chore: rename chrek to Dynamo Snapshot (#7028)

Signed-off-by: Schwinn Saereesitthipitak <17022745+galletas1712@users.noreply.github.com>

deploy/helm/charts/chrek/templates/seccomp-configmap.yaml → deploy/helm/charts/snapshot/templates/seccomp-configmap.yaml

@@ -5,10 +5,10 @@
 apiVersion: v1
 kind: ConfigMap
 metadata:
-  name: {{ include "chrek.fullname" . }}-seccomp
+  name: {{ include "snapshot.fullname" . }}-seccomp
   namespace: {{ .Release.Namespace }}
   labels:
-    {{- include "chrek.labels" . | nindent 4 }}
+    {{- include "snapshot.labels" . | nindent 4 }}
     app.kubernetes.io/component: seccomp
 data:
   block-iouring.json: |

deploy/helm/charts/chrek/templates/serviceaccount.yaml → deploy/helm/charts/snapshot/templates/serviceaccount.yaml

@@ -5,10 +5,10 @@
 apiVersion: v1
 kind: ServiceAccount
 metadata:
-  name: {{ include "chrek.serviceAccountName" . }}
+  name: {{ include "snapshot.serviceAccountName" . }}
   namespace: {{ .Release.Namespace }}
   labels:
-    {{- include "chrek.labels" . | nindent 4 }}
+    {{- include "snapshot.labels" . | nindent 4 }}
     app.kubernetes.io/component: checkpoint-agent
   {{- with .Values.serviceAccount.annotations }}
   annotations:

deploy/helm/charts/chrek/values.yaml → deploy/helm/charts/snapshot/values.yaml

@@ -12,7 +12,7 @@
 # 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.
-# Chrek - Checkpoint/Restore Infrastructure
+# Dynamo Snapshot - Checkpoint/Restore Infrastructure
 # This chart deploys the checkpoint storage (PVC) and CRIU agent (DaemonSet)
 # in a single namespace. Install this chart in each namespace where you want
 # to enable checkpoint/restore functionality for DynamoGraphDeployments.
@@ -27,7 +27,7 @@ storage:
     # Create a new PVC (set to false if using existing PVC)
     create: true
     # PVC name - must match operator configuration
-    name: chrek-pvc
+    name: snapshot-pvc
     # PVC size
     size: 100Gi
     # Storage class (leave empty for default)
@@ -48,14 +48,17 @@ storage:
     # OCI URI (e.g., oci://registry.io/repo/checkpoints)
     uri: ""
 
-# DaemonSet configuration for chrek (checkpoint/restore) agent
+# DaemonSet configuration for snapshot (checkpoint/restore) agent
 daemonset:
   # Container image
   image:
-    repository: nvcr.io/nvidia/ai-dynamo/chrek-agent
+    repository: nvcr.io/nvidia/ai-dynamo/snapshot-agent
     tag: 1.0.0
     pullPolicy: Always
 
+  # Snapshot agent and nsrestore log level (trace, debug, info, warn, error)
+  snapshotLogLevel: info
+
   # Image pull secrets
   imagePullSecrets:
     - name: ngc-secret

deploy/operator/api/config/v1alpha1/defaults.go

@@ -93,7 +93,7 @@ func SetDefaultsOperatorConfiguration(obj *OperatorConfiguration) {
 		obj.Checkpoint.Storage.Type = CheckpointStorageTypePVC
 	}
 	if obj.Checkpoint.Storage.PVC.PVCName == "" {
-		obj.Checkpoint.Storage.PVC.PVCName = "chrek-pvc"
+		obj.Checkpoint.Storage.PVC.PVCName = "snapshot-pvc"
 	}
 	if obj.Checkpoint.Storage.PVC.BasePath == "" {
 		obj.Checkpoint.Storage.PVC.BasePath = "/checkpoints"

deploy/operator/api/config/v1alpha1/types.go

@@ -265,7 +265,7 @@ type CheckpointStorageConfiguration struct {
 // CheckpointPVCConfig holds PVC storage configuration.
 type CheckpointPVCConfig struct {
 	// PVCName is the name of the PVC
-	// +kubebuilder:default="chrek-pvc"
+	// +kubebuilder:default="snapshot-pvc"
 	PVCName string `json:"pvcName"`
 	// BasePath is the base directory within the PVC
 	// +kubebuilder:default="/checkpoints"

deploy/operator/internal/checkpoint/dgd_integration.go

@@ -325,7 +325,7 @@ func InjectPodInfoVolumeMount(container *corev1.Container) {
 }
 
