Changes Compiled

Makefile

@@ -53,7 +53,7 @@ endif
 OPERATOR_SDK_VERSION ?= v1.36.1
 
 # Image URL to use all building/pushing image targets
-IMG ?= slinky.slurm.net/slurm-operator:$(VERSION)
+IMG ?= ebennerv/slurm-operator:$(VERSION)
 
 # Set the namespaces that helm tests will run against
 SLURM_NAMESPACE ?= slurm

cmd/manager/main.go

@@ -21,10 +21,10 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/healthz"
 	"sigs.k8s.io/controller-runtime/pkg/log/zap"
 
-	slinkyv1alpha1 "github.com/SlinkyProject/slurm-operator/api/v1alpha1"
-	"github.com/SlinkyProject/slurm-operator/internal/controller/cluster"
-	"github.com/SlinkyProject/slurm-operator/internal/controller/nodeset"
-	"github.com/SlinkyProject/slurm-operator/internal/resources"
+	slinkyv1alpha1 "github.com/togethercomputer/slurm-operator/api/v1alpha1"
+	"github.com/togethercomputer/slurm-operator/internal/controller/cluster"
+	"github.com/togethercomputer/slurm-operator/internal/controller/nodeset"
+	"github.com/togethercomputer/slurm-operator/internal/resources"
 	//+kubebuilder:scaffold:imports
 )
 

cmd/webhook/main.go

@@ -20,7 +20,7 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/log/zap"
 	"sigs.k8s.io/controller-runtime/pkg/metrics/server"
 
-	slinkyv1alpha1 "github.com/SlinkyProject/slurm-operator/api/v1alpha1"
+	slinkyv1alpha1 "github.com/togethercomputer/slurm-operator/api/v1alpha1"
 	//+kubebuilder:scaffold:imports
 )
 

go.mod

@@ -1,8 +1,6 @@
-module github.com/SlinkyProject/slurm-operator
+module github.com/togethercomputer/slurm-operator
 
-go 1.22.0
-
-toolchain go1.22.5
+go 1.25.0
 
 require (
 	github.com/SlinkyProject/slurm-client v0.1.0-20241108145209-7bfdd4923f75
@@ -126,3 +124,5 @@ require (
 	sigs.k8s.io/structured-merge-diff/v4 v4.4.1 // indirect
 	sigs.k8s.io/yaml v1.4.0 // indirect
 )
+
+replace k8s.io/kubernetes => k8s.io/kubernetes v1.30.3

helm/slurm-operator/templates/operator/deployment.yaml

@@ -27,6 +27,8 @@ spec:
       hostname: {{ include "slurm-operator.name" . }}
       priorityClassName: {{ default "" .Values.priorityClassName }}
       serviceAccountName: {{ include "slurm-operator.operator.serviceAccountName" . }}
+      tolerations: {{- toYaml .Values.operator.tolerations | nindent 6 }}
+      affinity: {{- toYaml .Values.operator.affinity | nindent 6 }}
       {{- include "slurm-operator.imagePullSecrets" . | nindent 6 }}
       containers:
         - name: slurm-operator

helm/slurm-operator/templates/operator/rbac.yaml

@@ -79,6 +79,7 @@ rules:
   - get
   - list
   - watch
+  - update
 - apiGroups:
   - ""
   resources:

helm/slurm-operator/templates/webhook/deployment.yaml

@@ -29,6 +29,8 @@ spec:
       hostname: {{ include "slurm-operator.webhook.name" . }}
       priorityClassName: {{ default "" .Values.priorityClassName }}
       serviceAccountName: {{ include "slurm-operator.webhook.serviceAccountName" . }}
+      tolerations: {{- toYaml .Values.webhook.tolerations | nindent 6 }}
+      affinity: {{- toYaml .Values.webhook.affinity | nindent 6 }}
       {{- include "slurm-operator.imagePullSecrets" . | nindent 6 }}
       containers:
         - name: webhook

helm/slurm-operator/values.yaml

@@ -75,6 +75,11 @@ operator:
   # Ref: https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/#affinity-and-anti-affinity
   affinity: {}
   #
+  # -- (list)
+  # Configure pod tolerations.
+  # Ref: https://kubernetes.io/docs/concepts/scheduling-eviction/taint-and-toleration/
+  tolerations: []
+  #
   # -- (object)
   # Set container resource requests and limits for Kubernetes Pod scheduling.
   # Ref: https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/#resource-requests-and-limits-of-pod-and-container

helm/slurm/templates/_slurm.tpl

@@ -217,6 +217,27 @@ Define slurmd capabilities
 - SYS_NICE
 {{- end }}
 
+{{/*
+Determine login image repository
+*/}}
+{{- define "slurm.login.image.repository" -}}
+{{- .Values.login.image.repository | default (printf "%s/sackd" (include "slurm.image.repository" .)) -}}
+{{- end }}
+
+{{/*
+Define login image tag
+*/}}
+{{- define "slurm.login.image.tag" -}}
+{{- .Values.login.image.tag | default (include "slurm.image.tag" .) -}}
+{{- end }}
+
+{{/*
+Determine login image reference (repo:tag)
+*/}}
+{{- define "slurm.login.imageRef" -}}
+{{- printf "%s:%s" (include "slurm.login.image.repository" .) (include "slurm.login.image.tag" .) | quote -}}
+{{- end }}
+
 {{/*
 Define restapi name
 */}}
@@ -300,6 +321,30 @@ Define cluster secret name
 {{- printf "%s-token-%s" .Release.Name (include "slurm.user" .) -}}
 {{- end }}
 
