Convert status in cluster-autoscaler-status to yaml and add error inf…

…o for backoff

Change-Id: Ic68e0d67b7ce9912b605b6c0a3356b4d0e177911

cluster-autoscaler/clusterstate/api/types.go

@@ -69,36 +69,148 @@ const (
 	ClusterAutoscalerBackoff ClusterAutoscalerConditionStatus = "Backoff"
 )
 
-// ClusterAutoscalerCondition describes some aspect of ClusterAutoscaler work.
-type ClusterAutoscalerCondition struct {
-	// Type defines the aspect that the condition describes. For example, it can be Health or ScaleUp/Down activity.
-	Type ClusterAutoscalerConditionType `json:"type,omitempty"`
-	// Status of the condition.
-	Status ClusterAutoscalerConditionStatus `json:"status,omitempty"`
-	// Message is a free text extra information about the condition. It may contain some
-	// extra debugging data, like why the cluster is unhealthy.
-	Message string `json:"message,omitempty"`
-	// Reason is a unique, one-word, CamelCase reason for the condition's last transition.
-	Reason string `json:"reason,omitempty"`
+// RegisteredUnreadyNodeCount contains node counts of registered but unready nodes.
+type RegisteredUnreadyNodeCount struct {
+	// Total number of registered but unready nodes.
+	Total int `json:"total" yaml:"total"`
+	// ResourceUnready is the number of registered but unready nodes due to a missing resource (e.g. GPU).
+	ResourceUnready int `json:"resourceUnready" yaml:"resourceUnready"`
+}
+
+// RegisteredNodeCount contains node counts of registered nodes.
+type RegisteredNodeCount struct {
+	Total      int                        `json:"total" yaml:"total"`
+	Ready      int                        `json:"ready" yaml:"ready"`
+	NotStarted int                        `json:"notStarted" yaml:"notStarted"`
+	Unready    RegisteredUnreadyNodeCount `json:"unready,omitempty" yaml:"unready,omitempty"`
+}
+
+// HealthConditionNodeCount contains number of nodes that satisfy different criteria for health condition.
+type HealthConditionNodeCount struct {
+	Registered       RegisteredNodeCount `json:"registered,omitempty" yaml:"registered,omitempty"`
+	LongUnregistered int                 `json:"longUnregistered" yaml:"longUnregistered"`
+	Unregistered     int                 `json:"unregistered" yaml:"unregistered"`
+}
+
+// ClusterHealthCondition contains information about health condition for the whole cluster.
+type ClusterHealthCondition struct {
+	// Status of cluster health.
+	Status ClusterAutoscalerConditionStatus `json:"status,omitempty" yaml:"status,omitempty"`
+	// NodeCounts contains number of nodes that satisfy different criteria for health condition.
+	NodeCounts HealthConditionNodeCount `json:"nodeCounts,omitempty" yaml:"nodeCounts,omitempty"`
 	// LastProbeTime is the last time we probed the condition.
-	LastProbeTime metav1.Time `json:"lastProbeTime,omitempty"`
+	LastProbeTime metav1.Time `json:"lastProbeTime,omitempty" yaml:"lastProbeTime,omitempty"`
 	// LastTransitionTime is the time since when the condition was in the given state.
-	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty"`
+	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty" yaml:"lastTransitionTime,omitempty"`
 }
 
-// ClusterAutoscalerStatus contains ClusterAutoscaler status.
-type ClusterAutoscalerStatus struct {
-	// NodeGroupStatuses contains status information of individual node groups on which CA works.
-	NodeGroupStatuses []NodeGroupStatus `json:"nodeGroupStatuses,omitempty"`
-	// ClusterwideConditions contains conditions that apply to the whole autoscaler.
-	ClusterwideConditions []ClusterAutoscalerCondition `json:"clusterwideConditions,omitempty"`
+// NodeGroupHealthCondition contains information about health condition for a node group.
+type NodeGroupHealthCondition struct {
+	// Status of node group health.
+	Status ClusterAutoscalerConditionStatus `json:"status,omitempty" yaml:"status,omitempty"`
+	// NodeCounts contains number of nodes that satisfy different criteria for health condition.
+	NodeCounts HealthConditionNodeCount `json:"nodeCounts,omitempty" yaml:"nodeCounts,omitempty"`
+	// CloudProviderTarget is the target size set by cloud provider.
+	CloudProviderTarget int `json:"cloudProviderTarget" yaml:"cloudProviderTarget"`
+	// MinSize is the CA max size of a node group.
+	MinSize int `json:"minSize" yaml:"minSize"`
+	// MaxSize is the CA max size of a node group.
+	MaxSize int `json:"maxSize" yaml:"maxSize"`
+	// LastProbeTime is the last time we probed the condition.
+	LastProbeTime metav1.Time `json:"lastProbeTime,omitempty" yaml:"lastProbeTime,omitempty"`
+	// LastTransitionTime is the time since when the condition was in the given state.
+	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty" yaml:"lastTransitionTime,omitempty"`
+}
+
+// ClusterScaleUpNodeCount contains number of nodes that satisfy different criteria for scale up condition of the whole cluster.
+type ClusterScaleUpNodeCount struct {
+	Ready      int `json:"ready" yaml:"ready"`
+	Registered int `json:"registered" yaml:"registered"`
+}
+
+// NodeGroupScaleUpNodeCount contains number of nodes that satisfy different criteria for scale up condition of a node group.
+type NodeGroupScaleUpNodeCount struct {
+	Ready               int `json:"ready" yaml:"ready"`
+	CloudProviderTarget int `json:"cloudProviderTarget" yaml:"cloudProviderTarget"`
+}
+
+// ClusterScaleUpCondition contains information about scale up condition for the whole cluster.
+type ClusterScaleUpCondition struct {
+	// Status of the scale up.
+	Status ClusterAutoscalerConditionStatus `json:"status,omitempty" yaml:"status,omitempty"`
+	// NodeCounts contains number of nodes that satisfy different criteria for scale up condition.
+	NodeCounts ClusterScaleUpNodeCount `json:"nodeCounts,omitempty" yaml:"nodeCounts,omitempty"`
+	// LastProbeTime is the last time we probed the condition.
+	LastProbeTime metav1.Time `json:"lastProbeTime,omitempty" yaml:"lastProbeTime,omitempty"`
+	// LastTransitionTime is the time since when the condition was in the given state.
+	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty" yaml:"lastTransitionTime,omitempty"`
+}
+
+// BackoffInfo contains error information that caused the backoff.
+type BackoffInfo struct {
+	// ErrorCode is a specific error code for error condition
+	ErrorCode string `json:"errorCode,omitempty" yaml:"errorCode,omitempty"`
+	// ErrorMessage is human readable description of error condition
+	ErrorMessage string `json:"errorMessage,omitempty" yaml:"errorMessage,omitempty"`
 }
 
