deployment_util.go

/*
Copyright 2016 The Kubernetes Authors.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

This file was copied and modified from the kubernetes/kubernetes project
https://github.com/kubernetes/kubernetes/release-1.8/pkg/controller/deployment/util/deployment_util.go

Modifications Copyright (c) 2017 SAP SE or an SAP affiliate company. All rights reserved.
*/

// Package controller is used to provide the core functionalities of machine-controller-manager
package controller

import (
	"fmt"
	"reflect"
	"sort"
	"strconv"
	"strings"
	"time"

	"github.com/golang/glog"

	common "github.com/gardener/machine-controller-manager/pkg/apis/cluster/common"
	"github.com/gardener/machine-controller-manager/pkg/apis/cluster/v1alpha1"
	v1alpha1client "github.com/gardener/machine-controller-manager/pkg/client/clientset/versioned/typed/cluster/v1alpha1"
	labelsutil "github.com/gardener/machine-controller-manager/pkg/util/labels"
	"k8s.io/api/core/v1"
	apiequality "k8s.io/apimachinery/pkg/api/equality"
	"k8s.io/apimachinery/pkg/api/meta"
	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
	"k8s.io/apimachinery/pkg/labels"
	"k8s.io/apimachinery/pkg/runtime"
	"k8s.io/apimachinery/pkg/types"
	"k8s.io/apimachinery/pkg/util/errors"
	intstrutil "k8s.io/apimachinery/pkg/util/intstr"
	"k8s.io/apimachinery/pkg/util/wait"
	"k8s.io/client-go/util/integer"

	v1alpha1listers "github.com/gardener/machine-controller-manager/pkg/client/listers/cluster/v1alpha1"
)

// MachineDeploymentListerExpansion allows custom methods to be added to MachineDeploymentLister.
type MachineDeploymentListerExpansion interface {
	GetMachineDeploymentsForMachineSet(is *v1alpha1.MachineSet) ([]*v1alpha1.MachineDeployment, error)
}

// MachineDeploymentNamespaceListerExpansion allows custom methods to be added to MachineDeploymentNamespaceLister.
type MachineDeploymentNamespaceListerExpansion interface{}

// GetMachineDeploymentsForMachineSet returns a list of Deployments that potentially
// match a MachineSet. Only the one specified in the MachineSet's ControllerRef
// will actually manage it.
// Returns an error only if no matching Deployments are found.
func (c *controller) GetMachineDeploymentsForMachineSet(is *v1alpha1.MachineSet) ([]*v1alpha1.MachineDeployment, error) {
	if len(is.Labels) == 0 {
		return nil, fmt.Errorf("no deployments found for MachineSet %v because it has no labels", is.Name)
	}

	// TODO: MODIFY THIS METHOD so that it checks for the machineTemplateSpecHash label
	dList, err := c.machineDeploymentLister.List(labels.Everything())
	if err != nil {
		return nil, err
	}

	var deployments []*v1alpha1.MachineDeployment
	for _, d := range dList {
		selector, err := metav1.LabelSelectorAsSelector(d.Spec.Selector)
		if err != nil {
			return nil, fmt.Errorf("invalid label selector: %v", err)
		}
		// If a deployment with a nil or empty selector creeps in, it should match nothing, not everything.
		if selector.Empty() || !selector.Matches(labels.Set(is.Labels)) {
			continue
		}
		deployments = append(deployments, d)
	}

	if len(deployments) == 0 {
		return nil, fmt.Errorf("could not find deployments set for MachineSet %s with labels: %v", is.Name, is.Labels)
	}

	return deployments, nil
}

const (
	// RevisionAnnotation is the revision annotation of a deployment's machine sets which records its rollout sequence
	RevisionAnnotation = "deployment.kubernetes.io/revision"
	// RevisionHistoryAnnotation maintains the history of all old revisions that a machine set has served for a deployment.
	RevisionHistoryAnnotation = "deployment.kubernetes.io/revision-history"
	// DesiredReplicasAnnotation is the desired replicas for a deployment recorded as an annotation
	// in its machine sets. Helps in separating scaling events from the rollout process and for
	// determining if the new machine set for a deployment is really saturated.
	DesiredReplicasAnnotation = "deployment.kubernetes.io/desired-replicas"
	// MaxReplicasAnnotation is the maximum replicas a deployment can have at a given point, which
	// is deployment.spec.replicas + maxSurge. Used by the underlying machine sets to estimate their
	// proportions in case the deployment has surge replicas.
	MaxReplicasAnnotation = "deployment.kubernetes.io/max-replicas"

	// RollbackRevisionNotFound is not found rollback event reason
	RollbackRevisionNotFound = "DeploymentRollbackRevisionNotFound"
	// RollbackTemplateUnchanged is the template unchanged rollback event reason
	RollbackTemplateUnchanged = "DeploymentRollbackTemplateUnchanged"
	// RollbackDone is the done rollback event reason
	RollbackDone = "DeploymentRollback"

	// MachineSetUpdatedReason is added in a deployment when one of its machine sets is updated as part
	// of the rollout process.
	MachineSetUpdatedReason = "MachineSetUpdated"
	// FailedISCreateReason is added in a deployment when it cannot create a new machine set.
	FailedISCreateReason = "MachineSetCreateError"
	// NewMachineSetReason is added in a deployment when it creates a new machine set.
	NewMachineSetReason = "NewMachineSetCreated"
	// FoundNewISReason is added in a deployment when it adopts an existing machine set.
	FoundNewISReason = "FoundNewMachineSet"
	// NewISAvailableReason is added in a deployment when its newest machine set is made available
	// ie. the number of new machines that have passed readiness checks and run for at least minReadySeconds
	// is at least the minimum available machines that need to run for the deployment.
	NewISAvailableReason = "NewMachineSetAvailable"
	// TimedOutReason is added in a deployment when its newest machine set fails to show any progress
	// within the given deadline (progressDeadlineSeconds).
	TimedOutReason = "ProgressDeadlineExceeded"
	// PausedMachineDeployReason is added in a deployment when it is paused. Lack of progress shouldn't be
	// estimated once a deployment is paused.
	PausedMachineDeployReason = "DeploymentPaused"
	// ResumedMachineDeployReason is added in a deployment when it is resumed. Useful for not failing accidentally
	// deployments that paused amidst a rollout and are bounded by a deadline.
	ResumedMachineDeployReason = "DeploymentResumed"

