Skip to content

Commit 3e74d90

Browse files
committed
check if nvidia driver installed and then if peermem is loaded. if not load the module nvidia_peermem
1 parent a5cd3e0 commit 3e74d90

File tree

10 files changed

+4
-134
lines changed

10 files changed

+4
-134
lines changed

playbooks/new_nodes.yml

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -50,7 +50,6 @@
5050
when: cluster_network|bool
5151
- include_role:
5252
name: nvidia_peermem
53-
when: cluster_network|bool
5453

5554
- hosts: bastion,slurm_backup,login,compute
5655
become: true

playbooks/resize_add.yml

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -48,7 +48,6 @@
4848
when: cluster_network|bool
4949
- include_role:
5050
name: nvidia_peermem
51-
when: cluster_network|bool
5251

5352
- hosts: bastion,slurm_backup,login,compute
5453
become: true

playbooks/roles/nvidia_peermem/.travis.yml

Lines changed: 0 additions & 29 deletions
This file was deleted.

playbooks/roles/nvidia_peermem/README.md

Lines changed: 0 additions & 38 deletions
This file was deleted.

playbooks/roles/nvidia_peermem/handlers/main.yml

Lines changed: 0 additions & 2 deletions
This file was deleted.

playbooks/roles/nvidia_peermem/meta/main.yml

Lines changed: 0 additions & 52 deletions
This file was deleted.
Lines changed: 4 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,15 +1,16 @@
11
---
22
- name: Check if nvidia drivers are installed
3-
shell: lsmod | grep nvidia_peermem | wc -l
4-
register: result
3+
shell: cat /sys/module/nvidia/version | wc -l
4+
register: gpu
55

66

77
- name: Check if nvidia_peermem module is loaded
88
shell: lsmod | grep nvidia_peermem | wc -l
99
register: result
10+
when: gpu.stdout == '1'
1011

1112

1213
- name: Load nvidia_peermem module
1314
become: true
1415
shell: modprobe nvidia_peermem
15-
when: register.stdout != 3
16+
when: gpu.stdout == '1' and result.stdout != '3'

playbooks/roles/nvidia_peermem/tests/inventory

Lines changed: 0 additions & 2 deletions
This file was deleted.

playbooks/roles/nvidia_peermem/tests/test.yml

Lines changed: 0 additions & 5 deletions
This file was deleted.

playbooks/site.yml

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -52,7 +52,6 @@
5252
when: cluster_network|bool
5353
- include_role:
5454
name: nvidia_peermem
55-
when: cluster_network|bool
5655

5756
- hosts: bastion
5857
become: true

0 commit comments

Comments
 (0)