Skip to content
Test Slurm Cluster

refactor: multi-version Slurm support and improved developer experience #8

Sign in to view logs

refactor: multi-version Slurm support and improved developer experience

refactor: multi-version Slurm support and improved developer experience #8

Workflow file for this run

.github/workflows/test.yml at fff36a3
name: Test Slurm Cluster
on:
push:
branches: [ main ]
pull_request:
branches: [ main ]
workflow_dispatch: # Allow manual trigger
jobs:
# Build images for all supported versions
build:
runs-on: ubuntu-latest
timeout-minutes: 30
strategy:
matrix:
slurm_version: ['25.05.3', '24.11.6']
steps:
- name: Checkout code
uses: actions/checkout@v4
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Build Docker image for Slurm ${{ matrix.slurm_version }}
run: |
echo "SLURM_VERSION=${{ matrix.slurm_version }}" > .env
docker compose build
- name: Save Docker image
run: |
docker save slurm-docker-cluster:${{ matrix.slurm_version }} | gzip > slurm-${{ matrix.slurm_version }}.tar.gz
- name: Upload image artifact
uses: actions/upload-artifact@v4
with:
name: slurm-image-${{ matrix.slurm_version }}
path: slurm-${{ matrix.slurm_version }}.tar.gz
retention-days: 1
# Test cluster functionality on all versions
test-cluster:
needs: build
runs-on: ubuntu-latest
timeout-minutes: 30
strategy:
matrix:
slurm_version: ['25.05.3', '24.11.6']
steps:
- name: Checkout code
uses: actions/checkout@v4
- name: Download image artifact
uses: actions/download-artifact@v4
with:
name: slurm-image-${{ matrix.slurm_version }}
- name: Load Docker image
run: |
docker load < slurm-${{ matrix.slurm_version }}.tar.gz
docker images
- name: Configure version
run: |
echo "SLURM_VERSION=${{ matrix.slurm_version }}" > .env
cat .env
- name: Start cluster with 2 nodes
run: |
docker compose up -d
echo "Waiting for services to start..."
sleep 15
- name: Wait for MySQL to be healthy
run: |
timeout 60 bash -c 'until docker compose ps mysql | grep -q "healthy"; do
echo "Waiting for MySQL..."
sleep 2
done'
echo "MySQL is healthy"
- name: Wait for slurmdbd to be healthy
run: |
timeout 60 bash -c 'until docker compose ps slurmdbd | grep -q "healthy"; do
echo "Waiting for slurmdbd..."
sleep 2
done'
echo "slurmdbd is healthy"
- name: Wait for slurmctld to be healthy
run: |
timeout 60 bash -c 'until docker compose ps slurmctld | grep -q "healthy"; do
echo "Waiting for slurmctld..."
sleep 2
done'
echo "slurmctld is healthy"
- name: Wait for slurmrestd to be healthy
run: |
timeout 60 bash -c 'until docker compose ps slurmrestd | grep -q "healthy"; do
echo "Waiting for slurmrestd..."
sleep 2
done'
echo "slurmrestd is healthy"
- name: Wait for compute nodes to be healthy
run: |
timeout 90 bash -c 'until docker compose ps c1 | grep -q "healthy" && docker compose ps c2 | grep -q "healthy"; do
echo "Waiting for compute nodes..."
sleep 2
done'
echo "All compute nodes are healthy"
echo "Waiting for cluster to auto-register with slurmdbd..."
sleep 10
- name: Show cluster status
run: |
echo "=== Container Status ==="
docker compose ps
echo ""
echo "=== Slurm Info ==="
docker exec slurmctld sinfo
echo ""
echo "=== Node Details ==="
docker exec slurmctld scontrol show nodes
- name: Test REST API
run: |
echo "=== Testing REST API via Unix Socket ==="
# Auto-detect REST API version based on Slurm version
if [ "${{ matrix.slurm_version }}" == "24.11.6" ]; then
API_VERSION="v0.0.41"
else
API_VERSION="v0.0.42"
fi
echo "Using REST API version: $API_VERSION"
docker exec slurmrestd curl -s --unix-socket /var/run/slurmrestd/slurmrestd.socket \
http://localhost/slurm/${API_VERSION}/ping
echo ""
echo "=== REST API Jobs Query ==="
docker exec slurmrestd curl -s --unix-socket /var/run/slurmrestd/slurmrestd.socket \
http://localhost/slurm/${API_VERSION}/jobs | head -20
- name: Run test suite
run: chmod +x test_cluster.sh && ./test_cluster.sh
- name: Show logs on failure
if: failure()
run: |
echo "=== Docker Compose Logs ==="
docker compose logs
echo ""
echo "=== Container Status ==="
docker compose ps
- name: Cleanup
if: always()
run: docker compose down -v