	// MinimumReplicasAvailable is added in a deployment when it has its minimum replicas required available.
	MinimumReplicasAvailable = "MinimumReplicasAvailable"
	// MinimumReplicasUnavailable is added in a deployment when it doesn't have the minimum required replicas
	// available.
	MinimumReplicasUnavailable = "MinimumReplicasUnavailable"
)

// NewMachineDeploymentCondition creates a new deployment condition.
func NewMachineDeploymentCondition(condType v1alpha1.MachineDeploymentConditionType, status v1alpha1.ConditionStatus, reason, message string) *v1alpha1.MachineDeploymentCondition {
	return &v1alpha1.MachineDeploymentCondition{
		Type:               condType,
		Status:             status,
		LastUpdateTime:     metav1.Now(),
		LastTransitionTime: metav1.Now(),
		Reason:             reason,
		Message:            message,
	}
}

// GetMachineDeploymentCondition returns the condition with the provided type.
func GetMachineDeploymentCondition(status v1alpha1.MachineDeploymentStatus, condType v1alpha1.MachineDeploymentConditionType) *v1alpha1.MachineDeploymentCondition {
	for i := range status.Conditions {
		c := status.Conditions[i]
		if c.Type == condType {
			return &c
		}
	}
	return nil
}

// GetMachineDeploymentConditionInternal returns the condition with the provided type.
// Avoiding Internal versions, use standard versions only.
// TODO: remove the duplicate
func GetMachineDeploymentConditionInternal(status v1alpha1.MachineDeploymentStatus, condType v1alpha1.MachineDeploymentConditionType) *v1alpha1.MachineDeploymentCondition {
	for i := range status.Conditions {
		c := status.Conditions[i]
		if c.Type == condType {
			return &c
		}
	}
	return nil
}

// SetMachineDeploymentCondition updates the deployment to include the provided condition. If the condition that
// we are about to add already exists and has the same status and reason then we are not going to update.
func SetMachineDeploymentCondition(status *v1alpha1.MachineDeploymentStatus, condition v1alpha1.MachineDeploymentCondition) {
	currentCond := GetMachineDeploymentCondition(*status, condition.Type)
	if currentCond != nil && currentCond.Status == condition.Status && currentCond.Reason == condition.Reason {
		return
	}
	// Do not update lastTransitionTime if the status of the condition doesn't change.
	if currentCond != nil && currentCond.Status == condition.Status {
		condition.LastTransitionTime = currentCond.LastTransitionTime
	}
	newConditions := filterOutDeploymentCondition(status.Conditions, condition.Type)
	status.Conditions = append(newConditions, condition)
}

// RemoveMachineDeploymentCondition removes the deployment condition with the provided type.
func RemoveMachineDeploymentCondition(status *v1alpha1.MachineDeploymentStatus, condType v1alpha1.MachineDeploymentConditionType) {
	status.Conditions = filterOutDeploymentCondition(status.Conditions, condType)
}

// filterOutDeploymentCondition returns a new slice of deployment conditions without conditions with the provided type.
func filterOutDeploymentCondition(conditions []v1alpha1.MachineDeploymentCondition, condType v1alpha1.MachineDeploymentConditionType) []v1alpha1.MachineDeploymentCondition {
	var newConditions []v1alpha1.MachineDeploymentCondition
	for _, c := range conditions {
		if c.Type == condType {
			continue
		}
		newConditions = append(newConditions, c)
	}
	return newConditions
}

// MachineSetToMachineDeploymentCondition converts a machine set condition into a deployment condition.
// Useful for promoting machine set failure conditions into deployments.
func MachineSetToMachineDeploymentCondition(cond v1alpha1.MachineSetCondition) v1alpha1.MachineDeploymentCondition {
	return v1alpha1.MachineDeploymentCondition{
		Type:               v1alpha1.MachineDeploymentConditionType(cond.Type),
		Status:             cond.Status,
		LastTransitionTime: cond.LastTransitionTime,
		LastUpdateTime:     cond.LastTransitionTime,
		Reason:             cond.Reason,
		Message:            cond.Message,
	}
}

// SetMachineDeploymentRevision updates the revision for a deployment.
func SetMachineDeploymentRevision(deployment *v1alpha1.MachineDeployment, revision string) bool {
	updated := false

	if deployment.Annotations == nil {
		deployment.Annotations = make(map[string]string)
	}
	if deployment.Annotations[RevisionAnnotation] != revision {
		deployment.Annotations[RevisionAnnotation] = revision
		updated = true
	}

	return updated
}

// MaxRevision finds the highest revision in the machine sets
func MaxRevision(allISs []*v1alpha1.MachineSet) int64 {
	max := int64(0)
	for _, is := range allISs {
		if v, err := Revision(is); err != nil {
			// Skip the machine sets when it failed to parse their revision information
			glog.V(4).Infof("Error: %v. Couldn't parse revision for machine set %#v, machine deployment controller will skip it when reconciling revisions.", err, is)
		} else if v > max {
			max = v
		}
	}
	return max
}

// LastRevision finds the second max revision number in all machine sets (the last revision)
func LastRevision(allISs []*v1alpha1.MachineSet) int64 {
	max, secMax := int64(0), int64(0)
	for _, is := range allISs {
		if v, err := Revision(is); err != nil {
			// Skip the machine sets when it failed to parse their revision information
			glog.V(4).Infof("Error: %v. Couldn't parse revision for machine set %#v, machine deployment controller will skip it when reconciling revisions.", err, is)
		} else if v >= max {
			secMax = max
			max = v
		} else if v > secMax {
			secMax = v
		}
	}
	return secMax
}

// Revision returns the revision number of the input object.
func Revision(obj runtime.Object) (int64, error) {
	acc, err := meta.Accessor(obj)
	if err != nil {
		return 0, err
	}
	v, ok := acc.GetAnnotations()[RevisionAnnotation]
	if !ok {
		return 0, nil
	}
	return strconv.ParseInt(v, 10, 64)
}

