|
| 1 | +.. license-header |
| 2 | + SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved. |
| 3 | + SPDX-License-Identifier: Apache-2.0 |
| 4 | +
|
| 5 | + Licensed under the Apache License, Version 2.0 (the "License"); |
| 6 | + you may not use this file except in compliance with the License. |
| 7 | + You may obtain a copy of the License at |
| 8 | +
|
| 9 | + http://www.apache.org/licenses/LICENSE-2.0 |
| 10 | +
|
| 11 | + Unless required by applicable law or agreed to in writing, software |
| 12 | + distributed under the License is distributed on an "AS IS" BASIS, |
| 13 | + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 14 | + See the License for the specific language governing permissions and |
| 15 | + limitations under the License. |
| 16 | +
|
| 17 | +.. headings # #, * *, =, -, ^, " |
| 18 | +
|
| 19 | +
|
| 20 | +.. _install-network-operator-gov-ready: |
| 21 | + |
| 22 | +######################################## |
| 23 | +NVIDIA Network Operator Government Ready |
| 24 | +######################################## |
| 25 | + |
| 26 | +The NVIDIA Network Operator now offers government-ready components for NVIDIA AI Enterprise customers. |
| 27 | +Government ready is NVIDIA's designation for software that meets applicable security requirements for deployment in your FedRAMP High or equivalent sovereign use case. |
| 28 | +For more information on NVIDIA's government-ready support, refer to the white paper `AI Software for Regulated Environments <https://docs.nvidia.com/ai-enterprise/planning-resource/ai-software-regulated-environments-white-paper/latest/index.html>`_. |
| 29 | + |
| 30 | +===================================== |
| 31 | +Supported Network Operator Components |
| 32 | +===================================== |
| 33 | +The government-ready NVIDIA Network Operator includes the following components: |
| 34 | + |
| 35 | +.. _fn1: #base-image |
| 36 | +.. |fn1| replace:: :sup:`1` |
| 37 | + |
| 38 | +.. list-table:: |
| 39 | + :header-rows: 1 |
| 40 | + |
| 41 | + * - Component |
| 42 | + - Version |
| 43 | + * - NVIDIA Network Operator |
| 44 | + - |network-operator-version| |
| 45 | + * - NVIDIA Network Operator Init Container |
| 46 | + - |doca-init-container-version| |
| 47 | + * - DOCA-OFED Driver Container |
| 48 | + - |doca-driver-version-stig| |
| 49 | + * - RDMA Shared Device Plugin |
| 50 | + - |k8s-rdma-shared-dev-plugin-version| |
| 51 | + * - IP Over Infiniband (IPoIB) CNI plugin |
| 52 | + - |ipoib-cni-version| |
| 53 | + * - SRIOV Network Operator Config |
| 54 | + - |sriovnetop-version| |
| 55 | + * - SRIOV Network Operator Config Config Daemon |
| 56 | + - |sriovnetop-config-daemon-stig-version| |
| 57 | + * - SR-IOV Network Device Plugin |
| 58 | + - |sriovnetop-sriov-device-plugin-version| |
| 59 | + * - SR-IOV CNI plugin |
| 60 | + - |sriovnetop-sriov-cni-version| |
| 61 | + * - InfiniBand SR-IOV CNI plugin |
| 62 | + - |sriovnetop-ib-sriov-cni-version| |
| 63 | + * - K8s CNI network plugins |
| 64 | + - |cni-plugins-version| |
| 65 | + * - Multus CNI |
| 66 | + - |multus-version| |
| 67 | + * - RDMA CNI plugin |
| 68 | + - |rdma-cni-repository| |
| 69 | + |
| 70 | +:sup:`1` |
| 71 | +Hardened for STIG/FIPS compliance |
| 72 | + |
| 73 | +Artifacts for these components are available from the `NVIDIA NGC Catalog <https://registry.ngc.nvidia.com/orgs/nvidia/teams/mellanox/containers/doca-driver-stig-fips>`_. |
| 74 | + |
| 75 | +.. note:: |
| 76 | + |
| 77 | + Not all Network Operator components and features are available as government-ready containers in the v25.10.0 release. |
