Move topology-manager configuration to kubespray

The steps performed by the existing ansible playbook can be
performed directly by kubespray.  In addtion, fix and enable the
topology-manager.sh test.

Issue-ID: MULTICLOUD-1324
Signed-off-by: Todd Malsbary <todd.malsbary@intel.com>
Change-Id: Iee2197c1fc3e35288796399cccff0d3ae0925a6c

1 files changed, 112 insertions, 0 deletions

diff --git a/kud/tests/topology-manager-sriov.sh b/kud/tests/topology-manager-sriov.sh
new file mode 100755
index 00000000..447a7c83
--- /dev/null
+++ b/kud/tests/topology-manager-sriov.sh
@@ -0,0 +1,112 @@
+#!/bin/bash
+# SPDX-license-identifier: Apache-2.0
+##############################################################################
+# Copyright (c) 2020
+# All rights reserved. This program and the accompanying materials
+# are made available under the terms of the Apache License, Version 2.0
+# which accompanies this distribution, and is available at
+# http://www.apache.org/licenses/LICENSE-2.0
+##############################################################################
+
+set -o errexit
+set -o nounset
+set -o pipefail
+
+source _common.sh
+source _functions.sh
+
+sriov_capable_nodes=$(kubectl get nodes -o json | jq -r '.items[] | select((.status.capacity."intel.com/intel_sriov_nic"!=null) and ((.status.capacity."intel.com/intel_sriov_nic"|tonumber)>=2)) | .metadata.name')
+if [ -z "$sriov_capable_nodes" ]; then
+    echo "Ethernet adaptor version is not set. Topology manager test case cannot run on this machine"
+    exit 0
+else
+    echo "NIC card specs match. Topology manager option avaiable for this version."
+fi
+
+pod_name=pod-topology-manager
+csar_id=bd55cccc-bf34-11ea-b3de-0242ac130004
+
+function create_pod_yaml {
+    local csar_id=$1
+    _checks_args $csar_id
+    pushd ${CSAR_DIR}/${csar_id}
+
+    cat << POD > $pod_name.yaml
+kind: Pod
+apiVersion: v1
+metadata:
+  name: $pod_name
+  annotations:
+    k8s.v1.cni.cncf.io/networks: sriov-intel
+spec:
+  containers:
+  - name: $pod_name
+    image: docker.io/centos/tools:latest
+    command:
+    - /sbin/init
+    resources:
+      limits:
+        cpu: "1"
+        memory: "500Mi"
+        intel.com/intel_sriov_nic: '1'
+      requests:
+        cpu: "1"
+        memory: "500Mi"
+        intel.com/intel_sriov_nic: '1'
+POD
+    popd
+}
+
+create_pod_yaml ${csar_id}
+kubectl delete pod $pod_name --ignore-not-found=true --now --wait
+kubectl create -f ${CSAR_DIR}/${csar_id}/$pod_name.yaml --validate=false
+
+status_phase=""
+while [[ $status_phase != "Running" ]]; do
+    new_phase=$(kubectl get pods $pod_name | awk 'NR==2{print $3}')
+    if [[ $new_phase != $status_phase ]]; then
+        echo "$(date +%H:%M:%S) - $pod_name : $new_phase"
+        status_phase=$new_phase
+    fi
+    if [[ $new_phase == "Running" ]]; then
+        echo "Pod is up and running.."
+    fi
+    if [[ $new_phase == "Err"* ]]; then
+        exit 1
+    fi
+done
+
+uid=$(kubectl get pod pod-topology-manager -o jsonpath='{.metadata.uid}')
+node_name=$(kubectl get pod $pod_name -o jsonpath='{.spec.nodeName}')
+node_ip=$(kubectl get node $node_name -o jsonpath='{.status.addresses[].address}')
+
+apt-get install -y jq
+cpu_core=$(ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null $node_ip -- cat /var/lib/kubelet/cpu_manager_state | jq -r --arg UID "${uid}" --arg POD_NAME "${pod_name}" '.entries[$UID][$POD_NAME]')
+numa_node_number=$(ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null $node_ip -- lscpu | grep "NUMA node(s)" | awk -F ':' '{print $2}')
+for (( node=0; node<$numa_node_number; node++ )); do
+    ranges=$(ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null $node_ip -- lscpu | grep "NUMA node"$node | awk -F ':' '{print $2}')
+    ranges=(${ranges//,/ })
+    for range in ${ranges[@]}; do
+        min=$(echo $range | awk -F '-' '{print $1}')
+        max=$(echo $range | awk -F '-' '{print $2}')
+        if [ $cpu_core -ge $min ] && [ $cpu_core -le $max ]; then
+            cpu_numa_node=$node
+        fi
+    done
+done
+
+vf_pci=$(kubectl exec -it $pod_name -- env | grep PCIDEVICE_INTEL_COM_INTEL_SRIOV_NIC | awk -F '=' '{print $2}' | sed 's/\r//g')
+vf_numa_node=$(ssh -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null $node_ip -- cat /sys/bus/pci/devices/$vf_pci/numa_node)
+
+echo "The allocated cpu core is:" $cpu_core
+echo "The numa node of the allocated cpu core is:" $cpu_numa_node
+echo "The PCI address of the allocated vf is:" $vf_pci
+echo "The numa node of the allocated vf is:" $vf_numa_node
+if [ $cpu_numa_node == $vf_numa_node ]; then
+    echo "The allocated cpu core and vf are on the same numa node"
+else
+    echo "The allocated cpu core and vf are on different numa nodes"
+fi
+
+kubectl delete pod $pod_name --now
+echo "Test complete."