Add static numa tomplogy for openshift depoyment

simplyblock_core/env_var

@@ -1,5 +1,5 @@
 SIMPLY_BLOCK_COMMAND_NAME=sbcli-dev
 SIMPLY_BLOCK_VERSION=19.2.30
 
-SIMPLY_BLOCK_DOCKER_IMAGE=public.ecr.aws/simply-block/simplyblock:R25.10-Hotfix
+SIMPLY_BLOCK_DOCKER_IMAGE=public.ecr.aws/simply-block/simplyblock:R25.10-Hotfix-os-fix
 SIMPLY_BLOCK_SPDK_ULTRA_IMAGE=public.ecr.aws/simply-block/ultra:R25.10-Hotfix-latest

simplyblock_core/scripts/charts/values.yaml

@@ -19,7 +19,7 @@ grafana:
 image:
   simplyblock: 
     repository: "public.ecr.aws/simply-block/simplyblock"
-    tag: "R25.10-Hotfix"
+    tag: "R25.10-Hotfix-os-fix"
     pullPolicy: "Always"
 
 ports:

simplyblock_web/templates/oc_storage_core_isolation.yaml.j2

@@ -33,89 +33,67 @@ spec:
           args:
             - |
               set -e
-
-              MARKER="/var/simplyblock/.cpu_isolation_applied"
-
-              echo "--- Installing jq ---"
-              apk add --no-cache jq
-
-              echo "--- Checking if node was already configured ---"
-
+              MARKER="/var/simplyblock/.cpu_topology_applied"
               if [[ -f "$MARKER" ]]; then
                   echo "[INFO] Node already configured. Skipping sleep and exiting..."
                   exit 0
               fi
 
-              echo "--- Reading isolated cores from config ---"
-              CONFIG_FILE="/var/simplyblock/sn_config_file"
-
-              if [[ ! -f "$CONFIG_FILE" ]]; then
-                  echo "[ERROR] Config file $CONFIG_FILE not found."
-                  exit 1
-              fi
-
-              ISOLATED_CORES=$(jq -r '.isolated_cores | join(",")' "$CONFIG_FILE")
-              if [[ -z "$ISOLATED_CORES" ]]; then
-                  echo "[ERROR] Could not extract isolated cores from $CONFIG_FILE"
-                  exit 1
-              fi
-
-              echo "[INFO] Isolated cores to apply: $ISOLATED_CORES"
-
+              SHORT_ID=$(echo "{{ HOSTNAME }}" | rev | cut -d'-' -f1 | rev)
+              # Create the Machine config pool
               cat <<EOF | kubectl apply -f -
               apiVersion: machineconfiguration.openshift.io/v1
               kind: MachineConfigPool
               metadata:
-                name: worker-isolated-{{ HOSTNAME }}
+                name: storage-$SHORT_ID
+                labels:
+                  custom-kubelet: cpumanager-enabled-$SHORT_ID
               spec:
                 machineConfigSelector:
                   matchExpressions:
                     - key: machineconfiguration.openshift.io/role
                       operator: In
                       values:
                         - worker
-                        - worker-isolated-{{ HOSTNAME }}
+                        - storage-$SHORT_ID
                 nodeSelector:
                   matchLabels:
                     kubernetes.io/hostname: {{ HOSTNAME }}
               EOF
 
-              # Create the MachineConfig for CPU isolation and IOMMU settings
+              # Create the Machine config
               cat <<EOF | kubectl apply -f -
               apiVersion: machineconfiguration.openshift.io/v1
               kind: MachineConfig
               metadata:
-                name: worker-isolated-{{ HOSTNAME }}
+                name: 00-storage-$SHORT_ID
                 labels:
-                  machineconfiguration.openshift.io/role: worker-isolated-{{ HOSTNAME }}
+                  machineconfiguration.openshift.io/role: storage-$SHORT_ID
               spec:
-                kernelArguments:
-                  - "nohz_full=${ISOLATED_CORES}"
-                  - "rcu_nocbs=${ISOLATED_CORES}"
-                  - "isolcpus=${ISOLATED_CORES}"
-                  - "intel_iommu=on"
-                  - "iommu=pt"
+                config:
+                  ignition:
+                    version: 3.2.0
               EOF
 
               # Create the KubeletConfig to enable static CPU manager
               cat <<EOF | kubectl apply -f -
               apiVersion: machineconfiguration.openshift.io/v1
               kind: KubeletConfig
               metadata:
-                name: set-static-cpu-manager-{{ HOSTNAME }}
+                name: kubelet-storage-$SHORT_ID
               spec:
                 machineConfigPoolSelector:
                   matchLabels:
-                    machineconfiguration.openshift.io/role: worker-isolated-{{ HOSTNAME }}
+                    custom-kubelet: cpumanager-enabled-$SHORT_ID
                 kubeletConfig:
                   cpuManagerPolicy: static
                   cpuManagerReconcilePeriod: 5s
+                  topologyManagerPolicy: single-numa-node
+                  reservedSystemCPUs: "0,1"
               EOF
 
-              echo "[INFO] Init setup and CPU isolation complete."
-
               echo "[INFO] Marking node as configured."
               touch "$MARKER"
 
-              echo "[INFO] Node is rebooting. Sleeping for 5 minutes to stop pipeline gracefully..."
+              echo "[INFO] Node is updating the kubeletconfig"
               sleep 300

simplyblock_web/templates/storage_deploy_spdk.yaml.j2

@@ -66,6 +66,13 @@ spec:
       volumeMounts:
         - name: foundationdb
           mountPath: /etc/foundationdb
+      resources:
+        requests:
+          cpu: "100m"
+          memory: "64Mi"
+        limits:
+          cpu: "100m"
+          memory: "64Mi"
 
   containers:
     - name: spdk-container
@@ -125,8 +132,10 @@ spec:
         limits:
           hugepages-2Mi: {{ MEM_GEGA }}Gi
           cpu: {{ CORES }}
+          memory: "8Gi"
         requests:
           hugepages-2Mi: {{ MEM_GEGA }}Gi
+          memory: "8Gi"
 
     - name: spdk-proxy-container
       image: {{ SIMPLYBLOCK_DOCKER_IMAGE }}
@@ -148,6 +157,13 @@ spec:
           value: "True"
         - name: TIMEOUT
           value: "300"
+      resources:
+        limits:
+          cpu: 1
+          memory: "128Mi"
+        requests:
+          cpu: 1
+          memory: "128Mi"
     {% if MODE == "docker" %}
     - name: fluentd
       image: public.ecr.aws/simply-block/fluentd-kubernetes-daemonset:v1.17.1-debian-graylog-1.2