 // InjectCheckpointIntoPodSpec injects checkpoint configuration into a pod spec for
-// external restore via the chrek DaemonSet. The pod image is expected to be a
+// external restore via the snapshot DaemonSet. The pod image is expected to be a
 // runtime-compatible restore image (runtime + CRIU tooling). For ready checkpoints,
 // the operator overrides command to `sleep infinity` so the watcher can trigger
 // external restore via nsenter + nsrestore.

deploy/operator/internal/checkpoint/dgd_integration_test.go

@@ -44,7 +44,7 @@ func testPVCConfig() *configv1alpha1.CheckpointConfiguration {
 		Storage: configv1alpha1.CheckpointStorageConfiguration{
 			Type: configv1alpha1.CheckpointStorageTypePVC,
 			PVC: configv1alpha1.CheckpointPVCConfig{
-				PVCName:  "chrek-pvc",
+				PVCName:  "snapshot-pvc",
 				BasePath: "/checkpoints",
 			},
 		},
@@ -113,7 +113,7 @@ func TestHelpers(t *testing.T) {
 func TestInjectionIdempotency(t *testing.T) {
 	// Volume injection is idempotent
 	podSpec := &corev1.PodSpec{Volumes: []corev1.Volume{{Name: consts.CheckpointVolumeName}, {Name: consts.PodInfoVolumeName}}}
-	InjectCheckpointVolume(podSpec, "chrek-pvc")
+	InjectCheckpointVolume(podSpec, "snapshot-pvc")
 	InjectPodInfoVolume(podSpec)
 	assert.Len(t, podSpec.Volumes, 2)
 
@@ -260,7 +260,7 @@ func TestInjectCheckpointIntoPodSpec(t *testing.T) {
 		for _, v := range podSpec.Volumes {
 			volNames[v.Name] = true
 			if v.Name == consts.CheckpointVolumeName {
-				assert.Equal(t, "chrek-pvc", v.PersistentVolumeClaim.ClaimName)
+				assert.Equal(t, "snapshot-pvc", v.PersistentVolumeClaim.ClaimName)
 			}
 		}
 		assert.True(t, volNames[consts.CheckpointVolumeName])
@@ -329,7 +329,7 @@ func TestInjectCheckpointIntoPodSpec(t *testing.T) {
 				Storage: configv1alpha1.CheckpointStorageConfiguration{Type: "pvc", PVC: configv1alpha1.CheckpointPVCConfig{BasePath: "/checkpoints"}},
 			}, "no PVC name"},
 			{"PVC base path missing", testPodSpec(), testInfo(), &configv1alpha1.CheckpointConfiguration{
-				Storage: configv1alpha1.CheckpointStorageConfiguration{Type: "pvc", PVC: configv1alpha1.CheckpointPVCConfig{PVCName: "chrek-pvc"}},
+				Storage: configv1alpha1.CheckpointStorageConfiguration{Type: "pvc", PVC: configv1alpha1.CheckpointPVCConfig{PVCName: "snapshot-pvc"}},
 			}, "no PVC base path"},
 			{"S3 URI missing", testPodSpec(), testInfo(), &configv1alpha1.CheckpointConfiguration{
 				Storage: configv1alpha1.CheckpointStorageConfiguration{Type: "s3"},

deploy/operator/internal/consts/consts.go

@@ -140,13 +140,13 @@ const (
 	ResourceStateUnknown  = "unknown"
 
 	// Checkpoint/restore constants
-	// CROSS-REFERENCE: Some constants below are duplicated in the chrek package at
-	// deploy/chrek/pkg/config/constants.go. If you change a value here, update there too.
+	// CROSS-REFERENCE: Some constants below are duplicated in the snapshot package at
+	// deploy/snapshot/pkg/config/constants.go. If you change a value here, update there too.
 
