Add AlmaLinux 8 custom AMI support

CHANGELOG.md

@@ -7,6 +7,8 @@ CHANGELOG
 **ENHANCEMENTS**
 - Improve cluster update resiliency on login nodes by reusing the head-node-driven orchestration already in place on compute nodes,
   removing the dependency on cfn-hup and cfn-init.
+- Add support for AlmaLinux 8 as a `CustomAmi` created through the `build-image` process. ParallelCluster does not
+  publish an official AlmaLinux 8 AMI.
 
 **CHANGES**
 - The validator `ClusterNameValidator` now enforces cluster names to be limited to 40 characters when using `ExternalSlurmdbd`, 

cli/src/pcluster/constants.py

@@ -32,9 +32,18 @@
     "rocky8",
     "rhel9",
     "rocky9",
+    "almalinux8",
 ]
-UNSUPPORTED_OSES_FOR_MICRO_NANO = ["ubuntu2204", "ubuntu2404", "rhel8", "rocky8", "rhel9", "rocky9"]
-UNSUPPORTED_OSES_FOR_P6E_GB200 = ["rhel8", "rocky8", "rhel9"]
+UNSUPPORTED_OSES_FOR_MICRO_NANO = [
+    "ubuntu2204",
+    "ubuntu2404",
+    "rhel8",
+    "rocky8",
+    "rhel9",
+    "rocky9",
+    "almalinux8",
+]
+UNSUPPORTED_OSES_FOR_P6E_GB200 = ["rhel8", "rocky8", "rhel9", "almalinux8"]
 SUPPORTED_OSES_FOR_P6E_GB200 = list(set(SUPPORTED_OSES) - set(UNSUPPORTED_OSES_FOR_P6E_GB200))
 UNSUPPORTED_OSES_FOR_P6_B300 = []
 SUPPORTED_OSES_FOR_P6_B300 = list(set(SUPPORTED_OSES) - set(UNSUPPORTED_OSES_FOR_P6_B300))
@@ -65,6 +74,7 @@
     "rocky8": CR_PLATFORM_LINUX_UNIX,
     "rhel9": CR_PLATFORM_RHEL,
     "rocky9": CR_PLATFORM_LINUX_UNIX,
+    "almalinux8": CR_PLATFORM_LINUX_UNIX,
 }
 
 OS_MAPPING = {
@@ -75,6 +85,7 @@
     "rocky8": {"user": "rocky"},
     "rhel9": {"user": "ec2-user"},
     "rocky9": {"user": "rocky"},
+    "almalinux8": {"user": "ec2-user"},
 }
 
 OS_TO_IMAGE_NAME_PART_MAP = {
@@ -85,9 +96,10 @@
     "rocky8": "rocky8-hvm",
     "rhel9": "rhel9-hvm",
     "rocky9": "rocky9-hvm",
+    "almalinux8": "almalinux8-hvm",
 }
 # We do not publicly publish/release Parallelcluster AMI of below OSSes
-PRIVATE_OSES = ["rocky8", "rocky9"]
+PRIVATE_OSES = ["rocky8", "rocky9", "almalinux8"]
 
 IMAGE_NAME_PART_TO_OS_MAP = {value: key for key, value in OS_TO_IMAGE_NAME_PART_MAP.items()}
 

cli/src/pcluster/resources/imagebuilder/parallelcluster.yaml

@@ -30,6 +30,7 @@ phases:
                 rocky.8*) echo 'rocky8' ;;
                 rhel.9*) echo 'rhel9' ;;
                 rocky.9*) echo 'rocky9' ;;
+                almalinux.8*) echo 'almalinux8' ;;
                 *) echo "Unsupported OS: $RELEASE" && exit {{ FailExitCode }} ;;
               esac
 