// SetNewMachineSetAnnotations sets new machine set's annotations appropriately by updating its revision and
// copying required deployment annotations to it; it returns true if machine set's annotation is changed.
func SetNewMachineSetAnnotations(deployment *v1alpha1.MachineDeployment, newIS *v1alpha1.MachineSet, newRevision string, exists bool) bool {
	// First, copy deployment's annotations (except for apply and revision annotations)
	annotationChanged := copyMachineDeploymentAnnotationsToMachineSet(deployment, newIS)
	// Then, update machine set's revision annotation
	if newIS.Annotations == nil {
		newIS.Annotations = make(map[string]string)
	}
	oldRevision, ok := newIS.Annotations[RevisionAnnotation]
	// The newRS's revision should be the greatest among all RSes. Usually, its revision number is newRevision (the max revision number
	// of all old RSes + 1). However, it's possible that some of the old RSes are deleted after the newRS revision being updated, and
	// newRevision becomes smaller than newRS's revision. We should only update newRS revision when it's smaller than newRevision.

	oldRevisionInt, err := strconv.ParseInt(oldRevision, 10, 64)
	if err != nil {
		if oldRevision != "" {
			glog.Warningf("Updating machine set revision OldRevision not int %s", err)
			return false
		}
		//If the RS annotation is empty then initialise it to 0
		oldRevisionInt = 0
	}
	newRevisionInt, err := strconv.ParseInt(newRevision, 10, 64)
	if err != nil {
		glog.Warningf("Updating machine set revision NewRevision not int %s", err)
		return false
	}
	if oldRevisionInt < newRevisionInt {
		newIS.Annotations[RevisionAnnotation] = newRevision
		annotationChanged = true
		glog.V(4).Infof("Updating machine set %q revision to %s", newIS.Name, newRevision)
	}
	// If a revision annotation already existed and this machine set was updated with a new revision
	// then that means we are rolling back to this machine set. We need to preserve the old revisions
	// for historical information.
	if ok && annotationChanged {
		revisionHistoryAnnotation := newIS.Annotations[RevisionHistoryAnnotation]
		oldRevisions := strings.Split(revisionHistoryAnnotation, ",")
		if len(oldRevisions[0]) == 0 {
			newIS.Annotations[RevisionHistoryAnnotation] = oldRevision
		} else {
			oldRevisions = append(oldRevisions, oldRevision)
			newIS.Annotations[RevisionHistoryAnnotation] = strings.Join(oldRevisions, ",")
		}
	}
	// If the new machine set is about to be created, we need to add replica annotations to it.
	if !exists && SetReplicasAnnotations(newIS, *(deployment.Spec.Replicas), *(deployment.Spec.Replicas)+MaxSurge(*deployment)) {
		annotationChanged = true
	}
	return annotationChanged
}

var annotationsToSkip = map[string]bool{
	v1.LastAppliedConfigAnnotation: true,
	RevisionAnnotation:             true,
	RevisionHistoryAnnotation:      true,
	DesiredReplicasAnnotation:      true,
	MaxReplicasAnnotation:          true,
}

// skipCopyAnnotation returns true if we should skip copying the annotation with the given annotation key
// TODO: How to decide which annotations should / should not be copied?
//       See https://github.com/kubernetes/kubernetes/pull/20035#issuecomment-179558615
func skipCopyAnnotation(key string) bool {
	return annotationsToSkip[key]
}

// copyDeploymentAnnotationsToMachineSet copies deployment's annotations to machine set's annotations,
// and returns true if machine set's annotation is changed.
// Note that apply and revision annotations are not copied.
func copyMachineDeploymentAnnotationsToMachineSet(deployment *v1alpha1.MachineDeployment, is *v1alpha1.MachineSet) bool {
	isAnnotationsChanged := false
	if is.Annotations == nil {
		is.Annotations = make(map[string]string)
	}
	for k, v := range deployment.Annotations {
		// newRS revision is updated automatically in getNewMachineSet, and the deployment's revision number is then updated
		// by copying its newRS revision number. We should not copy deployment's revision to its newRS, since the update of
		// deployment revision number may fail (revision becomes stale) and the revision number in newRS is more reliable.
		if skipCopyAnnotation(k) || is.Annotations[k] == v {
			continue
		}
		is.Annotations[k] = v
		isAnnotationsChanged = true
	}
	return isAnnotationsChanged
}

// SetMachineDeploymentAnnotationsTo sets deployment's annotations as given RS's annotations.
// This action should be done if and only if the deployment is rolling back to this rs.
// Note that apply and revision annotations are not changed.
func SetMachineDeploymentAnnotationsTo(deployment *v1alpha1.MachineDeployment, rollbackToIS *v1alpha1.MachineSet) {
	deployment.Annotations = getSkippedAnnotations(deployment.Annotations)
	for k, v := range rollbackToIS.Annotations {
		if !skipCopyAnnotation(k) {
			deployment.Annotations[k] = v
		}
	}
}

func getSkippedAnnotations(annotations map[string]string) map[string]string {
	skippedAnnotations := make(map[string]string)
	for k, v := range annotations {
		if skipCopyAnnotation(k) {
			skippedAnnotations[k] = v
		}
	}
	return skippedAnnotations
}

// FindActiveOrLatest returns the only active or the latest machine set in case there is at most one active
// machine set. If there are more active machine sets, then we should proportionally scale them.
func FindActiveOrLatest(newIS *v1alpha1.MachineSet, oldISs []*v1alpha1.MachineSet) *v1alpha1.MachineSet {
	if newIS == nil && len(oldISs) == 0 {
		return nil
	}

	sort.Sort(sort.Reverse(MachineSetsByCreationTimestamp(oldISs)))
	allISs := FilterActiveMachineSets(append(oldISs, newIS))

	switch len(allISs) {
	case 0:
		// If there is no active machine set then we should return the newest.
		if newIS != nil {
			return newIS
		}
		return oldISs[0]
	case 1:
		return allISs[0]
	default:
		return nil
	}
}