 	// Kubernetes labels
-	KubeLabelIsCheckpointSource = "nvidia.com/chrek-is-checkpoint-source" // Pod label that triggers DaemonSet auto-checkpoint
-	KubeLabelCheckpointHash     = "nvidia.com/chrek-checkpoint-hash"      // Checkpoint identity hash (= DynamoCheckpoint CR name)
-	KubeLabelIsRestoreTarget    = "nvidia.com/chrek-is-restore-target"    // Pod label that triggers DaemonSet auto-restore
+	KubeLabelIsCheckpointSource = "nvidia.com/snapshot-is-checkpoint-source" // Pod label that triggers DaemonSet auto-checkpoint
+	KubeLabelCheckpointHash     = "nvidia.com/snapshot-checkpoint-hash"      // Checkpoint identity hash (= DynamoCheckpoint CR name)
+	KubeLabelIsRestoreTarget    = "nvidia.com/snapshot-is-restore-target"    // Pod label that triggers DaemonSet auto-restore
 
 	// Environment variables injected into pods
 	EnvCheckpointStorageType  = "DYN_CHECKPOINT_STORAGE_TYPE"   // Storage backend (pvc, s3, oci) — checkpoint job pods only
@@ -159,7 +159,7 @@ const (
 	CheckpointVolumeName = "checkpoint-storage" // Pod-internal volume name for checkpoint PVC
 
 	// SeccompProfilePath is the localhost seccomp profile that blocks io_uring syscalls.
-	// Deployed to nodes by the chrek DaemonSet init container.
+	// Deployed to nodes by the snapshot DaemonSet init container.
 	SeccompProfilePath = "profiles/block-iouring.json"
 
 	// Pod identity (Downward API) ---

deploy/operator/internal/controller/dynamocheckpoint_controller_test.go

@@ -58,7 +58,7 @@ func checkpointTestConfig() *configv1alpha1.OperatorConfiguration {
 			Storage: configv1alpha1.CheckpointStorageConfiguration{
 				Type: configv1alpha1.CheckpointStorageTypePVC,
 				PVC: configv1alpha1.CheckpointPVCConfig{
-					PVCName:  "chrek-pvc",
+					PVCName:  "snapshot-pvc",
 					BasePath: "/checkpoints",
 				},
 			},
@@ -143,7 +143,7 @@ func TestBuildCheckpointJob(t *testing.T) {
 		volNames[v.Name] = true
 		if v.Name == consts.CheckpointVolumeName {
 			require.NotNil(t, v.PersistentVolumeClaim)
-			assert.Equal(t, "chrek-pvc", v.PersistentVolumeClaim.ClaimName)
+			assert.Equal(t, "snapshot-pvc", v.PersistentVolumeClaim.ClaimName)
 		}
 		if v.Name == consts.PodInfoVolumeName {
 			require.NotNil(t, v.DownwardAPI)

deploy/operator/internal/controller/dynamocomponentdeployment_controller_test.go

@@ -1298,7 +1298,7 @@ func TestDynamoComponentDeploymentReconciler_generatePodTemplateSpec_RestoreLabe
 					Storage: configv1alpha1.CheckpointStorageConfiguration{
 						Type: configv1alpha1.CheckpointStorageTypePVC,
 						PVC: configv1alpha1.CheckpointPVCConfig{
-							PVCName:  "chrek-pvc",
+							PVCName:  "snapshot-pvc",
 							BasePath: "/checkpoints",
 						},
 					},
@@ -1430,7 +1430,7 @@ func TestDynamoComponentDeploymentReconciler_generateDeployment_RestoreStrategy(
 					Storage: configv1alpha1.CheckpointStorageConfiguration{
 						Type: configv1alpha1.CheckpointStorageTypePVC,
 						PVC: configv1alpha1.CheckpointPVCConfig{
-							PVCName:  "chrek-pvc",
+							PVCName:  "snapshot-pvc",
 							BasePath: "/checkpoints",
 						},
 					},

