feat : Add orphan checkpoint retention policy

.github/workflows/tests.yaml

@@ -95,3 +95,12 @@ jobs:
 
  - name: Test Max Checkpoints Set to 1
  run: sudo -E bats -f "test_max_checkpoints_set_to_1" ./test/run_tests.bats
+
+ - name: Test Max Total Checkpoint Size
+ run: sudo -E bats -f "test_max_total_checkpoint_size" ./test/run_tests.bats
+
+ - name: Test Max Checkpoint Size
+ run: sudo -E bats -f "test_max_checkpoint_size" ./test/run_tests.bats
+
+ - name: Test orphan retention
+ run: sudo -E bats -f "test_orphan_retention_policy" ./test/run_tests.bats

api/v1/checkpointrestoreoperator_types.go

@@ -34,27 +34,41 @@ type CheckpointRestoreOperatorSpec struct {
 }
 
 type GlobalPolicySpec struct {
- MaxCheckpointsPerNamespaces *int `json:"maxCheckpointsPerNamespace,omitempty"`
- MaxCheckpointsPerPod *int `json:"maxCheckpointsPerPod,omitempty"`
- MaxCheckpointsPerContainer *int `json:"maxCheckpointsPerContainer,omitempty"`
+ RetainOrphan *bool `json:"retainOrphan,omitempty"`
+ MaxCheckpointsPerNamespaces *int `json:"maxCheckpointsPerNamespace,omitempty"`
+ MaxCheckpointsPerPod *int `json:"maxCheckpointsPerPod,omitempty"`
+ MaxCheckpointsPerContainer *int `json:"maxCheckpointsPerContainer,omitempty"`
+ MaxCheckpointSize *int `json:"maxCheckpointSize,omitempty"`
+ MaxTotalSizePerNamespace *int `json:"maxTotalSizePerNamespace,omitempty"`
+ MaxTotalSizePerPod *int `json:"maxTotalSizePerPod,omitempty"`
+ MaxTotalSizePerContainer *int `json:"maxTotalSizePerContainer,omitempty"`
 }
 
 type ContainerPolicySpec struct {
- Namespace string `json:"namespace,omitempty"`
- Pod string `json:"pod,omitempty"`
- Container string `json:"container,omitempty"`
- MaxCheckpoints *int64 `json:"maxCheckpoints,omitempty"`
+ Namespace string `json:"namespace,omitempty"`
+ Pod string `json:"pod,omitempty"`
+ Container string `json:"container,omitempty"`
+ RetainOrphan *bool `json:"retainOrphan,omitempty"`
+ MaxCheckpoints *int `json:"maxCheckpoints,omitempty"`
+ MaxCheckpointSize *int `json:"maxCheckpointSize,omitempty"`
+ MaxTotalSize *int `json:"maxTotalSize,omitempty"`
 }
 
 type PodPolicySpec struct {
- Namespace string `json:"namespace,omitempty"`
- Pod string `json:"pod,omitempty"`
- MaxCheckpoints *int64 `json:"maxCheckpoints,omitempty"`
+ Namespace string `json:"namespace,omitempty"`
+ Pod string `json:"pod,omitempty"`
+ RetainOrphan *bool `json:"retainOrphan,omitempty"`
+ MaxCheckpoints *int `json:"maxCheckpoints,omitempty"`
+ MaxCheckpointSize *int `json:"maxCheckpointSize,omitempty"`
+ MaxTotalSize *int `json:"maxTotalSize,omitempty"`
 }
 
 type NamespacePolicySpec struct {
- Namespace string `json:"namespace,omitempty"`
- MaxCheckpoints *int64 `json:"maxCheckpoints,omitempty"`
+ Namespace string `json:"namespace,omitempty"`
+ RetainOrphan *bool `json:"retainOrphan,omitempty"`
+ MaxCheckpoints *int `json:"maxCheckpoints,omitempty"`
+ MaxCheckpointSize *int `json:"maxCheckpointSize,omitempty"`
+ MaxTotalSize *int `json:"maxTotalSize,omitempty"`
 }
 
 // CheckpointRestoreOperatorStatus defines the observed state of CheckpointRestoreOperator
@@ -64,6 +78,8 @@ type CheckpointRestoreOperatorStatus struct {
 
