Retire legacy PXE provisioning paths

README.md

@@ -51,7 +51,8 @@ The cluster nodes will follow the requirements described by Slurm or Kubernetes.
 - CentOS 7, 8
 - Red Hat Enterprise Linux / Rocky Linux 8 and 9 for the DGX software stack through the `nvidia-dgx` role
 
-You may also install a supported operating system on all servers via a 3rd-party solution (i.e. [MAAS](https://maas.io/), [Foreman](https://www.theforeman.org/)) or utilize the provided [OS install container](docs/pxe/minimal-pxe-container.md).
+You may also install a supported operating system on all servers via a 3rd-party solution such as [MAAS](https://maas.io/) or [Foreman](https://www.theforeman.org/), or via an existing site-standard automated installer.
+For new Ubuntu 24.04 or DGX OS 7 deployments, prefer Ubuntu autoinstall/cloud-init or MAAS and then apply DeepOps roles after the OS is present.
 For DGX platform software installation on top of vanilla Ubuntu or Red Hat family operating systems, see the [DGX software stack role guide](docs/deepops/dgx-software-stack.md).
 
 ### Kubernetes

config.example/group_vars/all.yml

@@ -253,7 +253,9 @@ maas_adminusers:
 maas_dns_domain: 'deepops.local'
 maas_region_controller: '192.168.1.1'
 maas_region_controller_url: 'http://{{ maas_region_controller }}:5240/MAAS'
-maas_repo: 'ppa:maas/3.5'
+# MAAS 3.7 is the current Ubuntu 24.04 line. Keep the 3.5 PPA when the MAAS
+# controller itself still runs Ubuntu 22.04.
+maas_repo: "{{ 'ppa:maas/3.7' if ansible_distribution_version is version('24.04', '>=') else 'ppa:maas/3.5' }}"
 
 # Defines if maas user should generate ssh keys
 # Usable for remote KVM/libvirt power actions

docs/airgap/mirror-apt-repos.md

@@ -212,7 +212,7 @@ deb http://archive.ubuntu.com/ubuntu noble-security main restricted universe mul
 deb http://archive.ubuntu.com/ubuntu noble-updates main restricted universe multiverse
 deb http://archive.ubuntu.com/ubuntu noble-proposed main restricted universe multiverse
 deb http://archive.ubuntu.com/ubuntu noble-backports main restricted universe multiverse
-deb http://ppa.launchpad.net/maas/3.5/ubuntu noble main
+deb http://ppa.launchpad.net/maas/3.7/ubuntu noble main
 deb http://archive.canonical.com/ubuntu noble partner
 
 deb-src http://archive.ubuntu.com/ubuntu noble main restricted universe multiverse

docs/k8s-cluster/README.md

@@ -35,7 +35,8 @@ Instructions for deploying a GPU cluster with Kubernetes
 
 1. Install a supported operating system on all nodes.
 
-   Install a supported operating system on all servers via a 3rd-party solution (i.e. [MAAS](https://maas.io/), [Foreman](https://www.theforeman.org/)) or utilize the provided [OS install container](../pxe).
+   Install a supported operating system on all servers via a 3rd-party solution such as [MAAS](https://maas.io/) or [Foreman](https://www.theforeman.org/), or via an existing site-standard automated installer.
+   For new Ubuntu 24.04 or DGX OS 7 deployments, prefer Ubuntu autoinstall/cloud-init or MAAS and then apply DeepOps after the OS is present.
 
 2. Set up your provisioning machine.
 

docs/k8s-cluster/roce-perf-k8s.md

@@ -71,7 +71,8 @@ add switch PFC, ECN configuration
 
 2. Install a supported operating system on all nodes.
 
-   Install a supported operating system on all servers utilizing the [DGXie](/docs/pxe/dgxie-container.md) provisioning container, via a 3rd-party solution (i.e. [MAAS](https://maas.io/), [Foreman](https://www.theforeman.org/)), or server BMC/console.
+   Install a supported operating system on all servers via a 3rd-party solution such as [MAAS](https://maas.io/) or [Foreman](https://www.theforeman.org/), via an existing site-standard automated installer, or through server BMC/console.
+   For new Ubuntu 24.04 or DGX OS 7 deployments, prefer Ubuntu autoinstall/cloud-init or MAAS.
 
    > NOTE: During OS installation, it is ideal if the identical user/password is configured. Otherwise, follow step 4 below to create an identical user across all nodes in the cluster.
 

docs/pxe/README.md

@@ -14,10 +14,10 @@ Most of the playbooks in DeepOps are agnostic to the OS install tooling, assumin
 For example, DeepOps can be used to deploy a [Slurm cluster](../slurm-cluster/) or a [Kubernetes cluster](../k8s-cluster) regardless of how the OS was installed.
 This makes it relatively easy to integrate with an existing datacenter environment.
 
-However, DeepOps does provide tooling for several PXE installation mechanisms which can be used if an existing tool isn't already deployed.
-These include:
+DeepOps does not try to replace a site provisioning system.
+For environments without an existing bare-metal provisioning workflow, DeepOps provides MAAS setup guidance:
 
 - [MAAS](./maas.md), an open-source bare-metal provisioning tool developed by [Canonical](https://canonical.com/)
-- [DGXIE](./dgxie-container.md), a containerized deployment tool developed specifically to deploy NVIDIA DGX OS
-  - [DGXIE on Kubernetes](./dgxie-on-k8s.md)
-- A minimal [PXE container](./minimal-pxe-container.md) which wraps [Pixiecore](https://github.com/danderson/netboot/tree/master/pixiecore), an open source tool for network booting
+
+For new Ubuntu 24.04 or DGX OS 7 cluster deployments, prefer MAAS, an existing site provisioning system, or Ubuntu autoinstall/cloud-init.
+NVIDIA DGX OS 7 supports installing the DGX Software Stack on regular Ubuntu 24.04 for cluster deployments, which is a better fit for current automated installation tooling than the retired legacy DGX OS installer workflows.