deploy/chrek/Dockerfile → deploy/snapshot/Dockerfile

 # SPDX-FileCopyrightText: Copyright (c) 2024-2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
 # SPDX-License-Identifier: Apache-2.0
 #
-# Unified Dockerfile for chrek-agent and placeholder images.
+# Unified Dockerfile for snapshot-agent and placeholder images.
 #
 # Build targets:
-#   docker build --target agent -t chrek-agent:latest .
+#   docker build --target agent -t snapshot-agent:latest .
 #   docker build --target placeholder --build-arg BASE_IMAGE=<app-image> -t placeholder:latest .
 #
 # Optional targets for CI:
@@ -66,7 +66,7 @@ FROM go-base AS builder
 ARG TARGETOS=linux
 ARG TARGETARCH=amd64
 
-RUN CGO_ENABLED=0 GOOS=${TARGETOS} GOARCH=${TARGETARCH} go build -ldflags="-w -s" -o /chrek-agent ./cmd/agent
+RUN CGO_ENABLED=0 GOOS=${TARGETOS} GOARCH=${TARGETARCH} go build -ldflags="-w -s" -o /snapshot-agent ./cmd/agent
 RUN CGO_ENABLED=0 GOOS=${TARGETOS} GOARCH=${TARGETARCH} go build -ldflags="-w -s" -o /nsrestore ./cmd/nsrestore
 
 # =============================================================================
@@ -105,7 +105,7 @@ RUN git clone --branch ${CRIU_VERSION} https://github.com/checkpoint-restore/cri
 RUN git clone https://github.com/NVIDIA/cuda-checkpoint.git /tmp/cuda-checkpoint
 
 # =============================================================================
-# Stage: Agent - Final chrek-agent image
+# Stage: Agent - Final snapshot-agent image
 # =============================================================================
 FROM ${AGENT_BASE_IMAGE} AS agent
 
@@ -137,15 +137,15 @@ COPY --from=criu-builder /tmp/cuda-checkpoint/bin/x86_64_Linux/cuda-checkpoint /
 RUN chmod +x /usr/local/sbin/cuda-checkpoint
 
 # Copy the built binaries
-COPY --from=builder /chrek-agent /usr/local/bin/chrek-agent
+COPY --from=builder /snapshot-agent /usr/local/bin/snapshot-agent
 COPY --from=builder /nsrestore /usr/local/bin/nsrestore
 
 # Create directories
-RUN mkdir -p /checkpoints /var/run/chrek
+RUN mkdir -p /checkpoints /var/run/snapshot
 
 USER root
 
-ENTRYPOINT ["/usr/local/bin/chrek-agent"]
+ENTRYPOINT ["/usr/local/bin/snapshot-agent"]
 
 # =============================================================================
 # Stage: Placeholder - Runtime-compatible restore image (requires BASE_IMAGE arg)

deploy/chrek/Makefile → deploy/snapshot/Makefile

@@ -2,7 +2,7 @@
 # SPDX-License-Identifier: Apache-2.0
 
 # Image URL to use all building/pushing image targets
-IMG ?= nvcr.io/nvidian/dynamo-dev/chrek-agent:latest
+IMG ?= nvcr.io/nvidian/dynamo-dev/snapshot-agent:latest
 PLACEHOLDER_IMG ?= nvcr.io/nvidian/dynamo-dev/dynamo-vllm-placeholder:latest
 # PLACEHOLDER_BASE_IMG must be provided when building placeholder (no default)
 
@@ -54,12 +54,12 @@ lint-fix: golangci-lint ## Run golangci-lint linter and perform fixes.
 ##@ Build
 
 .PHONY: build
-build: fmt vet ## Build chrek-agent binary.
-	CGO_ENABLED=0 go build -ldflags="-w -s" -o bin/chrek-agent ./cmd/agent
+build: fmt vet ## Build snapshot-agent binary.
+	CGO_ENABLED=0 go build -ldflags="-w -s" -o bin/snapshot-agent ./cmd/agent
 
 .PHONY: run
-run: build ## Run chrek-agent from your host.
-	./bin/chrek-agent
+run: build ## Run snapshot-agent from your host.
+	./bin/snapshot-agent
 
 .PHONY: clean
 clean: ## Remove build artifacts.
@@ -69,15 +69,15 @@ clean: ## Remove build artifacts.
 ##@ Docker
 
 .PHONY: docker-build-agent
-docker-build-agent: ## Build chrek-agent docker image.
+docker-build-agent: ## Build snapshot-agent docker image.
 	$(CONTAINER_TOOL) build --target agent -t ${IMG} .
 