// GetDesiredReplicasAnnotation returns the number of desired replicas
func GetDesiredReplicasAnnotation(is *v1alpha1.MachineSet) (int32, bool) {
	return getIntFromAnnotation(is, DesiredReplicasAnnotation)
}

func getMaxReplicasAnnotation(is *v1alpha1.MachineSet) (int32, bool) {
	return getIntFromAnnotation(is, MaxReplicasAnnotation)
}

func getIntFromAnnotation(is *v1alpha1.MachineSet, annotationKey string) (int32, bool) {
	annotationValue, ok := is.Annotations[annotationKey]
	if !ok {
		return int32(0), false
	}
	intValue, err := strconv.Atoi(annotationValue)
	if err != nil {
		glog.V(2).Infof("Cannot convert the value %q with annotation key %q for the machine set %q", annotationValue, annotationKey, is.Name)
		return int32(0), false
	}
	return int32(intValue), true
}

// SetReplicasAnnotations sets the desiredReplicas and maxReplicas into the annotations
func SetReplicasAnnotations(is *v1alpha1.MachineSet, desiredReplicas, maxReplicas int32) bool {
	updated := false
	if is.Annotations == nil {
		is.Annotations = make(map[string]string)
	}
	desiredString := fmt.Sprintf("%d", desiredReplicas)
	if hasString := is.Annotations[DesiredReplicasAnnotation]; hasString != desiredString {
		is.Annotations[DesiredReplicasAnnotation] = desiredString
		updated = true
	}
	maxString := fmt.Sprintf("%d", maxReplicas)
	if hasString := is.Annotations[MaxReplicasAnnotation]; hasString != maxString {
		is.Annotations[MaxReplicasAnnotation] = maxString
		updated = true
	}
	return updated
}

// MaxUnavailable returns the maximum unavailable machines a rolling deployment can take.
func MaxUnavailable(deployment v1alpha1.MachineDeployment) int32 {
	if !IsRollingUpdate(&deployment) || *(deployment.Spec.Replicas) == 0 {
		return int32(0)
	}
	// Error caught by validation
	_, maxUnavailable, _ := ResolveFenceposts(deployment.Spec.Strategy.RollingUpdate.MaxSurge, deployment.Spec.Strategy.RollingUpdate.MaxUnavailable, *(deployment.Spec.Replicas))
	if maxUnavailable > *deployment.Spec.Replicas {
		return *deployment.Spec.Replicas
	}
	return maxUnavailable
}

// MinAvailable returns the minimum available machines of a given deployment
func MinAvailable(deployment *v1alpha1.MachineDeployment) int32 {
	if !IsRollingUpdate(deployment) {
		return int32(0)
	}
	return *(deployment.Spec.Replicas) - MaxUnavailable(*deployment)
}

// MaxSurge returns the maximum surge machines a rolling deployment can take.
func MaxSurge(deployment v1alpha1.MachineDeployment) int32 {
	if !IsRollingUpdate(&deployment) {
		return int32(0)
	}
	// Error caught by validation
	maxSurge, _, _ := ResolveFenceposts(deployment.Spec.Strategy.RollingUpdate.MaxSurge, deployment.Spec.Strategy.RollingUpdate.MaxUnavailable, *(deployment.Spec.Replicas))
	return maxSurge
}

// GetProportion will estimate the proportion for the provided machine set using 1. the current size
// of the parent deployment, 2. the replica count that needs be added on the machine sets of the
// deployment, and 3. the total replicas added in the machine sets of the deployment so far.
func GetProportion(is *v1alpha1.MachineSet, d v1alpha1.MachineDeployment, deploymentReplicasToAdd, deploymentReplicasAdded int32) int32 {
	if is == nil || *(is.Spec.Replicas) == 0 || deploymentReplicasToAdd == 0 || deploymentReplicasToAdd == deploymentReplicasAdded {
		return int32(0)
	}

	isFraction := getMachineSetFraction(*is, d)
	allowed := deploymentReplicasToAdd - deploymentReplicasAdded

	if deploymentReplicasToAdd > 0 {
		// Use the minimum between the machine set fraction and the maximum allowed replicas
		// when scaling up. This way we ensure we will not scale up more than the allowed
		// replicas we can add.
		return integer.Int32Min(isFraction, allowed)
	}
	// Use the maximum between the machine set fraction and the maximum allowed replicas
	// when scaling down. This way we ensure we will not scale down more than the allowed
	// replicas we can remove.
	return integer.Int32Max(isFraction, allowed)
}

// getMachineSetFraction estimates the fraction of replicas a machine set can have in
// 1. a scaling event during a rollout or 2. when scaling a paused deployment.
func getMachineSetFraction(is v1alpha1.MachineSet, d v1alpha1.MachineDeployment) int32 {
	// If we are scaling down to zero then the fraction of this machine set is its whole size (negative)
	if *(d.Spec.Replicas) == int32(0) {
		return -*(is.Spec.Replicas)
	}

	deploymentReplicas := *(d.Spec.Replicas) + MaxSurge(d)
	annotatedReplicas, ok := getMaxReplicasAnnotation(&is)
	if !ok {
		// If we cannot find the annotation then fallback to the current deployment size. Note that this
		// will not be an accurate proportion estimation in case other machine sets have different values
		// which means that the deployment was scaled at some point but we at least will stay in limits
		// due to the min-max comparisons in getProportion.
		annotatedReplicas = d.Status.Replicas
	}

	// We should never proportionally scale up from zero which means rs.spec.replicas and annotatedReplicas
	// will never be zero here.
	newISsize := (float64((*is.Spec.Replicas) * deploymentReplicas)) / float64(annotatedReplicas)
	return integer.RoundToInt32(newISsize) - *(is.Spec.Replicas)
}

