sync q1_dev to staging

discovery/roles/slurm_config/defaults/main.yml

@@ -29,12 +29,11 @@ conf_dict_items: {}
 
 __default_config:
   cgroup:
-    # CgroupAutomount: true
     CgroupPlugin: autodetect
-    ConstrainCores: true
-    ConstrainDevices: true
-    ConstrainRAMSpace: true
-    ConstrainSwapSpace: true
+    ConstrainCores: 'yes'
+    ConstrainDevices: 'yes'
+    ConstrainRAMSpace: 'yes'
+    ConstrainSwapSpace: 'yes'
   slurm:
     SlurmUser: "{{ slurm_user }}"
     SlurmctldPort: 6817

examples/slurm_conf/cgroup.conf

@@ -0,0 +1,8 @@
+# Sample Cgroup Configuration
+# This is a sample configuration - customize according to your environment
+# For more information, see https://slurm.schedmd.com/cgroup.conf.html
+CgroupPlugin=autodetect
+ConstrainCores=yes
+ConstrainDevices=yes
+ConstrainRAMSpace=yes
+ConstrainSwapSpace=yes

examples/slurm_conf/slurm.conf

@@ -0,0 +1,64 @@
+# Sample Slurm Configuration File
+# Replace values marked with <PLACEHOLDER> with your actual values
+# This is a sample configuration - customize according to your environment
+
+# By default, Omnia merges custom configuration sources with defaults
+# and existing configurations to ensure a complete and valid setup.
+
+# For supported conf parameters, see https://slurm.schedmd.com/slurm.conf.html
+
+# CLUSTER IDENTITY
+ClusterName=slurm_cluster
+SlurmctldHost=<CONTROLLER_HOSTNAME>
+
+# AUTHENTICATION
+AuthType=auth/munge
+CredType=cred/munge
+
+# SLURM USER
+SlurmUser=slurm
+
+# DIRECTORIES AND FILES
+StateSaveLocation=/var/spool/slurmctld
+SlurmdSpoolDir=/var/spool/slurmd
+SlurmctldPidFile=/var/run/slurmctld.pid
+SlurmdPidFile=/var/run/slurmd.pid
+Epilog=/etc/slurm/epilog.sh
+
+# PORTS
+SlurmctldPort=6817
+SlurmdPort=6818
+
+# PLUGINS
+PluginDir=/usr/lib64/slurm
+ProctrackType=proctrack/cgroup
+PrologFlags=contain
+TaskPlugin=task/cgroup
+MpiDefault=none
+JobAcctGatherType=jobacct_gather/linux
+JobAcctGatherFrequency=30
+
+# SCHEDULING
+SchedulerType=sched/backfill
+SelectType=select/cons_tres
+
+# TIMEOUTS
+SlurmctldTimeout=120
+SlurmdTimeout=300
+
+# PARAMETERS
+ReturnToService=2
+SlurmctldParameters=enable_configless
+
+# ACCOUNTING (Optional)
+AccountingStorageHost=<SLURMDBD_HOSTNAME>
+AccountingStoragePort=6819
+AccountingStorageType=accounting_storage/slurmdbd
+
+# COMPUTE NODES
+NodeName=<NODE_HOSTNAME> Sockets=2 CoresPerSocket=8 ThreadsPerCore=2 RealMemory=32000 State=UNKNOWN
+
+# PARTITIONS
+# Define at least one partition
+PartitionName=DEFAULT Nodes=ALL MaxTime=INFINITE State=UP
+PartitionName=normal Nodes=<NODE_LIST> Default=YES MaxTime=INFINITE State=UP

examples/slurm_conf/slurmdbd.conf

@@ -0,0 +1,23 @@
+# Sample SlurmDBD Configuration File
+# Replace values marked with <PLACEHOLDER> with your actual values
+# This is a sample configuration - customize according to your environment
+# For more information, see https://slurm.schedmd.com/slurmdbd.conf.html
+
+# Authentication
+AuthType=auth/munge
+SlurmUser=slurm
+
+# Database Daemon Configuration
+DbdHost=<DBD_HOST>
+DbdPort=6819
+LogFile=/var/log/slurm/slurmdbd.log
+PidFile=/var/run/slurmdbd.pid
+PluginDir=/usr/lib64/slurm
+
+# Database Connection
+StorageType=accounting_storage/mysql
+StorageHost=<DB_HOST>
+StoragePort=3306
+StorageLoc=slurm_acct_db
+StorageUser=slurm
+StoragePass=<db_password>

input/config/aarch64/rhel/10.0/openmpi.json

@@ -4,7 +4,11 @@
       { "package": "openmpi",
         "type": "tarball",
         "url": "https://download.open-mpi.org/release/open-mpi/v{{ openmpi_version.split('.')[:2] | join('.') }}/openmpi-{{ openmpi_version }}.tar.gz"
-      }
+      },
+      {"package": "pmix-devel", "type": "rpm", "repo_name": "aarch64_appstream"},
+      {"package": "munge-devel", "type": "rpm", "repo_name": "aarch64_codeready-builder"},
+      {"package": "gcc-c++", "type": "rpm", "repo_name": "aarch64_appstream"},
+      {"package": "make", "type": "rpm", "repo_name": "aarch64_appstream"}
     ]
   }
 }

input/config/aarch64/rhel/10.0/slurm_custom.json

@@ -2,11 +2,9 @@
     "slurm_custom": {
         "cluster": [
             {"package": "munge", "type": "rpm", "repo_name": "aarch64_appstream"},
-            {"package": "munge-devel", "type": "rpm", "repo_name": "aarch64_codeready-builder"},
             {"package": "firewalld", "type": "rpm", "repo_name": "aarch64_baseos"},
             {"package": "python3-firewall", "type": "rpm", "repo_name": "aarch64_baseos"},
             {"package": "pmix", "type": "rpm", "repo_name": "aarch64_appstream"},
-            {"package": "pmix-devel", "type": "rpm", "repo_name": "aarch64_appstream"},
             {"package": "nvcr.io/nvidia/hpc-benchmarks", "tag": "25.09", "type": "image"},
             {"package": "apptainer", "type": "rpm", "repo_name": "epel" },
 	        {"package": "doca-ofed", "type": "rpm_repo", "repo_name": "doca" }

input/omnia_config.yml

@@ -78,6 +78,9 @@
 #    <mapping> or <filepath>
 # <mapping> Supply the configuration values directly as a key–value map
 # <filepath> Supply the absolute path to a custom configuration file
+#            This path can be any path inside the omnia_core container.
+#            The default input path "/opt/omnia/input/project_default" 
+#            can also be used to place the custom conf files
 # The conf files supported by slurm are
 # slurm
 # cgroup
@@ -135,10 +138,9 @@ slurm_cluster:
     #   OR
 
     # config_sources:
-    #   slurm: /path/to/custom_slurm.conf
-    #   cgroup: /path/to/custom_cgroup.conf
-    #   slurmdbd: /path/to/custom_slurmdbd.conf
-    #   gres: /path/to/custom_gres.conf
+    #   slurm: /opt/omnia/input/project_default/slurm.conf
+    #   cgroup: /opt/omnia/input/project_default/cgroup.conf
+    #   slurmdbd: /opt/omnia/input/project_default/slurmdbd.conf
 
 # ----------------------------SERVICE K8S------------------------------------------------------
 # For service k8s cluster below parameters are required,(List)