-// NodeGroupStatus contains status of a group of nodes controlled by ClusterAutoscaler.
+// NodeGroupScaleUpCondition contains information about scale up condition for a node group.
+type NodeGroupScaleUpCondition struct {
+	// Status of the scale up.
+	Status ClusterAutoscalerConditionStatus `json:"status,omitempty" yaml:"status,omitempty"`
+	// NodeCounts contains number of nodes that satisfy different criteria for scale up condition.
+	NodeCounts NodeGroupScaleUpNodeCount `json:"nodeCounts,omitempty" yaml:"nodeCounts,omitempty"`
+	// LastProbeTime is the last time we probed the condition.
+	BackoffInfo   BackoffInfo `json:"backoffInfo,omitempty" yaml:"backoffInfo,omitempty"`
+	LastProbeTime metav1.Time `json:"lastProbeTime,omitempty" yaml:"lastProbeTime,omitempty"`
+	// LastTransitionTime is the time since when the condition was in the given state.
+	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty" yaml:"lastTransitionTime,omitempty"`
+}
+
+// ScaleDownCondition contains information about scale down condition for a node group or the whole cluster.
+type ScaleDownCondition struct {
+	// Status of the scale down.
+	Status ClusterAutoscalerConditionStatus `json:"status,omitempty" yaml:"status,omitempty"`
+	// Candidates number for the scale down.
+	Candidates int `json:"candidates,omitempty" yaml:"candidates,omitempty"`
+	// LastProbeTime is the last time we probed the condition.
+	LastProbeTime metav1.Time `json:"lastProbeTime,omitempty" yaml:"lastProbeTime,omitempty"`
+	// LastTransitionTime is the time since when the condition was in the given state.
+	LastTransitionTime metav1.Time `json:"lastTransitionTime,omitempty" yaml:"lastTransitionTime,omitempty"`
+}
+
+// ClusterWideStatus contains status that apply to the whole cluster.
+type ClusterWideStatus struct {
+	// Health contains information about health condition of the cluster.
+	Health ClusterHealthCondition `json:"health,omitempty" yaml:"health,omitempty"`
+	// ScaleUp contains information about scale up condition of the cluster.
+	ScaleUp ClusterScaleUpCondition `json:"scaleUp,omitempty" yaml:"scaleUp,omitempty"`
+	// ScaleDown contains information about scale down condition of the node group.
+	ScaleDown ScaleDownCondition `json:"scaleDown,omitempty" yaml:"scaleDown,omitempty"`
+}
+
+// NodeGroupStatus contains status of an individual node group on which CA works..
 type NodeGroupStatus struct {
-	// ProviderID is the cloud-provider-specific name of the node group. On GCE it will be equal
-	// to MIG url, on AWS it will be ASG name, etc.
-	ProviderID string `json:"providerID,omitempty"`
-	// Conditions is a list of conditions that describe the state of the node group.
-	Conditions []ClusterAutoscalerCondition `json:"conditions,omitempty"`
+	// Name of the node group.
+	Name string `json:"name,omitempty" yaml:"name,omitempty"`
+	// Health contains information about health condition of the node group.
+	Health NodeGroupHealthCondition `json:"health,omitempty" yaml:"health,omitempty"`
+	// ScaleUp contains information about scale up condition of the node group.
+	ScaleUp NodeGroupScaleUpCondition `json:"scaleUp,omitempty" yaml:"scaleUp,omitempty"`
+	// ScaleDown contains information about scale down condition of the node group.
+	ScaleDown ScaleDownCondition `json:"scaleDown,omitempty" yaml:"scaleDown,omitempty"`
+}
+
+// ClusterAutoscalerStatus contains ClusterAutoscaler status.
+type ClusterAutoscalerStatus struct {
+	// Time of the cluster autoscaler status.
+	Time string `json:"time,omitempty" yaml:"time,omitempty"`
+	// AutoscalerStatus contains status of ClusterAutoscaler (e.g. 'Initializing' & 'Running').
+	AutoscalerStatus string `json:"autoscalerStatus,omitempty" yaml:"autoscalerStatus,omitempty"`
+	// Message contains extra information about the status.
+	Message string `json:"message,omitempty" yaml:"message,omitempty"`
+	// ClusterWide contains conditions that apply to the whole cluster.
+	ClusterWide ClusterWideStatus `json:"clusterWide,omitempty" yaml:"clusterWide,omitempty"`
+	// NodeGroups contains status information of individual node groups on which CA works.
+	NodeGroups []NodeGroupStatus `json:"nodeGroups,omitempty" yaml:"nodeGroups,omitempty"`
 }