// GetAllMachineSets returns the old and new machine sets targeted by the given Deployment. It gets MachineList and MachineSetList from client interface.
// Note that the first set of old machine sets doesn't include the ones with no machines, and the second set of old machine sets include all old machine sets.
// The third returned value is the new machine set, and it may be nil if it doesn't exist yet.
func GetAllMachineSets(deployment *v1alpha1.MachineDeployment, c v1alpha1client.ClusterV1alpha1Interface) ([]*v1alpha1.MachineSet, []*v1alpha1.MachineSet, *v1alpha1.MachineSet, error) {
	isList, err := ListMachineSets(deployment, IsListFromClient(c))
	if err != nil {
		return nil, nil, nil, err
	}
	oldISes, allOldISes := FindOldMachineSets(deployment, isList)
	newIS := FindNewMachineSet(deployment, isList)
	return oldISes, allOldISes, newIS, nil
}

// GetOldMachineSets returns the old machine sets targeted by the given Deployment; get MachineList and MachineSetList from client interface.
// Note that the first set of old machine sets doesn't include the ones with no machines, and the second set of old machine sets include all old machine sets.
func GetOldMachineSets(deployment *v1alpha1.MachineDeployment, c v1alpha1client.ClusterV1alpha1Interface) ([]*v1alpha1.MachineSet, []*v1alpha1.MachineSet, error) {
	rsList, err := ListMachineSets(deployment, IsListFromClient(c))
	if err != nil {
		return nil, nil, err
	}
	oldRSes, allOldRSes := FindOldMachineSets(deployment, rsList)
	return oldRSes, allOldRSes, nil
}

// GetNewMachineSet returns a machine set that matches the intent of the given deployment; get MachineSetList from client interface.
// Returns nil if the new machine set doesn't exist yet.
func GetNewMachineSet(deployment *v1alpha1.MachineDeployment, c v1alpha1client.ClusterV1alpha1Interface) (*v1alpha1.MachineSet, error) {
	rsList, err := ListMachineSets(deployment, IsListFromClient(c))
	if err != nil {
		return nil, err
	}
	return FindNewMachineSet(deployment, rsList), nil
}

// IsListFromClient returns an rsListFunc that wraps the given client.
func IsListFromClient(c v1alpha1client.ClusterV1alpha1Interface) IsListFunc {
	return func(namespace string, options metav1.ListOptions) ([]*v1alpha1.MachineSet, error) {
		isList, err := c.MachineSets(namespace).List(options)
		if err != nil {
			return nil, err
		}
		var ret []*v1alpha1.MachineSet
		for i := range isList.Items {
			ret = append(ret, &isList.Items[i])
		}
		return ret, err
	}
}

// IsListFunc is used to list all machineSets for a given list option
// TODO: switch this to full namespacers
type IsListFunc func(string, metav1.ListOptions) ([]*v1alpha1.MachineSet, error)

// IsListFunc is used to list all machineList for a given listOptions
type machineListFunc func(string, metav1.ListOptions) (*v1alpha1.MachineList, error)

// ListMachineSets returns a slice of RSes the given deployment targets.
// Note that this does NOT attempt to reconcile ControllerRef (adopt/orphan),
// because only the controller itself should do that.
// However, it does filter out anything whose ControllerRef doesn't match.
func ListMachineSets(deployment *v1alpha1.MachineDeployment, getISList IsListFunc) ([]*v1alpha1.MachineSet, error) {
	// TODO: Right now we list machine sets by their labels. We should list them by selector, i.e. the machine set's selector
	//       should be a superset of the deployment's selector, see https://github.com/kubernetes/kubernetes/issues/19830.
	namespace := deployment.Namespace
	selector, err := metav1.LabelSelectorAsSelector(deployment.Spec.Selector)
	if err != nil {
		return nil, err
	}
	options := metav1.ListOptions{LabelSelector: selector.String()}
	all, err := getISList(namespace, options)
	if err != nil {
		return nil, err
	}
	// Only include those whose ControllerRef matches the Deployment.
	owned := make([]*v1alpha1.MachineSet, 0, len(all))
	for _, is := range all {
		if metav1.IsControlledBy(is, deployment) {
			owned = append(owned, is)
		}
	}
	return owned, nil
}

// ListMachineSetsInternal is ListMachineSets for v1alpha1.
// TODO: Remove the duplicate when call sites are updated to ListMachineSets.
func ListMachineSetsInternal(deployment *v1alpha1.MachineDeployment, getISList func(string, metav1.ListOptions) ([]*v1alpha1.MachineSet, error)) ([]*v1alpha1.MachineSet, error) {
	namespace := deployment.Namespace
	selector, err := metav1.LabelSelectorAsSelector(deployment.Spec.Selector)
	if err != nil {
		return nil, err
	}
	options := metav1.ListOptions{LabelSelector: selector.String()}
	all, err := getISList(namespace, options)
	if err != nil {
		return nil, err
	}
	// Only include those whose ControllerRef matches the Deployment.
	filtered := make([]*v1alpha1.MachineSet, 0, len(all))
	for _, is := range all {
		if metav1.IsControlledBy(is, deployment) {
			filtered = append(filtered, is)
		}
	}
	return filtered, nil
}

// ListMachines for given machineDeployment
func ListMachines(deployment *v1alpha1.MachineDeployment, isList []*v1alpha1.MachineSet, getMachineList machineListFunc) (*v1alpha1.MachineList, error) {
	namespace := deployment.Namespace
	selector, err := metav1.LabelSelectorAsSelector(deployment.Spec.Selector)
	if err != nil {
		return nil, err
	}
	options := metav1.ListOptions{LabelSelector: selector.String()}
	all, err := getMachineList(namespace, options)
	if err != nil {
		return all, err
	}
	// Only include those whose ControllerRef points to a MachineSet that is in
	// turn owned by this Deployment.
	isMap := make(map[types.UID]bool, len(isList))
	for _, is := range isList {
		isMap[is.UID] = true
	}
	owned := &v1alpha1.MachineList{Items: make([]v1alpha1.Machine, 0, len(all.Items))}
	for i := range all.Items {
		machine := &all.Items[i]
		controllerRef := metav1.GetControllerOf(machine)
		if controllerRef != nil && isMap[controllerRef.UID] {
			owned.Items = append(owned.Items, *machine)
		}
	}
	return owned, nil
}

