg/apis/storageos/v1"
 )
 
 var log = logf.Log.WithName("storageos.upgrade")
@@ -220,6 +220,10 @@ func (r *ReconcileStorageOSUpgrade) Reconcile(request reconcile.Request) (reconc
 	reconcilePeriod := 10 * time.Second
 	reconcileResult := reconcile.Result{RequeueAfter: reconcilePeriod}
 
+	// Return this for a immediate retry of the reconciliation loop with the
+	// same request object.
+	immediateRetryResult := reconcile.Result{Requeue: true}
+
 	// Fetch the StorageOSUpgrade instance
 	instance := &storageosv1.StorageOSUpgrade{}
 	err := r.client.Get(context.TODO(), request.NamespacedName, instance)
@@ -237,7 +241,7 @@ func (r *ReconcileStorageOSUpgrade) Reconcile(request reconcile.Request) (reconc
 			return reconcile.Result{}, nil
 		}
 		// Error reading the object - requeue the request.
-		return reconcileResult, err
+		return immediateRetryResult, err
 	}
 
 	r.SetCurrentUpgradeIfNone(instance)
@@ -256,16 +260,21 @@ func (r *ReconcileStorageOSUpgrade) Reconcile(request reconcile.Request) (reconc
 		// puller job, based on the cluster's node selector.
 		currentCluster, err := r.findCurrentCluster()
 		if err != nil {
+			log.Info("Failed to find current cluster", "error", err)
+			instance.Status.Completed = false
+			if updateErr := r.client.Status().Update(context.Background(), instance); updateErr != nil {
+				err = fmt.Errorf("%v,%v", updateErr, err)
+			}
 			// Re-queue if it fails to get the current cluster.
-			return reconcileResult, err
+			return immediateRetryResult, err
 		}
 		// Create image puller.
 		r.imagePuller = newImagePullJob(instance, currentCluster)
 		if err := controllerutil.SetControllerReference(instance, r.imagePuller, r.scheme); err != nil {
-			return reconcileResult, err
+			return immediateRetryResult, err
 		}
 		if err := r.client.Create(context.Background(), r.imagePuller); err != nil && !errors.IsAlreadyExists(err) {
-			return reconcileResult, fmt.Errorf("failed to create image puller job: %v", err)
+			return immediateRetryResult, fmt.Errorf("failed to create image puller job: %v", err)
 		}
 
 		r.recorder.Event(instance, corev1.EventTypeNormal, "PullImage", "Pulling the new container image")
@@ -285,7 +294,12 @@ func (r *ReconcileStorageOSUpgrade) Reconcile(request reconcile.Request) (reconc
 	}
 	// Re-queue if the image pull didn't complete.
 	if !r.imagePuller.Status.Completed {
-		return reconcileResult, fmt.Errorf("image pull didn't complete")
+		err := fmt.Errorf("image pull didn't complete")
+		instance.Status.Completed = false
+		if updateErr := r.client.Update(context.Background(), instance); updateErr != nil {
+			err = fmt.Errorf("%v, %v", updateErr, err)
+		}
+		return reconcileResult, err
 	}
 
 	// Find and pause the running cluster.