Skip to content

Build nightly image #52

Build nightly image

Build nightly image #52

Workflow file for this run

# NB: When run in a non-main branch (via workflow_dispatch), image scanning and distribution to other clouds does not happen
# on the basis that in this case a fatimage must be built and will be scanned.
name: Build nightly image
on:
workflow_dispatch:
inputs:
ci_cloud:
description: 'Select the CI_CLOUD'
required: true
type: choice
options:
- LEAFCLOUD
- SMS
- ARCUS
schedule:
- cron: '0 0 * * *' # Run at midnight
jobs:
build:
name: nightly-imagebuild
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}-${{ matrix.builds.label }}
cancel-in-progress: true
runs-on: ubuntu-22.04
strategy:
fail-fast: false # allow other matrix jobs to continue even if one fails
matrix:
builds:
- label: RL8-ofed
source_image_name: Rocky-8-GenericCloud-Base-8.9-20231119.0.x86_64.qcow2
inventory_groups: 'update,ofed'
- label: RL9-ofed
source_image_name: Rocky-9-GenericCloud-Base-9.4-20240523.0.x86_64.qcow2
inventory_groups: 'update,ofed'
- label: RL9-cuda
source_image_name: Rocky-9-GenericCloud-Base-9.4-20240523.0.x86_64.qcow2
inventory_groups: 'update,ofed,cuda'
env:
ANSIBLE_FORCE_COLOR: True
OS_CLOUD: openstack
CI_CLOUD: ${{ github.event.inputs.ci_cloud || vars.CI_CLOUD }}
IMAGE_VERSION: ${{ github.event_name == 'schedule' && 'latest' || github.ref_name }}
steps:
- uses: actions/checkout@v2
- name: Record settings for CI cloud
run: |
echo CI_CLOUD: ${{ env.CI_CLOUD }}
- name: Setup ssh
run: |
set -x
mkdir ~/.ssh
echo "${{ secrets[format('{0}_SSH_KEY', env.CI_CLOUD)] }}" > ~/.ssh/id_rsa
chmod 0600 ~/.ssh/id_rsa
shell: bash
- name: Add bastion's ssh key to known_hosts
run: cat environments/.stackhpc/bastion_fingerprints >> ~/.ssh/known_hosts
shell: bash
- name: Install ansible etc
run: dev/setup-env.sh
- name: Write clouds.yaml
run: |
mkdir -p ~/.config/openstack/
echo "${{ secrets[format('{0}_CLOUDS_YAML', env.CI_CLOUD)] }}" > ~/.config/openstack/clouds.yaml
shell: bash
- name: Setup environment
run: |
. venv/bin/activate
. environments/.stackhpc/activate
- name: Build fat image with packer
id: packer_build
run: |
set -x
. venv/bin/activate
. environments/.stackhpc/activate
cd packer/
packer init .
packer build \
-on-error=${{ vars.PACKER_ON_ERROR }} \
-var-file=$PKR_VAR_environment_root/${{ env.CI_CLOUD }}.pkrvars.hcl \
-var source_image_name=${{ matrix.builds.source_image_name }} \
-var image_name=${{ matrix.builds.label }} \
-var image_version=${{ env.IMAGE_VERSION }} \
-var inventory_groups=${{ matrix.builds.inventory_groups }} \
openstack.pkr.hcl
env:
PACKER_LOG: '1'
- name: Get image info and ensure it can be used for subsequent builds
id: manifest
run: |
. venv/bin/activate
IMAGE_ID=$(jq --raw-output '.builds[-1].artifact_id' packer/packer-manifest.json)
while ! openstack image show -f value -c name $IMAGE_ID; do
sleep 5
done
IMAGE_NAME=$(openstack image show -f value -c name $IMAGE_ID)
echo image: ${IMAGE_NAME} ${IMAGE_ID}
echo "image-name=${IMAGE_NAME}" >> "$GITHUB_OUTPUT"
echo "image-id=$IMAGE_ID" >> "$GITHUB_OUTPUT"
openstack image unset --property signature_verified $IMAGE_ID
- name: Download image
if: github.ref_name == 'main'
run: |
. venv/bin/activate
sudo mkdir /mnt/images
sudo chmod 777 /mnt/images
openstack image save --file /mnt/images/${{ steps.manifest.outputs.image-name }}.qcow2 ${{ steps.manifest.outputs.image-id }}
- name: Set up QEMU
if: github.ref_name == 'main'
uses: docker/setup-qemu-action@v3
- name: Install libguestfs
run: |
sudo apt -y update
sudo apt -y install libguestfs-tools
if: github.ref_name == 'main'
- name: Mount image
if: github.ref_name == 'main'
run: |
sudo mkdir -p './${{ steps.manifest.outputs.image-name }}'
sudo guestmount -a /mnt/images/${{ steps.manifest.outputs.image-name }}.qcow2 -i --ro -o allow_other './${{ steps.manifest.outputs.image-name }}'
- name: Run Trivy vulnerability scanner
if: github.ref_name == 'main'
