don't fail if RHOAI is not installed

bdattoma · Oct 25, 2024 · 58710b2 · 58710b2
1 parent 8b5622d
commit 58710b2
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 11 deletions.
diff --git a/ods_ci/tasks/Resources/Provisioning/GPU/AMD/amd_operator.sh b/ods_ci/tasks/Resources/Provisioning/GPU/AMD/amd_operator.sh
@@ -135,6 +135,13 @@ function wait_until_driver_image_is_built() {
 }
 
 function create_acceleratorprofile() {
+  echo "Creating AMD Accelerator Profile"
+  rhoai_ns=$(oc get namespace redhat-ods-applications --ignore-not-found  -oname)
+  if [ -n $rhoai_ns ];
+    then
+      echo "redhat-ods-applications namespace not found. Is RHOAI Installed? NVIDIA Accelerator Profile creation SKIPPED."
+      return 0
+  fi
   echo "Creating an Accelerator Profile for Dashboard"
   oc apply -f - <<EOF
   apiVersion: dashboard.opendatahub.io/v1
@@ -192,5 +199,5 @@ fi
 echo "Configuration of AMD GPU node and Operators completed"
 # the message appears in the logs, but the pod may get delete before our code next iteration checks the logs once again,
 # hence it'd fails to reach the pod. It happened to me
-# wait_while 1200 monitor_logs "$name" openshift-amd-gpu docker-build "Successfully pushed image-registry.openshift-image-registry.svc:5000/openshift-amd-gpu"
+wait_while 1200 monitor_logs "$name" openshift-amd-gpu docker-build "Successfully pushed image-registry.openshift-image-registry.svc:5000/openshift-amd-gpu"
 create_acceleratorprofile
diff --git a/ods_ci/tasks/Resources/Provisioning/GPU/NVIDIA/gpu_deploy.sh b/ods_ci/tasks/Resources/Provisioning/GPU/NVIDIA/gpu_deploy.sh
@@ -59,7 +59,13 @@ function rerun_accelerator_migration() {
   # 1. Delete the migration configmap
   # 2. Rollout restart dashboard deployment, so the configmap is created again and the migration run again
   # Context: https://github.com/opendatahub-io/odh-dashboard/issues/1938
-
+  echo "Creating NVIDIA Accelerator Profile via RHOAI Dashboard deployment rollout"
+  configmap=$(oc get configmap migration-gpu-status --ignore-not-found -n redhat-ods-applications -oname)
+  if [ -n $configmap ];
+    then
+      echo "migration-gpu-status not found. Is RHOAI Installed? NVIDIA Accelerator Profile creation SKIPPED."
+      return 0
+  fi
   echo "Deleting configmap migration-gpu-status"
   if ! oc delete configmap migration-gpu-status -n redhat-ods-applications;
     then
@@ -81,13 +87,13 @@ function rerun_accelerator_migration() {
   oc describe AcceleratorProfiles -n redhat-ods-applications
 }
 
-wait_until_pod_ready_status  "gpu-operator"
-oc apply -f "$GPU_INSTALL_DIR/../nfd_deploy.yaml"
-oc get csv -n nvidia-gpu-operator "$CSVNAME" -o jsonpath='{.metadata.annotations.alm-examples}' | jq .[0] > clusterpolicy.json
-oc apply -f clusterpolicy.json
-wait_until_pod_ready_status "nvidia-device-plugin-daemonset" 600
-wait_until_pod_ready_status "nvidia-container-toolkit-daemonset"
-wait_until_pod_ready_status "nvidia-dcgm-exporter"
-wait_until_pod_ready_status "gpu-feature-discovery"
-wait_until_pod_ready_status "nvidia-operator-validator"
+# wait_until_pod_ready_status  "gpu-operator"
+# oc apply -f "$GPU_INSTALL_DIR/../nfd_deploy.yaml"
+# oc get csv -n nvidia-gpu-operator "$CSVNAME" -o jsonpath='{.metadata.annotations.alm-examples}' | jq .[0] > clusterpolicy.json
+# oc apply -f clusterpolicy.json
+# wait_until_pod_ready_status "nvidia-device-plugin-daemonset" 600
+# wait_until_pod_ready_status "nvidia-container-toolkit-daemonset"
+# wait_until_pod_ready_status "nvidia-dcgm-exporter"
+# wait_until_pod_ready_status "gpu-feature-discovery"
+# wait_until_pod_ready_status "nvidia-operator-validator"
 rerun_accelerator_migration