 //+kubebuilder:object:root=true
 //+kubebuilder:subresource:status
+//+kubebuilder:rbac:groups="",resources=pods,verbs=get;list;watch
+//+kubebuilder:rbac:groups="",resources=namespaces,verbs=get;list;watch
 
 // CheckpointRestoreOperator is the Schema for the checkpointrestoreoperators API
 type CheckpointRestoreOperator struct {

config/crd/bases/criu.org_checkpointrestoreoperators.yaml

@@ -52,44 +52,69 @@ spec:
  properties:
  container:
  type: string
+ maxCheckpointSize:
+ type: integer
  maxCheckpoints:
- format: int64
+ type: integer
+ maxTotalSize:
  type: integer
  namespace:
  type: string
  pod:
  type: string
+ retainOrphan:
+ type: boolean
  type: object
  type: array
  globalPolicy:
  properties:
+ maxCheckpointSize:
+ type: integer
  maxCheckpointsPerContainer:
  type: integer
  maxCheckpointsPerNamespace:
  type: integer
  maxCheckpointsPerPod:
  type: integer
+ maxTotalSizePerContainer:
+ type: integer
+ maxTotalSizePerNamespace:
+ type: integer
+ maxTotalSizePerPod:
+ type: integer
+ retainOrphan:
+ type: boolean
  type: object
  namespacePolicies:
  items:
  properties:
+ maxCheckpointSize:
+ type: integer
  maxCheckpoints:
- format: int64
+ type: integer
+ maxTotalSize:
  type: integer
  namespace:
  type: string
+ retainOrphan:
+ type: boolean
  type: object
  type: array
  podPolicies:
  items:
  properties:
+ maxCheckpointSize:
+ type: integer
  maxCheckpoints:
- format: int64
+ type: integer
+ maxTotalSize:
  type: integer
  namespace:
  type: string
  pod:
  type: string
+ retainOrphan:
+ type: boolean
  type: object
  type: array
  type: object

config/rbac/role.yaml

@@ -4,6 +4,22 @@ kind: ClusterRole
 metadata:
  name: manager-role
 rules:
+- apiGroups:
+ - ""
+ resources:
+ - namespaces
+ verbs:
+ - get
+ - list
+ - watch
+- apiGroups:
+ - ""
+ resources:
+ - pods
+ verbs:
+ - get
+ - list
+ - watch
 - apiGroups:
  - criu.org
  resources:

config/samples/_v1_checkpointrestoreoperator.yaml

@@ -12,18 +12,30 @@ spec:
  checkpointDirectory: /var/lib/kubelet/checkpoints
  applyPoliciesImmediately: false
  globalPolicy:
+ retainOrphan: true
  maxCheckpointsPerNamespace: 50
  maxCheckpointsPerPod: 30
  maxCheckpointsPerContainer: 10
+ maxCheckpointSize: 10
+ maxTotalSizePerNamespace: 1000
+ maxTotalSizePerPod: 500
+ maxTotalSizePerContainer: 100
  # containerPolicies:
  # - namespace: <namespace>
  # pod: <pod_name>
  # container: <container_name>
+ # retainOrphan: false
  # maxCheckpoints: 5
+ # maxCheckpointSize: 10
+ # maxTotalSize: 100
  # podPolicies:
  # - namespace: <namespace>
  # pod: <pod_name>
  # maxCheckpoints: 10
+ # maxCheckpointSize: 10
+ # maxTotalSize: 500
  # namespacePolicies:
  # - namespace: <namespace>
  # maxCheckpoints: 15
+ # maxCheckpointSize: 10
+ # maxTotalSize: 1000