uses: aquasecurity/[email protected]
with:
scan-type: fs
scan-ref: "${{ steps.manifest.outputs.image-name }}"
scanners: "vuln"
format: sarif
output: "${{ steps.manifest.outputs.image-name }}.sarif"
# turn off secret scanning to speed things up
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
- name: Upload Trivy scan results to GitHub Security tab
if: github.ref_name == 'main'
uses: github/codeql-action/upload-sarif@v3
with:
sarif_file: "${{ steps.manifest.outputs.image-name }}.sarif"
category: "${{ matrix.build.label }}"
- name: Fail if scan has CRITICAL vulnerabilities
if: github.ref_name == 'main'
uses: aquasecurity/[email protected]
with:
scan-type: fs
scan-ref: "${{ steps.manifest.outputs.image-name }}"
scanners: "vuln"
format: table
exit-code: '1'
severity: 'CRITICAL'
ignore-unfixed: true
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
- name: Delete new image if Trivy scan fails
if: github.ref_name == 'main' && failure() && steps.packer_build.outcome == 'success' # Runs if the Trivy scan found crit vulnerabilities or failed
run: |
. venv/bin/activate
echo "Deleting new image due to critical vulnerabilities or scan failure ..."
openstack image delete "${{ steps.manifest.outputs.image-id }}"
- name: Delete old image
if: github.ref_name == 'main'
run: |
. venv/bin/activate
IMAGE_COUNT=$(openstack image list --name ${{ steps.manifest.outputs.image-name }} -f value -c ID | wc -l)
if [ "$IMAGE_COUNT" -gt 1 ]; then
OLD_IMAGE_ID=$(openstack image list --sort created_at:asc --name "${{ steps.manifest.outputs.image-name }}" -f value -c ID | head -n 1)
echo "Deleting old image ID: $OLD_IMAGE_ID"
openstack image delete "$OLD_IMAGE_ID"
else
echo "Only one image exists, skipping deletion."
fi
upload:
name: upload-nightly-targets
needs: build
if: github.ref_name == 'main'
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}-${{ matrix.builds.label }}-${{ matrix.target_cloud }}
cancel-in-progress: true
runs-on: ubuntu-22.04
strategy:
fail-fast: false
matrix:
target_cloud:
- LEAFCLOUD
- SMS
- ARCUS
builds:
- image: RL8-ofed-latest
- image: RL9-ofed-latest
- image: RL9-cuda-latest
exclude:
- target_cloud: LEAFCLOUD # why?? Should this not be source_cloud/vars.CI_CLOUD
env:
OS_CLOUD: openstack
SOURCE_CLOUD: ${{ github.event.inputs.ci_cloud || vars.CI_CLOUD }}
TARGET_CLOUD: ${{ matrix.target_cloud }}
steps:
- uses: actions/checkout@v2
- name: Record settings for CI cloud
run: |
echo SOURCE_CLOUD: ${{ env.SOURCE_CLOUD }}
echo TARGET_CLOUD: ${{ env.TARGET_CLOUD }}
- name: Install openstackclient
run: |
python3 -m venv venv
. venv/bin/activate
pip install -U pip
pip install $(grep -o 'python-openstackclient[><=0-9\.]*' requirements.txt)
shell: bash
- name: Write clouds.yaml
run: |
mkdir -p ~/.config/openstack/
echo "${{ secrets[format('{0}_CLOUDS_YAML', env.SOURCE_CLOUD)] }}" > ~/.config/openstack/source_clouds.yaml
echo "${{ secrets[format('{0}_CLOUDS_YAML', env.TARGET_CLOUD)] }}" > ~/.config/openstack/target_clouds.yaml
shell: bash
- name: Download source image
run: |
. venv/bin/activate
export OS_CLIENT_CONFIG_FILE=~/.config/openstack/source_clouds.yaml
openstack image save --file ${{ matrix.builds.image }} ${{ matrix.builds.image }}
shell: bash
- name: Upload to target cloud
run: |
. venv/bin/activate
export OS_CLIENT_CONFIG_FILE=~/.config/openstack/target_clouds.yaml
openstack image create "${{ matrix.builds.image }}" \
--file "${{ matrix.builds.image }}" \
--disk-format qcow2 \
shell: bash
- name: Delete old latest image from target cloud
run: |
. venv/bin/activate
export OS_CLIENT_CONFIG_FILE=~/.config/openstack/target_clouds.yaml
IMAGE_COUNT=$(openstack image list --name ${{ matrix.builds.image }} -f value -c ID | wc -l)
if [ "$IMAGE_COUNT" -gt 1 ]; then
OLD_IMAGE_ID=$(openstack image list --sort created_at:asc --name "${{ matrix.builds.image }}" -f value -c ID | head -n 1)
openstack image delete "$OLD_IMAGE_ID"
else
echo "Only one image exists, skipping deletion."
fi
shell: bash