@@ -55,14 +56,14 @@ phases:
               set -x
               # Get OS info
               . /etc/os-release
-              # Check input base AMI OS and get OS information, the output should be like amzn.2 | ubuntu.20.04 | ubuntu.22.04 | rhel.8.7 | rocky.8.8
+              # Check input base AMI OS and get OS information, the output should be like amzn.2023 | ubuntu.22.04 | rhel.8.10 | rocky.8.10 | almalinux.8.10
               RELEASE="${!ID}${!VERSION_ID:+.${!VERSION_ID}}"
 
               # Use OS name from previous step
               OS="{{ build.OperatingSystemName.outputs.stdout }}"
 
               # Set platform
-              [[ $OS =~ ^(alinux|rhel|rocky) ]] && PLATFORM='RHEL' || PLATFORM='DEBIAN'
+              [[ $OS =~ ^(alinux|almalinux|rhel|rocky) ]] && PLATFORM='RHEL' || PLATFORM='DEBIAN'
 
               # Set URLs
               COOKBOOK_NAME="aws-parallelcluster-cookbook-${CfnParamCookbookVersion}"
@@ -139,6 +140,7 @@ phases:
                 yum versionlock ${!KERNEL_PACKAGES_PREFIX} ${!KERNEL_PACKAGES_PREFIX}-core ${!KERNEL_PACKAGES_PREFIX}-modules ${!KERNEL_PACKAGES_PREFIX}-modules-extra
                 [[ $OS =~ rocky ]] && yum versionlock rocky-release rocky-repos
                 [[ $OS =~ rhel ]] && yum versionlock redhat-release
+                [[ $OS == almalinux8 ]] && yum versionlock almalinux-release almalinux-repos
               else
                 apt-get -y install linux-headers-$(uname -r) linux-modules-extra-$(uname -r)
                 apt-mark hold linux-aws linux-base linux-headers-aws linux-image-aws linux-modules-extra-aws
@@ -276,6 +278,7 @@ phases:
                 yum versionlock delete ${!KERNEL_PACKAGES_PREFIX} ${!KERNEL_PACKAGES_PREFIX}-core ${!KERNEL_PACKAGES_PREFIX}-modules ${!KERNEL_PACKAGES_PREFIX}-modules-extra
                 [[ $OS =~ rocky ]] && yum versionlock delete rocky-release rocky-repos
                 [[ $OS =~ rhel ]] && yum versionlock delete redhat-release
+                [[ $OS == almalinux8 ]] && yum versionlock delete almalinux-release almalinux-repos
               else
                 apt-mark unhold linux-aws linux-base linux-headers-aws linux-image-aws linux-modules-extra-aws
               fi

cli/src/pcluster/resources/imagebuilder/parallelcluster_tag.yaml

@@ -59,6 +59,8 @@ phases:
                 OS='rhel9'
               elif [ $(echo "${RELEASE}" | grep '^rocky\.9') ]; then
                 OS='rocky9'
+              elif [ $(echo "${RELEASE}" | grep '^almalinux\.8') ]; then
+                OS='almalinux8'
               fi
 
               echo ${OS}
@@ -74,7 +76,7 @@ phases:
               which aws
               if [[ $? -ne 0 ]]; then
                 echo "Installing unzip"
-                if [[ "{{ test.OperatingSystemName.outputs.stdout }}" =~ ^(alinux|rhel|rocky) ]]; then
+                if [[ "{{ test.OperatingSystemName.outputs.stdout }}" =~ ^(alinux|almalinux|rhel|rocky) ]]; then
                   yum -y install unzip
                 elif [[ "{{ test.OperatingSystemName.outputs.stdout }}" =~ ^ubuntu ]]; then
                   apt-get -y install unzip

cli/src/pcluster/resources/imagebuilder/parallelcluster_test.yaml

@@ -39,6 +39,7 @@ phases:
                 rocky.8*) echo rocky8;;
                 rhel.9*) echo rhel9;;
                 rocky.9*) echo rocky9;;
+                almalinux.8*) echo almalinux8;;
                 *) echo "Operating System '${RELEASE}' is not supported. Failing build." && exit 1;;
               esac
 
