Skip to content

Run only GPU tests for PR check #585

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Jul 2, 2024
Merged

Run only GPU tests for PR check #585

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a2f4fc5
Skip CPU tests for PR check
sutaakar Jul 2, 2024
262f23c
Upload KinD logs for PR check when Deploy CodeFlare stack step fails
sutaakar Jul 2, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 8 additions & 8 deletions .github/workflows/e2e_tests.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,6 +56,7 @@ jobs:
uses: ./common/github-actions/nvidia-gpu-setup

- name: Setup and start KinD cluster
id: kind-install
uses: ./common/github-actions/kind

- name: Install NVidia GPU operator for KinD
Expand Down Expand Up @@ -83,35 +84,34 @@ jobs:
export CODEFLARE_TEST_TIMEOUT_GPU_PROVISIONING=30m

export CODEFLARE_TEST_OUTPUT_DIR=${{ env.TEMP_DIR }}
echo "CODEFLARE_TEST_OUTPUT_DIR=${CODEFLARE_TEST_OUTPUT_DIR}" >> $GITHUB_ENV

set -euo pipefail
go test -timeout 60m -v ./test/e2e -json 2>&1 | tee ${CODEFLARE_TEST_OUTPUT_DIR}/gotest.log | gotestfmt
go test -timeout 60m -v -skip "^Test.*Cpu$" ./test/e2e -json 2>&1 | tee ${CODEFLARE_TEST_OUTPUT_DIR}/gotest.log | gotestfmt
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Curious to know if you prefer keeping two physical tests instead of one controlled with an environment variable. Is it for "reporting" purpose?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't have a strong opinion on that.
It is usable in downstream testing, where CPU and GPU tests are executed on different phases.
It also sufficiently distinguish those scenarios, so you can simply identify what option was executed from test name.

I kept it aligned with changes in opendatahub-io/distributed-workloads#176

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sounds good, was more to get your opinion, thanks.


- name: Print CodeFlare operator logs
if: always() && steps.deploy.outcome == 'success'
run: |
echo "Printing CodeFlare operator logs"
kubectl logs -n openshift-operators --tail -1 -l app.kubernetes.io/name=codeflare-operator | tee ${CODEFLARE_TEST_OUTPUT_DIR}/codeflare-operator.log
kubectl logs -n openshift-operators --tail -1 -l app.kubernetes.io/name=codeflare-operator | tee ${TEMP_DIR}/codeflare-operator.log

- name: Print Kueue operator logs
if: always() && steps.deploy.outcome == 'success'
run: |
echo "Printing Kueue operator logs"
KUEUE_CONTROLLER_POD=$(kubectl get pods -n kueue-system | grep kueue-controller | awk '{print $1}')
kubectl logs -n kueue-system --tail -1 ${KUEUE_CONTROLLER_POD} | tee ${CODEFLARE_TEST_OUTPUT_DIR}/kueue.log
kubectl logs -n kueue-system --tail -1 ${KUEUE_CONTROLLER_POD} | tee ${TEMP_DIR}/kueue.log

- name: Print KubeRay operator logs
if: always() && steps.deploy.outcome == 'success'
run: |
echo "Printing KubeRay operator logs"
kubectl logs -n ray-system --tail -1 -l app.kubernetes.io/name=kuberay | tee ${CODEFLARE_TEST_OUTPUT_DIR}/kuberay.log
kubectl logs -n ray-system --tail -1 -l app.kubernetes.io/name=kuberay | tee ${TEMP_DIR}/kuberay.log

- name: Export all KinD pod logs
uses: ./common/github-actions/kind-export-logs
if: always() && steps.deploy.outcome == 'success'
if: always() && steps.kind-install.outcome == 'success'
with:
output-directory: ${CODEFLARE_TEST_OUTPUT_DIR}
output-directory: ${TEMP_DIR}

- name: Upload logs
uses: actions/upload-artifact@v4
Expand All @@ -120,7 +120,7 @@ jobs:
name: logs
retention-days: 10
path: |
${{ env.CODEFLARE_TEST_OUTPUT_DIR }}/**/*.log
${{ env.TEMP_DIR }}/**/*.log

- name: Post notification about failure to a Slack channel in case of push event
if: failure() && github.event_name == 'push'
Expand Down