docs/retention_policy.md

@@ -27,6 +27,7 @@ spec:
  checkpointDirectory: /var/lib/kubelet/checkpoints
  applyPoliciesImmediately: false
  globalPolicy:
+ retainOrphan: true
  maxCheckpointsPerNamespace: 50
  maxCheckpointsPerPod: 30
  maxCheckpointsPerContainer: 10
@@ -35,11 +36,16 @@ spec:
  # - namespace: <namespace>
  # pod: <pod_name>
  # container: <container_name>
+ # retainOrphan: false # Set to false will delete all orphan checkpoints
  # maxCheckpoints: 5
+ # maxCheckpointSize: 6 # Maximum size of a single checkpoint in MB
+ # maxTotalSize: 20 # Maximum total size of checkpoints for the container in MB
  # podPolicies:
  # - namespace: <namespace>
  # pod: <pod_name>
  # maxCheckpoints: 10
+ # maxCheckpointSize: 8 # Maximum size of a single checkpoint in MB
+ # maxTotalSize: 50 # Maximum total size of checkpoints for the pod in MB
  # namespacePolicies:
  # - namespace: <namespace>
  # maxCheckpoints: 15` 
@@ -51,21 +57,35 @@ A sample configuration file is available [here](/config/samples/_v1_checkpointre
 - `checkpointDirectory`: Specifies the directory where checkpoints are stored.
 - `applyPoliciesImmediately`: If set to `true`, the policies are applied immediately. If `false` (default value), they are applied after new checkpoint creation.
 - `globalPolicy`: Defines global checkpoint retention limits.
+ - `retainOrphan`: If set to `true` (default), orphan checkpoints (checkpoints whose associated resources have been deleted) will be retained. If set to `false`, orphan checkpoints will be automatically deleted.
  - `maxCheckpointsPerNamespace`: Maximum number of checkpoints per namespace.
  - `maxCheckpointsPerPod`: Maximum number of checkpoints per pod.
  - `maxCheckpointsPerContainer`: Maximum number of checkpoints per container.
+ - `maxCheckpointSize`: Maximum size of a single checkpoint in MB.
+ - `maxTotalSizePerNamespace`: Maximum total size of checkpoints per namespace in MB.
+ - `maxTotalSizePerPod`: Maximum total size of checkpoints per pod in MB.
+ - `maxTotalSizePerContainer`: Maximum total size of checkpoints per container in MB.
 - `containerPolicies` (optional): Specific retention policies for containers.
  - `namespace`: Namespace of the container.
  - `pod`: Pod name of the container.
  - `container`: Container name.
+ - `retainOrphan`: If set to `true` (default), orphan checkpoints for this container will be retained. If set to `false`, orphan checkpoints will be deleted.
  - `maxCheckpoints`: Maximum number of checkpoints for the container.
+ - `maxCheckpointSize`: Maximum size of a single checkpoint in MB.
+ - `maxTotalSize`: Maximum total size of checkpoints for the container in MB.
 - `podPolicies` (optional): Specific retention policies for pods.
  - `namespace`: Namespace of the pod.
  - `pod`: Pod name.
+ - `retainOrphan`: If set to `true` (default), orphan checkpoints for this pod will be retained. If set to `false`, orphan checkpoints will be deleted.
  - `maxCheckpoints`: Maximum number of checkpoints for the pod.
+ - `maxCheckpointSize`: Maximum size of a single checkpoint in MB.
+ - `maxTotalSize`: Maximum total size of checkpoints for the pod in MB.
 - `namespacePolicies` (optional): Specific retention policies for namespaces.
  - `namespace`: Namespace name.
+ - `retainOrphan`: If set to `true` (default), orphan checkpoints for this namespace will be retained. If set to `false`, orphan checkpoints will be deleted.
  - `maxCheckpoints`: Maximum number of checkpoints for the namespace.
+ - `maxCheckpointSize`: Maximum size of a single checkpoint in MB.
+ - `maxTotalSize`: Maximum total size of checkpoints for the namespace in MB.
 
 ## Policy Hierarchy and Application
 

go.mod

@@ -9,6 +9,7 @@ require (
  github.com/go-logr/logr v1.4.2
  github.com/onsi/ginkgo/v2 v2.20.0
  github.com/onsi/gomega v1.34.1
+ k8s.io/api v0.30.3
  k8s.io/apimachinery v0.30.3
  k8s.io/client-go v0.30.3
  k8s.io/kubelet v0.30.3
@@ -71,7 +72,6 @@ require (
  gopkg.in/inf.v0 v0.9.1 // indirect
  gopkg.in/yaml.v2 v2.4.0 // indirect
  gopkg.in/yaml.v3 v3.0.1 // indirect
- k8s.io/api v0.30.3 // indirect
  k8s.io/apiextensions-apiserver v0.30.3 // indirect
  k8s.io/klog/v2 v2.130.1 // indirect
  k8s.io/kube-openapi v0.0.0-20240709000822-3c01b740850f // indirect