@@ -62,7 +63,7 @@ phases:
             - |
               set -v
               case '{{ test.OSName.outputs.stdout }}' in
-                alinux*|centos*|rhel*|rocky*) echo RHEL;;
+                alinux*|almalinux*|centos*|rhel*|rocky*) echo RHEL;;
                 ubuntu*) echo DEBIAN;;
               esac
 
@@ -90,7 +91,7 @@ phases:
               set -v
               ARCHITECTURE='{{ test.OSArchitecture.outputs.stdout }}'
               OS='{{ test.OSName.outputs.stdout }}'
-              if [ ${ARCHITECTURE} == 'arm64' ] && [[ ${OS} =~ ^(ubuntu(20|22)04|alinux(2|2023)|rhel8|rocky8|rhel9|rocky9)$ ]] || [ ${ARCHITECTURE} == 'x86_64' ]; then
+              if [ ${ARCHITECTURE} == 'arm64' ] && [[ ${OS} =~ ^(ubuntu(20|22)04|alinux(2|2023)|rhel8|rocky8|rhel9|rocky9|almalinux8)$ ]] || [ ${ARCHITECTURE} == 'x86_64' ]; then
                 echo "true"
               else
                 echo "false"
@@ -393,4 +394,4 @@ phases:
               cd /etc/chef/cookbooks/{{ loop.value }}
               inspec exec test --profiles-path . --controls /tag:install/ /tag:testami/ --no-distinct-exit
               [[ $? -ne 0 ]] && echo "InSpec tests failed: {{ loop.value }}" && exit 1
-              echo "InSpec tests passed: {{ loop.value }}"
+              echo "InSpec tests passed: {{ loop.value }}"

cli/src/pcluster/resources/imagebuilder/update_and_reboot.yaml

@@ -10,7 +10,7 @@ constants:
 phases:
   - name: build
     steps:
-      # Check input base AMI OS and get OS information, the output should be like amzn.2 | ubuntu.20.04 | ubuntu.22.04 | rhel.8.7 | rocky.8.8
+      # Check input base AMI OS and get OS information, the output should be like amzn.2023 | ubuntu.22.04 | rhel.8.10 | rocky.8.10 | almalinux.8.10
       - name: OperatingSystemRelease
         action: ExecuteBash
         inputs:
@@ -49,6 +49,8 @@ phases:
                 OS='rhel9'
               elif [ `echo "${!RELEASE}" | grep '^rocky\.9'` ]; then
                 OS='rocky9'
+              elif [ `echo "${!RELEASE}" | grep '^almalinux\.8'` ]; then
+                OS='almalinux8'
               else
                 echo "Operating System '${!RELEASE}' is not supported. Failing build."
                 exit {{ FailExitCode }}
@@ -65,7 +67,7 @@ phases:
                set -x
                OS='{{ build.OperatingSystemName.outputs.stdout }}'
 
-               if [ `echo "${!OS}" | grep -E '^(alinux|rhel|rocky)'` ]; then
+               if [ `echo "${!OS}" | grep -E '^(alinux|almalinux|rhel|rocky)'` ]; then
                  PLATFORM='RHEL'
                elif [ `echo "${!OS}" | grep -E '^ubuntu'` ]; then
                  PLATFORM='DEBIAN'
@@ -81,15 +83,15 @@ phases:
             - |
               set -x
               RELEASE='{{ build.OperatingSystemRelease.outputs.stdout }}'
-              if [ `echo "${!RELEASE}" | grep -Ev '^(amzn|ubuntu|rhel|rocky)'` ]; then
+              if [ `echo "${!RELEASE}" | grep -Ev '^(amzn|ubuntu|rhel|rocky|almalinux)'` ]; then
                 echo "This component does not support '${!RELEASE}'. Failing build."
                 exit {{ FailExitCode }}
               fi
 
-              # This component only supports aarch64 CPUs on Amazon Linux 2, Ubuntu2204, RHEL8, Rocky8, RHEL9 and Rocky9
+              # This component supports aarch64 CPUs on Amazon Linux 2023, Ubuntu, RHEL, Rocky Linux, and AlmaLinux 8
               ARCH=$(uname -m)
               if [[ `echo ${!ARCH}` == 'aarch64' ]]; then
-                if [ `echo "${!RELEASE}" | grep -Ev '^(amzn\.2|ubuntu\.20\.04|ubuntu\.22\.04|ubuntu\.24\.04|rhel\.8|rocky\.8|rhel\.9|rocky\.9)'` ]; then
+                if [ `echo "${!RELEASE}" | grep -Ev '^(amzn\.2023|ubuntu\.22\.04|ubuntu\.24\.04|rhel\.8|rocky\.8|rhel\.9|rocky\.9|almalinux\.8)'` ]; then
                   echo "This component does not support '${!RELEASE}' on ARM64 CPUs. Failing build."
                   exit {{ FailExitCode }}
                 fi
@@ -179,6 +181,8 @@ phases:
                     yum versionlock rocky-release rocky-repos
                   elif [[ ${!OS} == "rhel8" ]] || [[ ${!OS} == "rhel9" ]] ; then
                     yum versionlock redhat-release
+                  elif [[ ${!OS} == "almalinux8" ]] ; then
+                    yum versionlock almalinux-release almalinux-repos
                   fi
                   echo "Kernel version locked"
                 fi
@@ -197,7 +201,7 @@ phases:
               if [[ ${!PLATFORM} == RHEL ]]; then
                 yum -y update
 
-                if [[ ${!OS} == "rhel8" ]] || [[ ${!OS} == "rocky8" ]] || [[ ${!OS} == "rhel9" ]] || [[ ${!OS} == "rocky9" ]] || [[ ${!OS} == "alinux2023" ]] ; then
+                if [[ ${!OS} == "rhel8" ]] || [[ ${!OS} == "rocky8" ]] || [[ ${!OS} == "rhel9" ]] || [[ ${!OS} == "rocky9" ]] || [[ ${!OS} == "almalinux8" ]] || [[ ${!OS} == "alinux2023" ]] ; then
                   # package-cleanup has changed in RHEL8 and it works differently https://access.redhat.com/solutions/1227
                   yum remove -y $(yum repoquery --installonly --latest-limit=-2 -q)
                 else
@@ -243,6 +247,8 @@ phases:
                   yum versionlock delete rocky-release
                 elif [[ ${!OS} == "rhel8" ]] || [[ ${!OS} == "rhel9" ]] ; then
                   yum versionlock delete redhat-release
+                elif [[ ${!OS} == "almalinux8" ]] ; then
+                  yum versionlock delete almalinux-release almalinux-repos
                 fi
                 echo "Kernel version unlocked"
               fi

cli/src/pcluster/templates/cw_dashboard_builder.py

@@ -746,7 +746,9 @@ def _add_cw_log(self):
                 [
                     self._new_cw_log_widget(
                         title="system-messages",
-                        conditions=[Condition(["alinux2023", "rhel8", "rocky8", "rhel9", "rocky9"], base_os)],
+                        conditions=[
+                            Condition(["alinux2023", "rhel8", "rocky8", "rhel9", "rocky9", "almalinux8"], base_os)
+                        ],
                         filters=[self._new_filter(pattern=f"{head_private_ip}.*system-messages")],
                     ),
                     self._new_cw_log_widget(

cli/src/pcluster/validators/cluster_validators.py

@@ -51,7 +51,10 @@
 SHARED_STORAGE_NAME_MAX_LENGTH = 30
 NAME_REGEX = r"^[a-z][a-z0-9\-]*$"
 
-EFA_UNSUPPORTED_ARCHITECTURES_OSES = {"x86_64": [], "arm64": []}
+EFA_UNSUPPORTED_ARCHITECTURES_OSES = {
+    "x86_64": ["almalinux8"],
+    "arm64": ["almalinux8"],
+}
 
 EFS_MESSAGES = {
     "errors": {

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_disabled_efa_no_placement_group/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 The EC2 instance selected supports enhanced networking capabilities using Elastic Fabric Adapter (EFA). EFA enables you to run applications requiring high levels of inter-node communications at scale on AWS at no additional charge (https://docs.aws.amazon.com/parallelcluster/latest/ug/efa-v3.html).
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_efa_not_supported/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_enabled_efa_default_placement_group/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 The EC2 instance selected supports enhanced networking capabilities using Elastic Fabric Adapter (EFA). EFA enables you to run applications requiring high levels of inter-node communications at scale on AWS at no additional charge (https://docs.aws.amazon.com/parallelcluster/latest/ug/efa-v3.html).
 Enabling EFA requires compute instances to be placed within a Placement Group. Please specify an existing Placement Group name or leave it blank for ParallelCluster to create one.
 Allowed values for VPC ID:

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_enabled_efa_existing_placement_group/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 The EC2 instance selected supports enhanced networking capabilities using Elastic Fabric Adapter (EFA). EFA enables you to run applications requiring high levels of inter-node communications at scale on AWS at no additional charge (https://docs.aws.amazon.com/parallelcluster/latest/ug/efa-v3.html).
 Enabling EFA requires compute instances to be placed within a Placement Group. Please specify an existing Placement Group name or leave it blank for ParallelCluster to create one.
 Allowed values for VPC ID:

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_enabled_efa_non_existent_placement_group/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 The EC2 instance selected supports enhanced networking capabilities using Elastic Fabric Adapter (EFA). EFA enables you to run applications requiring high levels of inter-node communications at scale on AWS at no additional charge (https://docs.aws.amazon.com/parallelcluster/latest/ug/efa-v3.html).
 Enabling EFA requires compute instances to be placed within a Placement Group. Please specify an existing Placement Group name or leave it blank for ParallelCluster to create one.
 ERROR: non-existent-test-pg is not an acceptable value for Placement Group name

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_filtered_subnets_by_az/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_no_automation_no_errors/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_no_input_no_automation_no_errors/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_subnet_automation_no_errors/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_subnet_automation_no_errors_empty_vpc/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_vpc_automation_no_errors/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for Availability Zone:
 1. eu-west-1a
 2. eu-west-1b

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_vpc_automation_no_vpc_in_region/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 There are no VPC for the given region. Starting automatic creation of VPC and subnets...
 Allowed values for Availability Zone:
 1. eu-west-1a

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_vpc_automation_no_vpc_in_region_public/output.txt

@@ -36,6 +36,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 There are no VPC for the given region. Starting automatic creation of VPC and subnets...
 Allowed values for Availability Zone:
 1. eu-west-1a

cli/tests/pcluster/cli/configure/test_pcluster_configure/test_with_region_arg/output.txt

@@ -19,6 +19,7 @@ Allowed values for Operating System:
 5. rocky8
 6. rhel9
 7. rocky9
+8. almalinux8
 Allowed values for VPC ID:
   #  id            name                                 number_of_subnets
 ---  ------------  ---------------------------------  -------------------

cli/tests/pcluster/schemas/test_cluster_schema.py

@@ -69,6 +69,7 @@ def test_cluster_schema_slurm(mocker, test_datadir, config_file_name):
         (None, None, "Missing data for required field"),
         ("ubuntu2204", "ami-12345678", None),
         ("alinux2023", None, None),
+        ("almalinux8", "ami-12345678", None),
     ],
 )
 def test_image_schema(os, custom_ami, failure_message):
@@ -85,6 +86,9 @@ def test_image_schema(os, custom_ami, failure_message):
         image = ImageSchema().load(image_schema)
         assert_that(image.os).is_equal_to(os)
         assert_that(image.custom_ami).is_equal_to(custom_ami)
+        output_schema = ImageSchema()
+        output_schema.context = {"delete_defaults_when_dump": True}
+        assert_that(output_schema.dump(image)).is_equal_to(image_schema)
 
 
 @pytest.mark.parametrize(