| 78 | + |
| 79 | + |
| 80 | +Validated Kubernetes Distributions |
| 81 | +=================================== |
| 82 | + |
| 83 | +The government-ready NVIDIA Network Operator has been validated on the following Kubernetes distributions: |
| 84 | + |
| 85 | +- Canonical Kubernetes 1.34 with Ubuntu Pro 24.04 and FIPS-compliant kernel |
| 86 | + |
| 87 | +Install Government-Ready NVIDIA Network Operator |
| 88 | +============================================= |
| 89 | + |
| 90 | +Once you have your :ref:`gov-ready-prerequisites` configured, use the following steps to install the NVIDIA Network Operator on Canonical Kubernetes distributions: |
| 91 | + |
| 92 | +#. :ref:`install-nfd` |
| 93 | +#. :ref:`create-ngc-api-pull-secret` |
| 94 | +#. :ref:`create-ubuntu-pro-token-secret` |
| 95 | +#. :ref:`deploy-nvidia-network-operator-gov-ready` |
| 96 | + |
| 97 | +.. _gov-ready-prerequisites: |
| 98 | + |
| 99 | +Prerequisites |
| 100 | +------------- |
| 101 | + |
| 102 | +- An active NVIDIA AI Enterprise subscription and NGC API token to access Network Operator government-ready containers. |
| 103 | + Refer to `Generating Your NGC API Key <https://docs.nvidia.com/ngc/gpu-cloud/ngc-user-guide/index.html#generating-api-key>`_ in the NVIDIA NGC User Guide for more information on NGC API tokens. |
| 104 | + |
| 105 | +- An Ubuntu Pro token for Canonical Kubernetes deployments. |
| 106 | + This token is required for the driver container to download kernel headers and other necessary packages from the Canonical repository when using the FIPS-enabled kernel on Ubuntu 24.04. |
| 107 | + Refer to the `Ubuntu Pro documentation <https://documentation.ubuntu.com/pro-client/en/v30/howtoguides/get_token_and_attach/>`_ for more information on accessing Ubuntu Pro tokens. |
| 108 | + |
| 109 | +- The ``helm`` CLI installed on a client machine. |
| 110 | + |
| 111 | + You can run the following commands to install the Helm CLI: |
| 112 | + |
| 113 | + .. code-block:: console |
| 114 | +
|
| 115 | + $ curl -fsSL -o get_helm.sh https://raw.githubusercontent.com/helm/helm/master/scripts/get-helm-3 \ |
| 116 | + && chmod 700 get_helm.sh \ |
| 117 | + && ./get_helm.sh |
| 118 | +
|
| 119 | +- A namespace to deploy the NVIDIA Network Operator. |
| 120 | + The example install commands below use ``nvidia-network-operator`` as the namespace. |
| 121 | + |
| 122 | +- Optionally, Service Mesh for intra-cluster traffic encryption. |
| 123 | + By default, the NVIDIA Network Operator does not encrypt traffic between its controller (and operands) and the Kubernetes API server. |
| 124 | + If you wish to encrypt this communication, you should deploy and maintain a service mesh application within the Kubernetes cluster to enable secure traffic. |
| 125 | + |
| 126 | +.. _install-nfd: |
| 127 | + |
| 128 | +Install Node Feature Discovery (NFD) |
| 129 | +------------------------------------- |
| 130 | + |
| 131 | +NFD is an open-source project that is a dependency for the Operator on each node in your cluster. |
| 132 | +It must be deployed before installing the NVIDIA Network Operator. |
| 133 | + |
| 134 | +Network Operator does not maintain a government ready version of NFD, it is recommended that you install the upstream NFD version that aligns with the :ref:`operator-component-matrix`. |
| 135 | +The NFD container is built on top of a scratch image, providing a highly secure container environment. |
