fix (CI): Refactor SSH runners (#41991)

* Change ssh runner type

* Add wait step to SSH runner workflow

* Rename wait step to wait2 in ssh-runner.yml

* Remove wait step from ssh-runner.yml

Removed the wait step from the SSH runner workflow.

* Update runner type for single GPU A10 instance

* Update SSH runner version to 1.90.3

* Add sha256sum to ssh-runner workflow

* Update runner type and remove unused steps

Author: glegendre01

.github/workflows/ssh-runner.yml

@@ -4,7 +4,7 @@ on:
   workflow_dispatch:
     inputs:
       runner_type:
-        description: 'Type of runner to test (a10 or t4)'
+        description: 'Type of runner to test (a10)'
         required: true
       docker_image:
         description: 'Name of the Docker image'
@@ -36,14 +36,10 @@ jobs:
           NUM_GPUS: ${{ github.event.inputs.num_gpus }}
           RUNNER_TYPE: ${{ github.event.inputs.runner_type }}
         run: |
-          if [[ "$NUM_GPUS" == "single" && "$RUNNER_TYPE" == "t4" ]]; then
-            echo "RUNNER=aws-g4dn-4xlarge-cache" >> $GITHUB_ENV
-          elif [[ "$NUM_GPUS" == "multi" && "$RUNNER_TYPE" == "t4" ]]; then
-            echo "RUNNER=aws-g4dn-12xlarge-cache" >> $GITHUB_ENV
-          elif [[ "$NUM_GPUS" == "single" && "$RUNNER_TYPE" == "a10" ]]; then
-            echo "RUNNER=aws-g5-4xlarge-cache" >> $GITHUB_ENV
+          if [[ "$NUM_GPUS" == "single" && "$RUNNER_TYPE" == "a10" ]]; then
+            echo "RUNNER=aws-g5-4xlarge-cache-ssh" >> $GITHUB_ENV
           elif [[ "$NUM_GPUS" == "multi" && "$RUNNER_TYPE" == "a10" ]]; then
-            echo "RUNNER=aws-g5-12xlarge-cache" >> $GITHUB_ENV
+            echo "RUNNER=aws-g5-12xlarge-cache-ssh" >> $GITHUB_ENV
           else
             echo "RUNNER=" >> $GITHUB_ENV
           fi
@@ -61,8 +57,6 @@ jobs:
       group: ${{ needs.get_runner.outputs.RUNNER }}
     container:
       image: ${{ github.event.inputs.docker_image }}
-      options: --gpus all --privileged --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
-
     steps:
       - name: Update clone
         working-directory: /transformers
@@ -106,7 +100,7 @@ jobs:
           else
             echo "SLACKCHANNEL=${{ secrets.SLACK_CIFEEDBACK_CHANNEL }}" >> $GITHUB_ENV
           fi
-
+        
       - name: Tailscale # In order to be able to SSH when a test fails
         uses: huggingface/tailscale-action@main
         with: