Add support for ParallelCluster 3.11.0 (#260)

* Add ParallelCluster 3.11.0 support

Resolves #258

* Add support for APC 3.11.0

Disable spack pyxis plugin so that external login nodes work.
The spack plugin has to be built on each login node OS and arch.

Related to #259

Resolves #258

source/cdk/config_schema.py

@@ -87,6 +87,9 @@
 #     * Upgrade Python to 3.9.19 (from 3.9.17).
 # 3.10.1:
 #     * Build fix for China regions
+# 3.11.0:
+#     * Add support for ap-southeast-3
+#     * login node enhancements
 MIN_PARALLEL_CLUSTER_VERSION = parse_version('3.6.0')
 # Update source/resources/default_config.yml with latest version when this is updated.
 PARALLEL_CLUSTER_VERSIONS = [
@@ -102,7 +105,16 @@
     '3.9.3',
     '3.10.0',
     '3.10.1',
+    '3.11.0',
 ]
+PARALLEL_CLUSTER_ENROOT_VERSIONS = {
+    # This can be found on the head node by running 'yum info enroot'
+    '3.11.0':  '3.4.1', # confirmed
+}
+PARALLEL_CLUSTER_PYXIS_VERSIONS = {
+    # This can be found on the head node at /opt/parallelcluster/sources
+    '3.11.0':  '0.20.0', # confirmed
+}
 PARALLEL_CLUSTER_MUNGE_VERSIONS = {
     # This can be found on the head node at /opt/parallelcluster/sources
     # This can be found on the head node at /etc/chef/local-mode-cache/cache/
@@ -118,6 +130,7 @@
     '3.9.3':   '0.5.15', # confirmed
     '3.10.0':  '0.5.16', # confirmed
     '3.10.1':  '0.5.16', # confirmed
+    '3.11.0':  '0.5.16', # confirmed
 }
 PARALLEL_CLUSTER_PYTHON_VERSIONS = {
     # This can be found on the head node at /opt/parallelcluster/pyenv/versions
@@ -133,6 +146,7 @@
     '3.9.3':   '3.9.17', # confirmed
     '3.10.0':  '3.9.19', # confirmed
     '3.10.1':  '3.9.19', # confirmed
+    '3.11.0':  '3.9.20', # confirmed
 }
 PARALLEL_CLUSTER_SLURM_VERSIONS = {
     # This can be found on the head node at /etc/chef/local-mode-cache/cache/
@@ -148,6 +162,7 @@
     '3.9.3':   '23.11.7', # confirmed
     '3.10.0':  '23.11.7', # confirmed
     '3.10.1':  '23.11.7', # confirmed
+    '3.11.0':  '23.11.10', # confirmed
 }
 PARALLEL_CLUSTER_PC_SLURM_VERSIONS = {
     # This can be found on the head node at /etc/chef/local-mode-cache/cache/
@@ -163,6 +178,7 @@
     '3.9.3':   '23-11-7-1', # confirmed
     '3.10.0':  '23-11-7-1', # confirmed
     '3.10.1':  '23-11-7-1', # confirmed
+    '3.11.0':  '23-11-10-1', # confirmed
 }
 SLURM_REST_API_VERSIONS = {
     '23-02-2-1': '0.0.39',
@@ -173,6 +189,7 @@
     '23-02-7-1': '0.0.39',
     '23-11-4-1': '0.0.39',
     '23-11-7-1': '0.0.39',
+    '23-11-10-1': '0.0.39',
 }
 
 def get_parallel_cluster_version(config):

source/resources/lambdas/CreateParallelCluster/CreateParallelCluster.py

@@ -150,8 +150,8 @@ def lambda_handler(event, context):
                 logger.info("Create call succeeded.")
                 logger.info(f"response={response}")
             except Exception as e:
-                logger.exception("ParallelCluster create exception. Ignoring exception so config stack deployment doesn't fail,")
-                message = f"ParallelCluster create exception:\n\n{e}"
+                logger.exception(f"ParallelCluster create exception. Ignoring exception so config stack deployment doesn't fail\n\nException:\n{e}\n\nException dict:\n{e.__dict__}")
+                message = f"ParallelCluster create exception:\n\nException:\n{e}\n\nException dict:\n{e.__dict__}"
                 sns_client.publish(
                     TopicArn = environ['ErrorSnsTopicArn'],
                     Subject = f"{cluster_name} CreateParallelCluster create failed",

source/resources/playbooks/roles/ParallelClusterHeadNode/tasks/config-pyxis.yml

@@ -0,0 +1,15 @@
+---
+# APC 3.11.0 added the pyxis Slurm Spank plugin
+# This needs to also be built on login nodes.
+# Currently, I am not doing that so until then, disable the Spank plugin by
+# renaming plugstack.conf
+
+- name: Rename plugstack.conf to plugstack.conf.back
+  shell:
+    cmd: |
+      set -ex
+
+      plugstack_conf=/opt/slurm/etc/plugstack.conf
+      if [[ -e  $plugstack_conf ]]; then
+          mv $plugstack_conf ${plugstack_conf}.back
+      fi

source/resources/playbooks/roles/ParallelClusterHeadNode/tasks/main.yml

@@ -7,3 +7,4 @@
 - { include_tasks: config-slurmrestd.yml, tags: slurmrestd }
 - { include_tasks: config-licenses.yml, tags: licenses }
 - { include_tasks: config-slurmdb-accounts.yml, tags: accounts }
+- { include_tasks: config-pyxis.yml }