Merge pull request #2935 from tpdownes/fix_a3_localssd

Update permissions for local SSD mounts to allow non-superuser write access
GoogleCloudPlatform · Aug 15, 2024 · 0ef79d3 · 0ef79d3
2 parents 2bb483b + da0a133
commit 0ef79d3
Show file tree

Hide file tree

Showing 4 changed files with 4 additions and 4 deletions.
diff --git a/community/examples/hpc-slurm-local-ssd-v5-legacy.yaml b/community/examples/hpc-slurm-local-ssd-v5-legacy.yaml
@@ -48,7 +48,7 @@ deployment_groups:
       local_ssd_filesystem:
         fs_type: ext4
         mountpoint: /mnt/localssd
-        permissions: "0755" # must quote numeric filesystem permissions!
+        permissions: "1777" # must quote numeric filesystem permissions!
 
   - id: compute_node_group
     source: community/modules/compute/schedmd-slurm-gcp-v5-node-group

diff --git a/community/examples/hpc-slurm-local-ssd.yaml b/community/examples/hpc-slurm-local-ssd.yaml
@@ -48,7 +48,7 @@ deployment_groups:
       local_ssd_filesystem:
         fs_type: ext4
         mountpoint: /mnt/localssd
-        permissions: "0755" # must quote numeric filesystem permissions!
+        permissions: "1777" # must quote numeric filesystem permissions!
 
   - id: nodeset
     source: community/modules/compute/schedmd-slurm-gcp-v6-nodeset

diff --git a/examples/machine-learning/a3-highgpu-8g/ml-slurm-a3-2-cluster.yaml b/examples/machine-learning/a3-highgpu-8g/ml-slurm-a3-2-cluster.yaml
@@ -115,7 +115,7 @@ deployment_groups:
       # mount local SSD filesystems
       local_ssd_filesystem:
         mountpoint: /mnt/localssd
-        permissions: 1777
+        permissions: "1777" # must quote numeric filesystem permissions!
       runners:
       - type: ansible-local
         destination: enable_nvidia_dcgm.yml

diff --git a/examples/machine-learning/a3-megagpu-8g/slurm-a3mega-cluster.yaml b/examples/machine-learning/a3-megagpu-8g/slurm-a3mega-cluster.yaml
@@ -87,7 +87,7 @@ deployment_groups:
       # mount local SSD filesystems
       local_ssd_filesystem:
         mountpoint: /mnt/localssd
-        permissions: "0755" # must quote numeric filesystem permissions!
+        permissions: "1777" # must quote numeric filesystem permissions!
       runners:
       - type: shell
         destination: setup_aperture.sh