NVIDIA
NVIDIA
NVSM-Provision
Container
NVIDIA
NVIDIA
NVSM-Provision

NVIDIA System Management (NVSM) Provision is a part of multinode software framework for provisioning the aggregator and NVIDIA DGX nodes cluster.

NVIDIA System Management (NVSM) Provision is a part of multinode software framework for provisioning the aggregator and NVIDIA DGX nodes cluster.

Overview

The NVSM provision container is a key component in the NVSM multi-node architecture. It contains Ansible playbooks that handle the provisioning of both the aggregator node and compute nodes in the cluster.

Key Functions

  • Provisions the aggregator node and compute nodes in the cluster
  • Copies certificates from the aggregator node to compute nodes
  • Restarts compute nodes' NVSM services to establish connection with the aggregator node
  • Manages the multinode inventory file

Prerequisites

Before running the provision container, ensure:

  1. Docker and docker-compose are installed on the aggregator node:
    apt-get install docker
    apt-get install docker-compose
    
  2. All required containers are loaded:
    docker load -i nvsm-prometheus_25.03.05.tar.gz
    docker load -i nvsm-grafana_25.03.05.tar.gz
    docker load -i nvsm-provision_25.03.05.tar.gz
    docker load -i nvsm-aggregator_25.03.05.tar.gz
    

Inventory File Management

The provision container uses an inventory file (/etc/nvsm/aggregator/inventory.yaml) that contains:

  • BMC IP addresses
  • Host IP addresses
  • Passwords for compute nodes
  • Node-specific configurations

The inventory file can be encrypted using admin defined password.

Running the Provision Container

docker run -it --rm -v /etc/nvsm/aggregator:/etc/nvsm/mn nvcr.io/nvstaging/nvsm/nvsm-provision:25.03.05

Post-Provisioning

After provisioning:

  1. Check that all containers are running:
   sudo docker ps
  1. Verify NVSM services are active
   sudo docker exec nvsm-aggregator nvsm status
   sudo docker exec nvsm-aggregator nvsm_lifecycle status

Important Notes

  • The provisioning process restarts NVSM services on compute nodes
  • NVSM commands may error out during the initial NVSM INIT phase (3-5 minutes)
  • Ensure time synchronization between aggregator and compute nodes

References

Publisher
NVIDIA
NVIDIA
Latest Taglatest
UpdatedJanuary 30, 2026 UTC
Compressed Size271.48 MB
Multinode SupportNo
Multi-Arch SupportNo

NVIDIA uses cookies to improve your experience on our web site. We and our third-party partners also use cookies and other tools to collect and record information you provide as well as information about your interactions with our websites for performance improvement, analytics, and to assist in marketing efforts. By clicking "Accept All", you consent to our use of cookies and other tools as described in our Cookie Policy. You can manage your cookie settings by clicking on "Manage Settings." By continuing to use this site or by clicking one of the buttons below, you agree to our Terms of Service (which contains important waivers). Please see our Privacy Policy for more information on our privacy practices.