+{{/*
+Define login name
+*/}}
+{{- define "slurm.login.name" -}}
+{{ printf "%s-login" .Release.Name }}
+{{- end }}
+
+{{/*
+Define login labels
+*/}}
+{{- define "slurm.login.labels" -}}
+app.kubernetes.io/component: login
+{{ include "slurm.login.selectorLabels" . }}
+{{ include "slurm.labels" . }}
+{{- end }}
+
+{{/*
+Define login selectorLabels
+*/}}
+{{- define "slurm.login.selectorLabels" -}}
+app.kubernetes.io/name: login
+app.kubernetes.io/instance: {{ .Release.Name }}
+{{- end }}
+
 {{/*
 Define token name
 */}}

helm/slurm/templates/accounting/accounting-statefulset.yaml

@@ -46,6 +46,10 @@ spec:
       affinity:
         {{- .Values.accounting.affinity | toYaml | nindent 8 }}
       {{- end }}{{- /* if .Values.accounting.affinity */}}
+      {{- if .Values.accounting.tolerations }}
+      tolerations:
+        {{- .Values.accounting.tolerations | toYaml | nindent 8 }}
+      {{- end }}{{- /* if .Values.accounting.tolerations */}}
       {{- include "slurm.imagePullSecrets" $ | nindent 6 }}
       initContainers:
         - name: init

helm/slurm/templates/compute/compute-nodeset.yaml

@@ -108,6 +108,8 @@ spec:
           ports:
             - name: slurmd
               containerPort: {{ include "slurm.compute.port" $ }}
+            - name: ssh
+              containerPort: 22
           startupProbe:
             exec:
               command:
@@ -135,6 +137,14 @@ spec:
             - name: {{ $vct.metadata.name }}
               mountPath: {{ $vct.spec.mountPath | default (printf "/mnt/%s" $vct.metadata.name ) }}
             {{- end }}{{- /* range $vct := $nodeset.volumeClaimTemplates */}}
+            - mountPath: /dev/shm
+              name: dshm
+            {{- if and $nodeset.persistence $nodeset.persistence.existingDataClaims }}
+            {{- range $nodeset.persistence.existingDataClaims }}
+            - name: {{ .name }}
+              mountPath: {{ .mountPath }}
+            {{- end }}{{- /* range $nodeset.persistence.existingDataClaims */}}
+            {{- end }}{{- /* if $nodeset.persistence.existingDataClaims */}}
       volumes:
         {{- include "slurm.volumes" $ | nindent 8 }}
         - name: slurm-config
@@ -143,6 +153,17 @@ spec:
             sources:
               - secret:
                   name: {{ include "slurm.auth.secretName" $ }}
+        - emptyDir:
+            medium: Memory
+            sizeLimit: {{ $nodeset.shmSize }}
+          name: dshm
+        {{- if and $nodeset.persistence $nodeset.persistence.existingDataClaims }}
+        {{- range $nodeset.persistence.existingDataClaims }}
+        - name: {{ .name }}
+          persistentVolumeClaim:
+            claimName: {{ .name }}
+        {{- end }}{{- /* range $nodeset.persistence.existingDataClaims */}}
+        {{- end }}{{- /* if $nodeset.persistence.existingDataClaims */}}
   {{- if $nodeset.volumeClaimTemplates }}
   volumeClaimTemplates:
     {{- $nodeset.volumeClaimTemplates | toYaml | nindent 4 }}

helm/slurm/templates/controller/controller-statefulset.yaml

@@ -39,6 +39,10 @@ spec:
       affinity:
         {{- .Values.controller.affinity | toYaml | nindent 8 }}
       {{- end }}{{- /* if .Values.controller.affinity */}}
+      {{- if .Values.controller.tolerations }}
+      tolerations:
+        {{- .Values.controller.tolerations | toYaml | nindent 8 }}
+      {{- end }}{{- /* if .Values.controller.tolerations */}}
       {{- include "slurm.imagePullSecrets" . | nindent 6 }}
       {{- if .Values.accounting.enabled }}
       securityContext:

helm/slurm/templates/login/login-deployment.yaml

@@ -0,0 +1,105 @@
+{{- /*
+SPDX-FileCopyrightText: Copyright (C) SchedMD LLC.
+SPDX-License-Identifier: Apache-2.0
+*/}}
+
+{{- if .Values.login.enabled }}
+apiVersion: apps/v1
+kind: Deployment
+metadata:
+  name: {{ include "slurm.login.name" . }}
+  namespace: {{ .Release.Namespace }}
+  labels:
+    {{- include "slurm.login.labels" . | nindent 4 }}
+spec:
+  replicas: {{ .Values.login.replicas | default 1 }}
+  selector:
+    matchLabels:
+      {{- include "slurm.login.selectorLabels" . | nindent 6 }}
+  template:
+    metadata:
+      annotations:
+        kubectl.kubernetes.io/default-container: login
+      labels:
+        {{- include "slurm.login.labels" . | nindent 8 }}
+    spec:
+      priorityClassName: {{ .Values.login.priorityClassName | default .Values.priorityClassName }}
+      automountServiceAccountToken: false
+      enableServiceLinks: false
+      dnsConfig:
+        {{- include "slurm.dnsConfig" . | nindent 8 }}
+      initContainers:
+        - name: init
+          image: {{ include "slurm.authcred.imageRef" . }}
+          imagePullPolicy: {{ .Values.authcred.imagePullPolicy | default (include "slurm.imagePullPolicy" .) }}
+          {{- with .Values.authcred.resources }}
+          resources:
+            {{- toYaml . | nindent 12 }}
+          {{- end }}{{- /* with .Values.authcred.resources */}}
+          env:
+            - name: SLURM_USER
+              value: {{ include "slurm.user" . }}
+          command:
+            - bash
+            - -c
+            - |
+              {{- range .Files.Lines "scripts/init.sh" }}
+              {{ . }}
+              {{- end }}{{- /* range .Files.Lines "scripts/init.sh" */}}
+          volumeMounts:
+            {{- include "slurm.init.volumeMounts" . | nindent 12 }}
+      containers:
+        - name: login
+          image: {{ include "slurm.login.imageRef" . }}
+          imagePullPolicy: {{ .Values.login.imagePullPolicy | default (include "slurm.imagePullPolicy" .) }}
+          {{- with .Values.login.securityContext }}
+          securityContext:
+            {{- toYaml . | nindent 12 }}
+          {{- end }}
+          env:
+            - name: SACKD_OPTIONS
+              value: -D --conf-server={{- printf "%s:%s" (include "slurm.controller.name" .) (include "slurm.controller.port" .) }}
+          ports:
+            - name: ssh
+              containerPort: 22
+          {{- with .Values.login.resources }}
+          resources:
+            {{- toYaml . | nindent 12 }}
+          {{- end }}{{- /* with .Values.login.resources */}}
+          readinessProbe:
+            exec:
+              command:
+                - test
+                - -S
+                - /run/slurm/sack.socket
+          volumeMounts:
+            {{- include "slurm.volumeMounts" . | nindent 12 }}
+            - name: authsocket
+              mountPath: /run/slurm
+            {{- with .Values.login.extraVolumeMounts }}
+            {{- toYaml . | nindent 12 }}
+            {{- end }}{{- /* range $vol := $.Values.login.extraVolumeMounts */}}
+      {{- with .Values.accounting.nodeSelector }}
+      nodeSelector:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}{{- /* with .Values.accounting.nodeSelector */}}
+      {{- with .Values.login.affinity }}
+      affinity:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}{{- /* with .Values.login.affinity */}}
+      {{- with .Values.login.tolerations }}
+      tolerations:
+        {{- toYaml . | nindent 8 }}
+      {{- end }}{{- /* with .Values.login.tolerations */}}
+      volumes:
+        {{- include "slurm.volumes" . | nindent 8 }}
+        - name: slurm-config
+          projected:
+            defaultMode: 0600
+            sources:
+              - secret:
+                  name: {{ include "slurm.auth.secretName" . }}
+        {{- with .Values.login.extraVolumes }}
+        {{- toYaml . | nindent 8 }}
+        {{- end }}{{- /* with .Values.login.volumes */}}
+{{- end }}{{- /* if .Values.login.enabled */}}

helm/slurm/templates/login/login-service.yaml

@@ -0,0 +1,18 @@
+{{- if .Values.login.enabled }}
+apiVersion: v1
+kind: Service
+metadata:
+  name: slurm-login
+  namespace: {{ .Release.Namespace }}
+  labels:
+    {{- include "slurm.login.labels" . | nindent 4 }}
+spec:
+  type: ClusterIP
+  selector:
+    {{- include "slurm.login.selectorLabels" . | nindent 4 }}
+  ports:
+    - name: ssh
+      port: 22
+      targetPort: ssh
+      protocol: TCP
+{{- end }}{{- /* if .Values.login.enabled */}}

helm/slurm/templates/restapi/restapi-deployment.yaml

@@ -38,6 +38,10 @@ spec:
       affinity:
         {{- .Values.restapi.affinity | toYaml | nindent 8 }}
       {{- end }}{{- /* if .Values.restapi.affinity */}}
+      {{- if .Values.restapi.tolerations }}
+      tolerations:
+        {{- .Values.restapi.tolerations | toYaml | nindent 8 }}
+      {{- end }}{{- /* if .Values.restapi.tolerations */}}
       {{- include "slurm.imagePullSecrets" . | nindent 6 }}
       initContainers:
         - name: init