| 136 | +For information on NFD CVEs and security updates, refer to the `NFD GitHub repository <https://github.com/kubernetes-sigs/node-feature-discovery/security>`_. |
| 137 | + |
| 138 | +Refer to the NFD documentation for `installation instructions <https://kubernetes-sigs.github.io/node-feature-discovery/stable/get-started/index.html>`_. |
| 139 | + |
| 140 | + |
| 141 | +.. _create-ngc-api-pull-secret: |
| 142 | + |
| 143 | +Create NGC API Pull Secret |
| 144 | +--------------------------- |
| 145 | + |
| 146 | +Add a Docker registry secret for downloading the Network Operator artifacts from NVIDIA NGC in the same namespace where you are planning to deploy the NVIDIA Network Operator. |
| 147 | +Update ``ngc-api-key`` in the command below with your NGC API key. |
| 148 | + |
| 149 | +.. code-block:: console |
| 150 | +
|
| 151 | + $ kubectl create secret -n nvidia-network-operator docker-registry ngc-secret \ |
| 152 | + --docker-server=nvcr.io \ |
| 153 | + --docker-username='$oauthtoken' \ |
| 154 | + --docker-password=<ngc-api-key> |
| 155 | +
|
| 156 | +.. _create-ubuntu-pro-token-secret: |
| 157 | + |
| 158 | +Create Ubuntu Pro Token Secret |
| 159 | +------------------------------- |
| 160 | + |
| 161 | +Create a Kubernetes secret to hold the value of your Ubuntu Pro token secret. |
| 162 | +This secret will be used in the install command in the next step. |
| 163 | + |
| 164 | +The Ubuntu Pro Token is required for the driver container to download kernel headers and other necessary packages from the Canonical repository when using the FIPS-enabled kernel on Ubuntu 24.04. |
| 165 | + |
| 166 | +.. _deploy-nvidia-network-operator-gov-ready: |
| 167 | + |
| 168 | +Install NVIDIA Network Operator Government-Ready Components |
| 169 | +-------------------------------------------------------- |
| 170 | + |
| 171 | +#. Label your ``nvidia-network-operator`` namespace for the Operator to set the enforcement policy to privilege. |
| 172 | + |
| 173 | + .. code-block:: console |
| 174 | +
|
| 175 | + $ kubectl label --overwrite ns nvidia-network-operator pod-security.kubernetes.io/enforce=privileged |
| 176 | +
|
| 177 | +#. Add the NVIDIA Helm repository: |
| 178 | + |
| 179 | + .. code-block:: console |
| 180 | +
|
| 181 | + $ helm repo add nvidia https://helm.ngc.nvidia.com/nvidia \ |
| 182 | + && helm repo update |
| 183 | +
|
| 184 | +#. Install the NVIDIA Network Operator with SR-IOV Network Operator. |
| 185 | + |
| 186 | + .. code-block:: console |
| 187 | +
|
| 188 | + $ helm install network-operator nvidia/network-operator \ |
| 189 | + --namespace nvidia-network-operator \ |
| 190 | + --set sriov-network-operator.images.sriovConfigDaemon=doca-driver-stig-fips \ |
| 191 | + --set sriov-network-operator.imagePullSecrets={ngc-secret} \ |
| 192 | + --set sriovNetworkOperator.enabled=true |
| 193 | + --set nfd.enabled=true |
| 194 | +
|
| 195 | +.. _update-ubuntu-pro-token-in-nicclusterpolicy: |
| 196 | + |
| 197 | +Update Ubuntu Pro Token in NicClusterPolicy |
| 198 | +========================================= |
| 199 | + |
| 200 | +``UBUNTU_PRO_TOKEN`` environment variable of the Network Operator NicClusterPolicy should be configured. |
| 201 | + |
| 202 | +.. code-block:: console |
| 203 | +
|
| 204 | + $ kubectl edit nicclusterpolicy nic-cluster-policy |
| 205 | +
|
| 206 | +Then update the secret with your new Ubuntu Pro Token. |
| 207 | +This token is required for the driver container to download kernel headers and other necessary packages from the Canonical repository when using the FIPS-enabled kernel on Ubuntu 24.04. |
| 208 | + |
0 commit comments