Merge pull request #104 from at88mph/resource-fix

Node Resource querying

Author: at88mph

helm/applications/skaha/Chart.yaml

@@ -21,13 +21,13 @@ type: application
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: 1.1.1
+version: 1.1.2
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "1.1.2"
+appVersion: "1.1.3"
 
 dependencies:
   - name: "redis"

helm/applications/skaha/README.md

@@ -42,8 +42,9 @@ A Helm chart to install the Skaha web service of the CANFAR Science Platform
 | deployment.skaha.sessions.expirySeconds | string | `"345600"` |  |
 | deployment.skaha.sessions.imagePullPolicy | string | `"Always"` |  |
 | deployment.skaha.sessions.initContainerImage | string | `"redis:8.2.2-bookworm"` |  |
+| deployment.skaha.sessions.nodeLabelSelector | string | `""` | Used to identify Kubernetes Worker Nodes when querying for resources available to User Sessions. |
 | deployment.skaha.sessions.kueue | object | `{}` |  |
-| deployment.skaha.sessions.maxCount | string | `"3"` |  |
+| deployment.skaha.sessions.maxCount | string | `"5"` | Maximum number of concurrent interactive user sessions.  Does not apply to `headless` User Sessions. |
 | deployment.skaha.sessions.maxEphemeralStorage | string | `"200Gi"` |  |
 | deployment.skaha.sessions.minEphemeralStorage | string | `"20Gi"` |  |
 | deployment.skaha.sessions.persistentVolumeClaimName | string | `"skaha-workload-cavern-pvc"` |  |

helm/applications/skaha/templates/skaha-tomcat-deployment.yaml

@@ -72,6 +72,10 @@ spec:
         - name: SKAHA_HEADLESS_PRIORITY_CLASS
           value: "{{ . }}"
         {{- end }}
+        {{- with .Values.deployment.skaha.sessions.nodeLabelSelector }}
+        - name: SKAHA_WORKER_NODE_LABEL_SELECTOR
+          value: "{{ . }}"
+        {{- end }}
         {{- with .Values.deployment.skaha.sessions.kueue }}
         {{- range $queueConfigKey, $queueConfig := . }}
         - name: SKAHA_QUEUE_{{ $queueConfigKey | upper }}_NAME

helm/applications/skaha/values.yaml

@@ -15,7 +15,7 @@ skahaWorkload:
 deployment:
   hostname: myhost.example.com  # Change this!
   skaha:
-    image: images.opencadc.org/platform/skaha:1.1.2
+    image: images.opencadc.org/platform/skaha:1.1.3
     imagePullPolicy: Always
 
     # Cron string for the image caching cron job schedule. Defaults to every half hour.
@@ -103,6 +103,19 @@ deployment:
       # initContainerImage:
       initContainerImage: "redis:8.2.2-bookworm"
 
+      # Optionally set the node label selector to identify Kubernetes Worker Nodes.  This is used to accurately query for available 
+      # resources from schedulable Nodes by eliminating, for example, Nodes that are only cordoned for Web APIs.
+      # Example:
+      #   nodeLabelSelector: "node-role.kubernetes.io/node-type=worker"
+      #
+      # Example (multiple labels ANDed):
+      #   nodeLabelSelector: "node-role.kubernetes.io/node-type=worker,environment=production"
+      #
+      #
+      # Example (multiple labels ORed):
+      #   nodeLabelSelector: "node-role.kubernetes.io/node-type in (worker,worker-gpu)"
+      nodeLabelSelector:
+
       # Optionally configure the Kueue system to handle large workloads.  Configurable per session type (e.g. desktop, notebook, etc.).
       # Leaving this empty will default to submitting Jobs to the base Kubernetes system.
       # Provide a default configuration for all session types, or omit and only configure specific session types.