NVIDIA · michael-balint · Jul 27, 2026 · Jul 20, 2026
diff --git a/config.example/group_vars/slurm-cluster.yml b/config.example/group_vars/slurm-cluster.yml
@@ -71,7 +71,7 @@ slurm_cluster_install_nvidia_driver: yes
 # NFS                                                                          #
 ################################################################################
 # Default exports:
-# - /home: shared home directories, needed for Open OnDemand and best practice
+# - /home: shared home directories, best practice for a Slurm cluster
 # - /sw:   shared space for software installs, used for Spack or EasyBuild
 nfs_exports:
   - path: /home
@@ -159,24 +159,6 @@ slurm_cluster_install_openmpi: false
 openmpi_install_prefix: "/usr/local"
 openmpi_configure: "./configure --prefix={{ openmpi_install_prefix }} --disable-dependency-tracking --disable-getpwuid --with-pmix={{ pmix_install_prefix }} --with-hwloc={{ hwloc_install_prefix }} --with-pmi={{ slurm_install_prefix }} --with-slurm={{ slurm_install_prefix }} --with-libevent=/usr"
 
-################################################################################
-# Open OnDemand                                                                #
-################################################################################
-install_open_ondemand: no
-# The OOD Linux-host adapter runs interactive apps inside an Apptainer/Singularity
-# container. DeepOps no longer installs Singularity (the wrapper role is retired;
-# Enroot/Pyxis is the supported Slurm runtime), so install Apptainer/Singularity on
-# the login nodes yourself before enabling this adapter.
-ood_install_linuxhost_adapter: no
-
-servername: '{{ ansible_fqdn }}'
-httpd_port: 9050
-httpd_listen_addr_port:
-  - 9050
-httpd_use_rewrites: false
-node_uri: /node
-rnode_uri: /rnode
-
 ################################################################################
 # Allow the User Permission to Set GPU Clocks                                  #
 ################################################################################

diff --git a/docs/deepops/architecture.md b/docs/deepops/architecture.md
@@ -99,7 +99,6 @@ Roles run as part of the various playbooks. Includes default values and tasks as
 │   ├── nvidia-mig-manager
 │   ├── nvidia-network-operator
 │   ├── nvidia-peer-memory
-│   ├── ood-wrapper
 │   ├── openmpi
 │   ├── openshift
 │   ├── prometheus

diff --git a/docs/deepops/testing.md b/docs/deepops/testing.md
@@ -162,6 +162,6 @@ galaxy_info:
 
 The Slurm and Kubernetes deployment guides both document cluster verification steps. These should be run during the installation process to validate a GPU workload can be executed on the cluster.
 
-Additional services such as Kubeflow, Open OnDemand, or Monitoring may have additional validation steps that are documented in the corresponding DeepOps READMEs and the official documentation.
+Additional services such as Kubeflow or Monitoring may have additional validation steps that are documented in the corresponding DeepOps READMEs and the official documentation.
 
 For workloads that can be used as post-deployment validation, see the example workloads for [k8s](../../workloads/examples/k8s/) and [slurm](../../workloads/examples/slurm).
diff --git a/docs/misc/network-ports.md b/docs/misc/network-ports.md
@@ -52,4 +52,3 @@ or listing known ports that we configure within DeepOps itself.
   - Grafana web interface: 3000/tcp
   - Prometheus node exporter: 9100/tcp
   - DCGM node exporter: 9400/tcp
-- Open OnDemand: 9050/tcp
diff --git a/docs/slurm-cluster/README.md b/docs/slurm-cluster/README.md
@@ -187,7 +187,7 @@ Pay attention to the playbook output in the terminal. The NCCL compilation and s
 
 Now that Slurm is installed, try a ["Hello World" example using MPI](../../workloads/examples/slurm/mpi-hello/README.md).
 
-Read through the [slurm usage guide](slurm-usage.md) and [Open OnDemand guide](ood.md) for more information.
+Read through the [slurm usage guide](slurm-usage.md) for more information.
 
 ## Prolog and Epilog
 
@@ -232,9 +232,9 @@ For information about configuring a shared NFS filesystem on your Slurm cluster,
 You may optionally choose to install a tool for managing additional packages on your Slurm cluster.
 See the documentation on [software modules](./software-modules.md) for information on how to set this up.
 
-## Installing Open on Demand
+## Open OnDemand
 
-[Open OnDemand](https://openondemand.org/) can be installed by setting the `install_open_ondemand` variable to yes before running the `slurm-cluster.yml` playbook.
+DeepOps no longer bundles an Open OnDemand installer. Sites that want the [Open OnDemand](https://openondemand.org/) HPC portal on their Slurm cluster should deploy it directly from the upstream project, following its own [installation documentation](https://osc.github.io/ood-documentation/latest/installation.html).
 
 ## Pyxis and Enroot
 

diff --git a/docs/slurm-cluster/ood-images/ood-01.png b/docs/slurm-cluster/ood-images/ood-01.png
diff --git a/docs/slurm-cluster/ood-images/ood-02.png b/docs/slurm-cluster/ood-images/ood-02.png
diff --git a/docs/slurm-cluster/ood-images/ood-03.png b/docs/slurm-cluster/ood-images/ood-03.png
diff --git a/docs/slurm-cluster/ood-images/ood-04.png b/docs/slurm-cluster/ood-images/ood-04.png
diff --git a/docs/slurm-cluster/ood-images/ood-05.png b/docs/slurm-cluster/ood-images/ood-05.png
diff --git a/docs/slurm-cluster/ood-images/ood-06.png b/docs/slurm-cluster/ood-images/ood-06.png
diff --git a/docs/slurm-cluster/ood-images/ood-07.png b/docs/slurm-cluster/ood-images/ood-07.png
diff --git a/docs/slurm-cluster/ood-images/ood-08.png b/docs/slurm-cluster/ood-images/ood-08.png
diff --git a/docs/slurm-cluster/ood-images/ood-09.png b/docs/slurm-cluster/ood-images/ood-09.png
diff --git a/docs/slurm-cluster/ood.md b/docs/slurm-cluster/ood.md
diff --git a/playbooks/slurm-cluster.yml b/playbooks/slurm-cluster.yml
@@ -137,10 +137,6 @@
     hostlist: "{{ rsyslog_client_group | default('slurm-node') }}"
   when: slurm_enable_rsyslog_client|default(true)
 
-- name: Install Open OnDemand
-  import_playbook: slurm-cluster/open-ondemand.yml
-  when: install_open_ondemand
-
 - name: Set Permissions to adjust GPU Clocks speeds
   import_playbook: utilities/gpu-clocks.yml
   when: allow_user_set_gpu_clocks

diff --git a/playbooks/slurm-cluster/open-ondemand.yml b/playbooks/slurm-cluster/open-ondemand.yml
diff --git a/roles/ood-wrapper/defaults/main.yml b/roles/ood-wrapper/defaults/main.yml
diff --git a/roles/ood-wrapper/tasks/client.yml b/roles/ood-wrapper/tasks/client.yml
diff --git a/roles/ood-wrapper/tasks/linuxhost-adapter.yml b/roles/ood-wrapper/tasks/linuxhost-adapter.yml
diff --git a/roles/ood-wrapper/tasks/main.yml b/roles/ood-wrapper/tasks/main.yml