 .PHONY: docker-build-agent-lint
-docker-build-agent-lint: ## Build chrek-agent docker image up to lint stage.
+docker-build-agent-lint: ## Build snapshot-agent docker image up to lint stage.
 	$(CONTAINER_TOOL) build --target linter -t ${IMG}-lint .
 
 .PHONY: docker-build-agent-test
-docker-build-agent-test: ## Build chrek-agent docker image up to test stage.
+docker-build-agent-test: ## Build snapshot-agent docker image up to test stage.
 	$(CONTAINER_TOOL) build --target tester -t ${IMG}-test .
 
 .PHONY: docker-build-placeholder
@@ -97,7 +97,7 @@ endif
 		-t ${PLACEHOLDER_IMG} .
 
 .PHONY: docker-push-agent
-docker-push-agent: ## Push chrek-agent docker image.
+docker-push-agent: ## Push snapshot-agent docker image.
 	$(CONTAINER_TOOL) push ${IMG}
 
 .PHONY: docker-push-placeholder

deploy/chrek/cmd/agent/config.go → deploy/snapshot/cmd/agent/config.go

@@ -8,11 +8,11 @@ import (
 
 	"gopkg.in/yaml.v3"
 
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/types"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/types"
 )
 
 // ConfigMapPath is the default path where the ConfigMap is mounted.
-const ConfigMapPath = "/etc/chrek/config.yaml"
+const ConfigMapPath = "/etc/snapshot/config.yaml"
 
 // LoadConfig loads the agent configuration from a YAML file.
 func LoadConfig(path string) (*types.AgentConfig, error) {

deploy/chrek/cmd/agent/main.go → deploy/snapshot/cmd/agent/main.go

-// Package main provides the chrek DaemonSet agent.
+// Package main provides the snapshot DaemonSet agent.
 // The agent watches for pods with checkpoint/restore labels on its node
 // and triggers operations via the orchestrators.
 package main
@@ -12,9 +12,9 @@ import (
 	"github.com/containerd/containerd"
 	"github.com/go-logr/logr"
 
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/common"
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/logging"
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/watcher"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/common"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/logging"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/watcher"
 )
 
 func main() {
@@ -41,7 +41,7 @@ func main() {
 	sigChan := make(chan os.Signal, 1)
 	signal.Notify(sigChan, syscall.SIGINT, syscall.SIGTERM)
 
-	agentLog.Info("Starting chrek agent",
+	agentLog.Info("Starting snapshot agent",
 		"node", cfg.NodeName,
 		"checkpoint_dir", cfg.BasePath,
 		"watch_namespace", cfg.RestrictedNamespace,

deploy/chrek/cmd/nsrestore/main.go → deploy/snapshot/cmd/nsrestore/main.go

@@ -8,8 +8,8 @@ import (
 
 	"github.com/go-logr/logr"
 
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/logging"
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/orchestrate"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/logging"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/orchestrate"
 )
 
 func main() {

deploy/chrek/go.mod → deploy/snapshot/go.mod

-module github.com/ai-dynamo/dynamo/deploy/chrek
+module github.com/ai-dynamo/dynamo/deploy/snapshot
 
 go 1.25.0
 

deploy/chrek/pkg/common/mounts.go → deploy/snapshot/pkg/common/mounts.go

@@ -10,7 +10,7 @@ import (
 	"github.com/moby/sys/mountinfo"
 	specs "github.com/opencontainers/runtime-spec/specs-go"
 
-	"github.com/ai-dynamo/dynamo/deploy/chrek/pkg/types"
+	"github.com/ai-dynamo/dynamo/deploy/snapshot/pkg/types"
 )
 
 // ReadMountInfo reads and parses mountinfo for a container process via /host/proc.