Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions charts/snapshot/templates/_helpers.tpl
Original file line number Diff line number Diff line change
Expand Up @@ -101,3 +101,6 @@ reads for rootfs-diff capture, and CRI-O config.json fallback).
{{- define "snapshot.runtimeStorageDir" -}}
{{- if eq .Values.runtime.type "crio" -}}/var/lib/containers{{- else -}}/var/lib/containerd{{- end -}}
{{- end }}

{{- define "snapshot.pageBrokerControlPath" -}}/pagebroker/control{{- end -}}
{{- define "snapshot.pageBrokerStagingPath" -}}/pagebroker/staging{{- end -}}
6 changes: 6 additions & 0 deletions charts/snapshot/templates/configmap.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,12 @@ data:
overlay:
exclusions: {{ toYaml .Values.config.overlay.exclusions | nindent 8 }}

pageBroker:
enabled: {{ .Values.pageBroker.enabled }}
{{- if .Values.pageBroker.enabled }}
controlSocketPath: {{ printf "%s/pagebroker.sock" (include "snapshot.pageBrokerControlPath" .) | quote }}
{{- end }}

restore:
restoreTimeoutSeconds: {{ .Values.config.restore.restoreTimeoutSeconds }}

Expand Down
30 changes: 30 additions & 0 deletions charts/snapshot/templates/daemonset.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,8 @@
{{- $socket := include "snapshot.runtimeSocket" . -}}
{{- $socketDir := $socket | dir -}}
{{- $storageDir := include "snapshot.runtimeStorageDir" . -}}
{{- $pageBrokerControlPath := include "snapshot.pageBrokerControlPath" . -}}
{{- $pageBrokerStagingPath := include "snapshot.pageBrokerStagingPath" . -}}
# SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: Apache-2.0

Expand Down Expand Up @@ -104,6 +106,24 @@ spec:
mountPath: /host-seccomp
{{- end }}
containers:
{{- if .Values.pageBroker.enabled }}
- name: pagebroker
image: "{{ .Values.pageBroker.image.repository }}:{{ .Values.pageBroker.image.tag }}"
imagePullPolicy: {{ .Values.pageBroker.image.pullPolicy }}
args:
- {{ printf "%s/pagebroker.sock" $pageBrokerControlPath | quote }}
- {{ $pageBrokerStagingPath | quote }}
- {{ .Values.storage.pvc.basePath | quote }}
- --max-concurrent-requests
- {{ .Values.pageBroker.maxConcurrentRequests | quote }}
resources:
{{- toYaml .Values.pageBroker.resources | nindent 12 }}
volumeMounts:
- name: pagebroker
mountPath: /pagebroker
- name: checkpoints
mountPath: {{ .Values.storage.pvc.basePath }}
{{- end }}
Comment thread
dfeigin-nv marked this conversation as resolved.
- name: agent
image: "{{ .Values.image.agent.repository }}:{{ .Values.image.agent.tag | default .Chart.AppVersion }}"
imagePullPolicy: {{ .Values.image.agent.pullPolicy }}
Expand All @@ -127,6 +147,10 @@ spec:
- name: config
mountPath: /etc/snapshot
readOnly: true
{{- if .Values.pageBroker.enabled }}
- name: pagebroker
mountPath: /pagebroker
{{- end }}
{{- if eq .Values.storage.type "pvc" }}
# Every agent owns direct access to the shared checkpoint store.
- name: checkpoints
Expand Down Expand Up @@ -173,6 +197,12 @@ spec:
- name: config
configMap:
name: {{ include "snapshot.fullname" . }}-config
{{- if .Values.pageBroker.enabled }}
- name: pagebroker
emptyDir:
medium: Memory
sizeLimit: {{ .Values.pageBroker.stagingSizeLimit | quote }}
{{- end }}
{{- if .Values.seccomp.deploy }}
# Seccomp profile ConfigMap (used by initContainer)
- name: seccomp-profiles
Expand Down
16 changes: 16 additions & 0 deletions charts/snapshot/values.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -75,6 +75,22 @@ storage:
oci:
uri: ""

pageBroker:
enabled: false
maxConcurrentRequests: 16
stagingSizeLimit: 64Gi
resources:
requests:
cpu: 8
memory: 64Gi
limits:
cpu: 32
memory: 72Gi
image:
repository: nvcr.io/nvidia/ai-dynamo/pagebroker
tag: 1.4.0
pullPolicy: Always

# DaemonSet configuration for the checkpoint/restore agent
daemonset:
# Agent and nsrestore log level (trace, debug, info, warn, error)
Expand Down