Kserve Smoke Test #22
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
| name: "Kserve Smoke Test" | |
| on: | |
| schedule: | |
| - cron: "20 4 * * 1" # once a week | |
| workflow_dispatch: | |
| # push: | |
| # branches: [main] | |
| # paths: | |
| # - "!docs" | |
| # - "!demo" | |
| # pull_request: | |
| env: | |
| KIND_CLUSTER_NAME: "kserve-testing" | |
| ISVC_NAME: "caikit-tgis-isvc" | |
| jobs: | |
| smoke-test: | |
| name: Kserve Smoke Test | |
| runs-on: ubuntu-latest | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Free Disk Space | |
| uses: jlumbroso/[email protected] | |
| with: | |
| tool-cache: false | |
| android: true | |
| dotnet: true | |
| haskell: true | |
| large-packages: true | |
| swap-storage: true | |
| - name: Create k8s Kind Cluster | |
| uses: helm/[email protected] | |
| with: | |
| cluster_name: ${{ env.KIND_CLUSTER_NAME }} | |
| config: test/kserve/kind_config.yaml | |
| - name: Install kserve | |
| run: | | |
| # Use the hack/quick_install.sh script from the kserve repo | |
| curl https://raw.githubusercontent.com/opendatahub-io/kserve/v0.12.1.4/hack/quick_install.sh | bash | |
| - name: Patch Istio Ingressway | |
| run: | | |
| # bind the istio ingressway ports on the host so that we can access its services | |
| kubectl patch deployments.apps -o yaml -n istio-system istio-ingressgateway -p '{ | |
| "spec": | |
| {"template": | |
| {"spec": | |
| {"containers": | |
| [ | |
| { | |
| "name": | |
| "istio-proxy", | |
| "ports": | |
| [ | |
| { | |
| "containerPort":8080, | |
| "hostPort":80 | |
| }, | |
| { | |
| "containerPort":8443, | |
| "hostPort":443 | |
| } | |
| ] | |
| } | |
| ] | |
| } | |
| } | |
| } | |
| }' | |
| - name: Setup testing namespace | |
| run: | | |
| source demo/kserve/scripts/env.sh | |
| kubectl create namespace ${TEST_NS} | |
| kubectl config set-context "kind-${KIND_CLUSTER_NAME}" --namespace ${TEST_NS} | |
| - name: Setup flan-t5-small model volume | |
| run: | | |
| kubectl apply -f test/kserve/setup.yaml | |
| max_retries=10 | |
| wait_time=60s | |
| until kubectl wait --for=jsonpath='{.status.phase}'=Succeeded pod/setup-flan-t5-small --timeout ${wait_time}; do | |
| echo "Current status:" | |
| kubectl describe pod,pv,pvc | |
| max_retries=$((max_retries-1)) | |
| if [[ $max_retries -le 0 ]]; then | |
| echo "Failed to setup model" | |
| kubectl logs pod/setup-flan-t5-small --all-containers | |
| exit 1 | |
| fi | |
| echo "-------------------" | |
| done | |
| - name: Deploy ServingRuntime/InferenceService | |
| run: | | |
| kubectl apply -f test/kserve/caikit-tgis-serving.yaml | |
| max_retries=10 | |
| wait_time=60s | |
| until kubectl wait isvc/${ISVC_NAME} --for=condition=Ready --timeout=${wait_time}; do | |
| echo "Current status:" | |
| kubectl describe isvc,servingruntime,pod | |
| max_retries=$((max_retries-1)) | |
| if [[ $max_retries -le 0 ]]; then | |
| exit 1 | |
| fi | |
| echo "-------------------" | |
| done | |
| - name: Install caikit-nlp-client | |
| run: | | |
| pip install caikit-nlp-client | |
| - name: Perform test inference (http) | |
| run: | | |
| export ISVC_URL="$(oc get isvc ${ISVC_NAME} -o jsonpath='{.status.components.predictor.url}')" | |
| export ISVC_HOSTNAME=$(echo $ISVC_URL | cut -d/ -f3-) | |
| echo "Querying ISVC at: ${ISVC_URL}" | |
| # We can't query the service via hostname, we need to add the entry to /etc/hosts | |
| echo "127.0.0.1 ${ISVC_HOSTNAME}" | sudo tee -a /etc/hosts | |
| python test/smoke-test.py |