-
Notifications
You must be signed in to change notification settings - Fork 115
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Signed-off-by: Ryan Cook <[email protected]>
- Loading branch information
1 parent
bc86300
commit 72ebe7c
Showing
1 changed file
with
8 additions
and
29 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -3,24 +3,6 @@ name: Training Bootc image builds | |
on: | ||
workflow_dispatch: | ||
|
||
pull_request: | ||
branches: | ||
- main | ||
paths: | ||
- .github/workflows/training_bootc.yaml | ||
- 'training/common/**' | ||
- 'training/amd-bootc/**' | ||
- 'training/intel-bootc/**' | ||
- 'training/nvidia-bootc/**' | ||
- 'training/intel-bootc/**' | ||
- 'training/instructlab/**' | ||
- 'training/vllm/**' | ||
# push: | ||
# branches: | ||
# - main | ||
# paths: | ||
# - .github/workflows/model_image_build_push.yaml | ||
|
||
concurrency: | ||
group: ${{ github.workflow }} | ||
cancel-in-progress: false | ||
|
@@ -52,7 +34,7 @@ jobs: | |
mode: start | ||
github-token: ${{ secrets.GH_PERSONAL_ACCESS_TOKEN }} | ||
ec2-image-id: ami-04cec38d48a5be576 | ||
ec2-instance-type: m7i.8xlarge | ||
ec2-instance-type: m7i.12xlarge | ||
subnet-id: subnet-0b1e1d94240813658 | ||
security-group-id: sg-055105753f5e8bd83 | ||
|
||
|
@@ -118,14 +100,18 @@ jobs: | |
arch: amd64 | ||
runs-on: ${{ needs.start-runner.outputs.label }} | ||
if: "!contains(github.event.pull_request.labels.*.name, 'hold-tests')" | ||
needs: nvidia-bootc-builder-image | ||
needs: [ nvidia-bootc-builder-image, start-runner ] | ||
steps: | ||
- uses: actions/[email protected] | ||
|
||
- name: mkdir root/.docker directory | ||
run: | | ||
mkdir -p ~/.docker | ||
- name: generate a ssh key - USER SHOULD INJECT THEIR OWN AND REBUILD IF THEY USE THIS IMAGE and overwrite the existing one | ||
run: | | ||
ssh-keygen -t rsa -b 4096 -f ~/.ssh/id_rsa -N "" <<<y | ||
- name: Login to Container Registry | ||
run: podman login -u ${{ secrets.REGISTRY_USER }} -p ${{ secrets.REGISTRY_PASSWORD }} ${{ env.REGISTRY }} | ||
|
||
|
@@ -162,15 +148,13 @@ jobs: | |
context: training/intel-bootc | ||
arch: amd64 | ||
gpu: intel | ||
pull-images: quay.io/ai-lab/vllm:latest quay.io/ai-lab/deepspeed-trainer:latest | ||
- image_name: amd-bootc | ||
context: training/amd-bootc | ||
arch: amd64 | ||
gpu: amd | ||
pull-images: quay.io/ai-lab/vllm:latest | ||
if: "!contains(github.event.pull_request.labels.*.name, 'hold-tests')" | ||
runs-on: ${{ needs.start-runner.outputs.label }} | ||
needs: nvidia-bootc-builder-image | ||
needs: [ start-runner, nvidia-bootc-builder-image ] | ||
continue-on-error: true | ||
steps: | ||
- uses: actions/[email protected] | ||
|
@@ -182,11 +166,6 @@ jobs: | |
- name: Login to Container Registry | ||
run: podman login -u ${{ secrets.REGISTRY_USER }} -p ${{ secrets.REGISTRY_PASSWORD }} ${{ env.REGISTRY }} | ||
|
||
- name: pull images | ||
id: pull_image | ||
working-directory: ${{ matrix.context }} | ||
run: podman pull ${{ matrix.pull-images }} | ||
|
||
- name: generate the local OCI assets | ||
run: | | ||
cd training | ||
|
@@ -232,7 +211,7 @@ jobs: | |
with: | ||
aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }} | ||
aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }} | ||
aws-region: ${{ env.REGION }} | ||
aws-region: ${{ env.AWS_REGION }} | ||
- name: Stop EC2 runner | ||
uses: machulav/ec2-github-runner@v2 | ||
with: | ||
|