// EqualIgnoreHash returns true if two given machineTemplateSpec are equal, ignoring the diff in value of Labels[machine-template-hash]
// We ignore machine-template-hash because the hash result would be different upon machineTemplateSpec API changes
// (e.g. the addition of a new field will cause the hash code to change)
// Note that we assume input machineTemplateSpecs contain non-empty labels
func EqualIgnoreHash(template1, template2 *v1alpha1.MachineTemplateSpec) bool {
	t1Copy := template1.DeepCopy()
	t2Copy := template2.DeepCopy()
	// First, compare template.Labels (ignoring hash)
	labels1, labels2 := t1Copy.Labels, t2Copy.Labels
	if len(labels1) > len(labels2) {
		labels1, labels2 = labels2, labels1
	}
	// We make sure len(labels2) >= len(labels1)
	for k, v := range labels2 {
		if labels1[k] != v && k != v1alpha1.DefaultMachineDeploymentUniqueLabelKey {
			return false
		}
	}
	// Then, compare the templates without comparing their labels
	t1Copy.Labels, t2Copy.Labels = nil, nil
	return apiequality.Semantic.DeepEqual(t1Copy, t2Copy)
}

// FindNewMachineSet returns the new RS this given deployment targets (the one with the same machine template).
func FindNewMachineSet(deployment *v1alpha1.MachineDeployment, isList []*v1alpha1.MachineSet) *v1alpha1.MachineSet {
	sort.Sort(MachineSetsByCreationTimestamp(isList))
	for i := range isList {
		if EqualIgnoreHash(&isList[i].Spec.Template, &deployment.Spec.Template) {
			// In rare cases, such as after cluster upgrades, Deployment may end up with
			// having more than one new MachineSets that have the same template as its template,
			// see https://github.com/kubernetes/kubernetes/issues/40415
			// We deterministically choose the oldest new MachineSet.
			return isList[i]
		}
	}
	// new MachineSet does not exist.
	return nil
}

// FindOldMachineSets returns the old machine sets targeted by the given Deployment, with the given slice of RSes.
// Note that the first set of old machine sets doesn't include the ones with no machines, and the second set of old machine sets include all old machine sets.
func FindOldMachineSets(deployment *v1alpha1.MachineDeployment, isList []*v1alpha1.MachineSet) ([]*v1alpha1.MachineSet, []*v1alpha1.MachineSet) {
	var requiredISs []*v1alpha1.MachineSet
	var allISs []*v1alpha1.MachineSet
	newIS := FindNewMachineSet(deployment, isList)
	for _, is := range isList {
		// Filter out new machine set
		if newIS != nil && is.UID == newIS.UID {
			continue
		}
		allISs = append(allISs, is)
		if *(is.Spec.Replicas) != 0 {
			requiredISs = append(requiredISs, is)
		}
	}
	return requiredISs, allISs
}

// WaitForMachineSetUpdated polls the machine set until it is updated.
func WaitForMachineSetUpdated(c v1alpha1listers.MachineSetLister, desiredGeneration int64, namespace, name string) error {
	return wait.PollImmediate(1*time.Second, 1*time.Minute, func() (bool, error) {
		is, err := c.MachineSets(namespace).Get(name)
		if err != nil {
			return false, err
		}
		return is.Status.ObservedGeneration >= desiredGeneration, nil
	})
}

// WaitForMachinesHashPopulated polls the machine set until updated and fully labeled.
func WaitForMachinesHashPopulated(c v1alpha1listers.MachineSetLister, desiredGeneration int64, namespace, name string) error {
	return wait.PollImmediate(1*time.Second, 1*time.Minute, func() (bool, error) {
		is, err := c.MachineSets(namespace).Get(name)
		if err != nil {
			return false, err
		}
		return is.Status.ObservedGeneration >= desiredGeneration &&
			is.Status.FullyLabeledReplicas == *(is.Spec.Replicas), nil
	})
}

// LabelMachinesWithHash labels all machines in the given machineList with the new hash label.
func LabelMachinesWithHash(machineList *v1alpha1.MachineList, c v1alpha1client.ClusterV1alpha1Interface, machineLister v1alpha1listers.MachineLister, namespace, name, hash string) error {
	for _, machine := range machineList.Items {
		// Ignore inactive Machines.
		if !IsMachineActive(&machine) {
			continue
		}
		// Only label the machine that doesn't already have the new hash
		if machine.Labels[v1alpha1.DefaultMachineDeploymentUniqueLabelKey] != hash {
			_, err := UpdateMachineWithRetries(c.Machines(machine.Namespace), machineLister, machine.Namespace, machine.Name,
				func(machineToUpdate *v1alpha1.Machine) error {
					// Precondition: the machine doesn't contain the new hash in its label.
					if machineToUpdate.Labels[v1alpha1.DefaultMachineDeploymentUniqueLabelKey] == hash {
						return errors.ErrPreconditionViolated
					}
					machineToUpdate.Labels = labelsutil.AddLabel(machineToUpdate.Labels, v1alpha1.DefaultMachineDeploymentUniqueLabelKey, hash)
					return nil
				})
			if err != nil {
				return fmt.Errorf("error in adding template hash label %s to machine %q: %v", hash, machine.Name, err)
			}
			glog.V(4).Infof("Labeled machine %s/%s of MachineSet %s/%s with hash %s.", machine.Namespace, machine.Name, namespace, name, hash)
		}
	}
	return nil
}

// SetFromMachineSetTemplate sets the desired MachineTemplateSpec from a machine set template to the given deployment.
func SetFromMachineSetTemplate(deployment *v1alpha1.MachineDeployment, template v1alpha1.MachineTemplateSpec) *v1alpha1.MachineDeployment {
	deployment.Spec.Template.ObjectMeta = template.ObjectMeta
	deployment.Spec.Template.Spec = template.Spec
	deployment.Spec.Template.ObjectMeta.Labels = labelsutil.CloneAndRemoveLabel(
		deployment.Spec.Template.ObjectMeta.Labels,
		v1alpha1.DefaultMachineDeploymentUniqueLabelKey)
	return deployment
}

