Update SSH configrations. Add bastion server.

README.md

@@ -89,15 +89,15 @@ additional users `user1`, `user2`, and `user3` to the slurm cluster.
 slurm_users = [{
   name = "user1"
   uid = 1001
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user1@crusoe.ai"
+  ssh_pubkey = "ssh-ed25519 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA user1@crusoe.ai"
 }, {
   name = "user2"
   uid = 1002
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user2@crusoe.ai"
+  ssh_pubkey = "ssh-ed25519 uAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA ser2@crusoe.ai"
 }, {
   name = "user3"
   uid = 1003
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user3@crusoe.ai"
+  ssh_pubkey = "ssh-ed25519 AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA user3@crusoe.ai"
 }]
 ```
 

ansible/roles/nfs_server/tasks/main.yml

@@ -5,18 +5,18 @@
   when: item.mount == "/home"
   register: home_mount_exists
 
-- name: Create a ext4 filesystem on /dev/vdb
+- name: Create a xfs filesystem on /dev/vdb
   tags: nfs_server
   community.general.filesystem:
-    fstype: ext4
+    fstype: xfs
     dev: /dev/vdb
 
 - name: Mount vdb to /mnt/home
   tags: nfs_server
   ansible.posix.mount:
     path: /mnt/home
     src: /dev/vdb
-    fstype: ext4
+    fstype: xfs
     state: mounted
   when: home_mount_exists is skipped
 
@@ -30,15 +30,15 @@
   ansible.posix.mount:
     path: /mnt/home
     src: /dev/vdb
-    fstype: ext4
+    fstype: xfs
     state: unmounted
 
 - name: Mount vdb to /home
   tags: nfs_server
   ansible.posix.mount:
     path: /home
     src: /dev/vdb
-    fstype: ext4
+    fstype: xfs
     state: mounted
 
 - name: Install nfs-server

ansible/roles/raid0/files/raid0.sh

@@ -17,15 +17,15 @@ udevadm settle
 blkid -p -u filesystem /dev/md/ephemeral >> /dev/null
 res=$?
 if [ $res -eq 2 ]; then
-    echo "info: creating ext4 fs on md dev"
-    mkfs.ext4 /dev/md/ephemeral
+    echo "info: creating xfs fs on md dev"
+    mkfs.xfs /dev/md/ephemeral
     if [ $? -ne 0 ]; then
-        echo "error: failed to create ext4 fs on md dev"
+        echo "error: failed to create xfs fs on md dev"
         exit 1
     fi
 elif [ $res -ne 0 ]; then
     echo "error: failed to probe fs on md dev"
     exit 1
 else
-    echo "info: md dev is already formatted with an ext4 fs"
+    echo "info: md dev is already formatted with an xfs fs"
 fi

ansible/roles/raid0/templates/raid0.mount

@@ -5,7 +5,7 @@ After=raid0.service
 [Mount]
 What=/dev/md/ephemeral
 Where={{ raid0_mount_path }}
-Type=ext4
+Type=xfs
 Options=defaults
 
 [Install]

examples/h100-80gb-sxm-ib.tfvars

@@ -10,16 +10,10 @@ slurm_compute_node_ib_network_id = "474dbfe2-ffb3-44ce-87a4-cea659fddea3"
 slurm_compute_node_count = 4
 
 # slurm users configuration
-slurm_users = [{
-  name = "user1"
-  uid = 1001
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user1@crusoe.ai"
-}, {
-  name = "user2"
-  uid = 1002
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user2@crusoe.ai"
-}, {
-  name = "user3"
-  uid = 1003
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user3@crusoe.ai"
-}]
+slurm_users = [
+    {
+      name = "crusoe-admin"
+      uid = 75123
+      ssh_pubkey = "from=\"35.232.249.237\" ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIM//Zq9sHLhFsP1gPZ+NuiUh8/4Vh5SVtISEKtz+c4el admin@managed-slurm-bastion-prod"
+    }
+]

examples/l40s-48gb.tfvars

@@ -9,16 +9,10 @@ slurm_compute_node_type = "l40s-48gb.10x"
 slurm_compute_node_count = 4
 
 # slurm users configuration
-slurm_users = [{
-  name = "user1"
-  uid = 1001
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user1@crusoe.ai"
-}, {
-  name = "user2"
-  uid = 1002
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user2@crusoe.ai"
-}, {
-  name = "user3"
-  uid = 1003
-  ssh_pubkey = "ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIIjPRr0iVR4mgzJy0ehnM5hWX4O86hM1bVTgdi5g3nkZ user3@crusoe.ai"
-}]
+slurm_users = [
+    {
+      name = "crusoe-admin"
+      uid = 75123
+      ssh_pubkey = "from=\"35.232.249.237\" ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIM//Zq9sHLhFsP1gPZ+NuiUh8/4Vh5SVtISEKtz+c4el admin@managed-slurm-bastion-prod"
+    }
+]

variables.tf

@@ -124,7 +124,13 @@ variable "slurm_users" {
     uid = number
     ssh_pubkey = string
   }))
-  default     = []
+  default     = [
+    {
+      name = "crusoe-admin"
+      uid = 75123
+      ssh_pubkey = "from=\"35.232.249.237\" ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIM//Zq9sHLhFsP1gPZ+NuiUh8/4Vh5SVtISEKtz+c4el admin@managed-slurm-bastion-prod"
+    }
+  ]
 }
 
 variable "partitions" {