// GetReplicaCountForMachineSets returns the sum of Replicas of the given machine sets.
func GetReplicaCountForMachineSets(MachineSets []*v1alpha1.MachineSet) int32 {
	totalReplicas := int32(0)
	for _, is := range MachineSets {
		if is != nil {
			totalReplicas += *(is.Spec.Replicas)
		}
	}
	return totalReplicas
}

// GetActualReplicaCountForMachineSets returns the sum of actual replicas of the given machine sets.
func GetActualReplicaCountForMachineSets(MachineSets []*v1alpha1.MachineSet) int32 {
	totalActualReplicas := int32(0)
	for _, is := range MachineSets {
		if is != nil {
			totalActualReplicas += is.Status.Replicas
		}
	}
	return totalActualReplicas
}

// GetReadyReplicaCountForMachineSets returns the number of ready machines corresponding to the given machine sets.
func GetReadyReplicaCountForMachineSets(MachineSets []*v1alpha1.MachineSet) int32 {
	totalReadyReplicas := int32(0)
	for _, is := range MachineSets {
		if is != nil {
			totalReadyReplicas += is.Status.ReadyReplicas
		}
	}
	return totalReadyReplicas
}

// GetAvailableReplicaCountForMachineSets returns the number of available machines corresponding to the given machine sets.
func GetAvailableReplicaCountForMachineSets(MachineSets []*v1alpha1.MachineSet) int32 {
	totalAvailableReplicas := int32(0)
	for _, is := range MachineSets {
		if is != nil {
			totalAvailableReplicas += is.Status.AvailableReplicas
		}
	}
	return totalAvailableReplicas
}

// IsRollingUpdate returns true if the strategy type is a rolling update.
func IsRollingUpdate(deployment *v1alpha1.MachineDeployment) bool {
	return deployment.Spec.Strategy.Type == common.RollingUpdateMachineDeploymentStrategyType
}

// MachineDeploymentComplete considers a deployment to be complete once all of its desired replicas
// are updated and available, and no old machines are running.
func MachineDeploymentComplete(deployment *v1alpha1.MachineDeployment, newStatus *v1alpha1.MachineDeploymentStatus) bool {
	return newStatus.UpdatedReplicas == *(deployment.Spec.Replicas) &&
		newStatus.Replicas == *(deployment.Spec.Replicas) &&
		newStatus.AvailableReplicas == *(deployment.Spec.Replicas) &&
		newStatus.ObservedGeneration >= deployment.Generation
}

// MachineDeploymentProgressing reports progress for a deployment. Progress is estimated by comparing the
// current with the new status of the deployment that the controller is observing. More specifically,
// when new machines are scaled up or become ready or available, or old machines are scaled down, then we
// consider the deployment is progressing.
func MachineDeploymentProgressing(deployment *v1alpha1.MachineDeployment, newStatus *v1alpha1.MachineDeploymentStatus) bool {
	oldStatus := deployment.Status

	// Old replicas that need to be scaled down
	oldStatusOldReplicas := oldStatus.Replicas - oldStatus.UpdatedReplicas
	newStatusOldReplicas := newStatus.Replicas - newStatus.UpdatedReplicas

	return (newStatus.UpdatedReplicas > oldStatus.UpdatedReplicas) ||
		(newStatusOldReplicas < oldStatusOldReplicas) ||
		newStatus.ReadyReplicas > deployment.Status.ReadyReplicas ||
		newStatus.AvailableReplicas > deployment.Status.AvailableReplicas
}

// used for unit testing
var nowFn = func() time.Time { return time.Now() }

// MachineDeploymentTimedOut considers a deployment to have timed out once its condition that reports progress
// is older than progressDeadlineSeconds or a Progressing condition with a TimedOutReason reason already
// exists.
func MachineDeploymentTimedOut(deployment *v1alpha1.MachineDeployment, newStatus *v1alpha1.MachineDeploymentStatus) bool {
	if deployment.Spec.ProgressDeadlineSeconds == nil {
		return false
	}

	// Look for the Progressing condition. If it doesn't exist, we have no base to estimate progress.
	// If it's already set with a TimedOutReason reason, we have already timed out, no need to check
	// again.
	condition := GetMachineDeploymentCondition(*newStatus, v1alpha1.MachineDeploymentProgressing)
	if condition == nil {
		return false
	}
	// If the previous condition has been a successful rollout then we shouldn't try to
	// estimate any progress. Scenario:
	//
	// * progressDeadlineSeconds is smaller than the difference between now and the time
	//   the last rollout finished in the past.
	// * the creation of a new MachineSet triggers a resync of the Deployment prior to the
	//   cached copy of the Deployment getting updated with the status.condition that indicates
	//   the creation of the new MachineSet.
	//
	// The Deployment will be resynced and eventually its Progressing condition will catch
	// up with the state of the world.
	if condition.Reason == NewISAvailableReason {
		return false
	}
	if condition.Reason == TimedOutReason {
		return true
	}

	// Look at the difference in seconds between now and the last time we reported any
	// progress or tried to create a machine set, or resumed a paused deployment and
	// compare against progressDeadlineSeconds.
	from := condition.LastUpdateTime
	now := nowFn()
	delta := time.Duration(*deployment.Spec.ProgressDeadlineSeconds) * time.Second
	timedOut := from.Add(delta).Before(now)

	glog.V(4).Infof("MachineDeployment %q timed out (%t) [last progress check: %v - now: %v]", deployment.Name, timedOut, from, now)
	return timedOut
}

// NewISNewReplicas calculates the number of replicas a deployment's new IS should have.
// When one of the followings is true, we're rolling out the deployment; otherwise, we're scaling it.
// 1) The new RS is saturated: newRS's replicas == deployment's replicas
// 2) Max number of machines allowed is reached: deployment's replicas + maxSurge == all RSs' replicas
func NewISNewReplicas(deployment *v1alpha1.MachineDeployment, allISs []*v1alpha1.MachineSet, newIS *v1alpha1.MachineSet) (int32, error) {
	switch deployment.Spec.Strategy.Type {
	case common.RollingUpdateMachineDeploymentStrategyType:
		// Check if we can scale up.
		maxSurge, err := intstrutil.GetValueFromIntOrPercent(deployment.Spec.Strategy.RollingUpdate.MaxSurge, int(*(deployment.Spec.Replicas)), true)
		if err != nil {
			return 0, err
		}
		// Find the total number of machines
		currentMachineCount := GetReplicaCountForMachineSets(allISs)
		maxTotalMachines := *(deployment.Spec.Replicas) + int32(maxSurge)
		if currentMachineCount >= maxTotalMachines {
			// Cannot scale up.
			return *(newIS.Spec.Replicas), nil
		}
		// Scale up.
		scaleUpCount := maxTotalMachines - currentMachineCount
		// Do not exceed the number of desired replicas.
		scaleUpCount = int32(integer.IntMin(int(scaleUpCount), int(*(deployment.Spec.Replicas)-*(newIS.Spec.Replicas))))
		return *(newIS.Spec.Replicas) + scaleUpCount, nil
	case common.RecreateMachineDeploymentStrategyType:
		return *(deployment.Spec.Replicas), nil
	default:
		return 0, fmt.Errorf("machine deployment type %v isn't supported", deployment.Spec.Strategy.Type)
	}
}

// IsSaturated checks if the new machine set is saturated by comparing its size with its deployment size.
// Both the deployment and the machine set have to believe this machine set can own all of the desired
// replicas in the deployment and the annotation helps in achieving that. All machines of the MachineSet
// need to be available.
func IsSaturated(deployment *v1alpha1.MachineDeployment, is *v1alpha1.MachineSet) bool {
	if is == nil {
		return false
	}
	desiredString := is.Annotations[DesiredReplicasAnnotation]
	desired, err := strconv.Atoi(desiredString)
	if err != nil {
		return false
	}
	return *(is.Spec.Replicas) == *(deployment.Spec.Replicas) &&
		int32(desired) == *(deployment.Spec.Replicas) &&
		is.Status.AvailableReplicas == *(deployment.Spec.Replicas)
}

// WaitForObservedMachineDeployment polls for deployment to be updated so that deployment.Status.ObservedGeneration >= desiredGeneration.
// Returns error if polling timesout.
func WaitForObservedMachineDeployment(getDeploymentFunc func() (*v1alpha1.MachineDeployment, error), desiredGeneration int64, interval, timeout time.Duration) error {
	// TODO: This should take clientset.Interface when all code is updated to use clientset. Keeping it this way allows the function to be used by callers who have client.Interface.
	return wait.PollImmediate(interval, timeout, func() (bool, error) {
		deployment, err := getDeploymentFunc()
		if err != nil {
			return false, err
		}
		return deployment.Status.ObservedGeneration >= desiredGeneration, nil
	})
}

// WaitForObservedDeploymentInternal polls for deployment to be updated so that deployment.Status.ObservedGeneration >= desiredGeneration.
// Returns error if polling timesout.
// TODO: remove the duplicate
func WaitForObservedDeploymentInternal(getDeploymentFunc func() (*v1alpha1.MachineDeployment, error), desiredGeneration int64, interval, timeout time.Duration) error {
	return wait.Poll(interval, timeout, func() (bool, error) {
		deployment, err := getDeploymentFunc()
		if err != nil {
			return false, err
		}
		return deployment.Status.ObservedGeneration >= desiredGeneration, nil
	})
}

// ResolveFenceposts resolves both maxSurge and maxUnavailable. This needs to happen in one
// step. For example:
//
// 2 desired, max unavailable 1%, surge 0% - should scale old(-1), then new(+1), then old(-1), then new(+1)
// 1 desired, max unavailable 1%, surge 0% - should scale old(-1), then new(+1)
// 2 desired, max unavailable 25%, surge 1% - should scale new(+1), then old(-1), then new(+1), then old(-1)
// 1 desired, max unavailable 25%, surge 1% - should scale new(+1), then old(-1)
// 2 desired, max unavailable 0%, surge 1% - should scale new(+1), then old(-1), then new(+1), then old(-1)
// 1 desired, max unavailable 0%, surge 1% - should scale new(+1), then old(-1)
func ResolveFenceposts(maxSurge, maxUnavailable *intstrutil.IntOrString, desired int32) (int32, int32, error) {
	surge, err := intstrutil.GetValueFromIntOrPercent(maxSurge, int(desired), true)
	if err != nil {
		return 0, 0, err
	}
	unavailable, err := intstrutil.GetValueFromIntOrPercent(maxUnavailable, int(desired), false)
	if err != nil {
		return 0, 0, err
	}

	if surge == 0 && unavailable == 0 {
		// Validation should never allow the user to explicitly use zero values for both maxSurge
		// maxUnavailable. Due to rounding down maxUnavailable though, it may resolve to zero.
		// If both fenceposts resolve to zero, then we should set maxUnavailable to 1 on the
		// theory that surge might not work due to quota.
		unavailable = 1
	}

	return int32(surge), int32(unavailable), nil
}

// statusUpdateRequired checks for if status update is required comparing two MachineDeployment statuses
func statusUpdateRequired(old v1alpha1.MachineDeploymentStatus, new v1alpha1.MachineDeploymentStatus) bool {

	if old.AvailableReplicas == new.AvailableReplicas &&
		old.CollisionCount == new.CollisionCount &&
		len(old.FailedMachines) == len(new.FailedMachines) &&
		old.ObservedGeneration == new.ObservedGeneration &&
		old.ReadyReplicas == new.ReadyReplicas &&
		old.Replicas == new.Replicas &&
		old.UpdatedReplicas == new.UpdatedReplicas &&
		reflect.DeepEqual(old.Conditions, new.Conditions) {
		// If all conditions are matching

		// Iterate through all new failed machines and check if there
		// exists an older machine with same name and description
		for _, newMachine := range new.FailedMachines {
			found := false

			for _, oldMachine := range old.FailedMachines {
				if oldMachine.Name == newMachine.Name &&
					oldMachine.LastOperation.Description == newMachine.LastOperation.Description {
					found = true
					continue
				}
			}

			if !found {
				return true
			}
		}

		return false
	}

	return true
}