Merge pull request #600 from GoogleCloudPlatform/release-candidate

Release v1.6.0

Author: heyealex

Makefile

@@ -14,11 +14,21 @@ ENG = ./cmd/... ./pkg/...
 TERRAFORM_FOLDERS=$(shell find ./modules ./community/modules ./tools -type f -name "*.tf" -not -path '*/\.*' -exec dirname "{}" \; | sort -u)
 PACKER_FOLDERS=$(shell find ./modules ./community/modules ./tools -type f -name "*.pkr.hcl" -not -path '*/\.*' -exec dirname "{}" \; | sort -u)
 
+ifneq (, $(shell which git))
+## GIT IS PRESENT
+ifneq (,$(wildcard .git))
+## GIT DIRECTORY EXISTS
+GIT_TAG_VERSION=$(shell git tag --points-at HEAD)
+GIT_BRANCH=$(shell git branch --show-current)
+GIT_COMMIT_INFO=$(shell git describe --tags --dirty --long)
+endif
+endif
+
 # RULES MEANT TO BE USED DIRECTLY
 
 ghpc: warn-go-version warn-terraform-version warn-packer-version $(shell find ./cmd ./pkg ghpc.go -type f)
 	$(info **************** building ghpc ************************)
-	go build ghpc.go
+	@go build -ldflags="-X 'main.gitTagVersion=$(GIT_TAG_VERSION)' -X 'main.gitBranch=$(GIT_BRANCH)' -X 'main.gitCommitInfo=$(GIT_COMMIT_INFO)'" ghpc.go
 
 install-user:
 	$(info ******** installing ghpc in ~/bin *********************)

README.md

@@ -204,6 +204,59 @@ In the right side, expand the Filters view and then filter by label, specifying
 
 ## Troubleshooting
 
+### Network is unreachable (Slurm V5)
+
+Slurm requires access to google APIs to function. This can be achieved through one of the following methods:
+
+1. Create a [Cloud NAT](https://cloud.google.com/nat) (preferred).
+2. Setting `disable_controller_public_ips: false` &
+   `disable_login_public_ips: false` on the controller and login nodes
+   respectively.
+3. Enable
+   [private access to Google APIs](https://cloud.google.com/vpc/docs/private-access-options).
+
+By default the Toolkit VPC module will create an associated Cloud NAT so this is
+typically seen when working with the pre-existing-vpc module. If no access
+exists you will see the following errors:
+
+When you ssh into the login node or controller you will see the following
+message:
+
+```text
+*** Slurm setup failed! Please view log: /slurm/scripts/setup.log ***
+```
+
+> **_NOTE:_**: Many different potential issues could be indicated by the above
+> message, so be sure to verify issue in logs.
+
+To confirm the issue, ssh onto the controller and call `sudo cat /slurm/scripts/setup.log`. Look for
+the following logs:
+
+```text
+google_metadata_script_runner: startup-script: ERROR: [Errno 101] Network is unreachable
+google_metadata_script_runner: startup-script: OSError: [Errno 101] Network is unreachable
+google_metadata_script_runner: startup-script: ERROR: Aborting setup...
+google_metadata_script_runner: startup-script exit status 0
+google_metadata_script_runner: Finished running startup scripts.
+```
+
+You may also notice mount failure logs on the login node:
+
+```text
+INFO: Waiting for '/usr/local/etc/slurm' to be mounted...
+INFO: Waiting for '/home' to be mounted...
+INFO: Waiting for '/opt/apps' to be mounted...
+INFO: Waiting for '/etc/munge' to be mounted...
+ERROR: mount of path '/usr/local/etc/slurm' failed: <class 'subprocess.CalledProcessError'>: Command '['mount', '/usr/local/etc/slurm']' returned non-zero exit status 32.
+ERROR: mount of path '/opt/apps' failed: <class 'subprocess.CalledProcessError'>: Command '['mount', '/opt/apps']' returned non-zero exit status 32.
+ERROR: mount of path '/home' failed: <class 'subprocess.CalledProcessError'>: Command '['mount', '/home']' returned non-zero exit status 32.
+ERROR: mount of path '/etc/munge' failed: <class 'subprocess.CalledProcessError'>: Command '['mount', '/etc/munge']' returned non-zero exit status 32.
+```
+
+> **_NOTE:_**: The above logs only indicate that something went wrong with the
+> startup of the controller. Check logs on the controller to be sure it is a
+> network issue.
+
 ### Failure to Create Auto Scale Nodes (Slurm)
 
 If your deployment succeeds but your jobs fail with the following error:

cmd/root.go

@@ -23,8 +23,16 @@ import (
 	"github.com/spf13/cobra"
 )
 
+// Git references when use Makefile
 var (
-	rootCmd = &cobra.Command{
+	GitTagVersion string
+	GitBranch     string
+	GitCommitInfo string
+)
+
+var (
+	annotation = make(map[string]string)
+	rootCmd    = &cobra.Command{
 		Use:   "ghpc",
 		Short: "A blueprint and deployment engine for HPC clusters in GCP.",
 		Long: `gHPC provides a flexible and simple to use interface to accelerate
@@ -34,12 +42,28 @@ HPC deployments on the Google Cloud Platform.`,
 				log.Fatalf("cmd.Help function failed: %s", err)
 			}
 		},
-		Version: "v1.5.0",
+		Version:     "v1.6.0",
+		Annotations: annotation,
 	}
 )
 
 // Execute the root command
 func Execute() error {
+	if len(GitCommitInfo) > 0 {
+		if len(GitTagVersion) == 0 {
+			GitTagVersion = "- not built from oficial release"
+		}
+		if len(GitBranch) == 0 {
+			GitBranch = "detached HEAD"
+		}
+		annotation["version"] = GitTagVersion
+		annotation["branch"] = GitBranch
+		annotation["commitInfo"] = GitCommitInfo
+		rootCmd.SetVersionTemplate(`ghpc version {{index .Annotations "version"}}
+Built from '{{index .Annotations "branch"}}' branch.
+Commit info: {{index .Annotations "commitInfo"}}
+`)
+	}
 	return rootCmd.Execute()
 }
 

community/examples/AMD/README.md

@@ -75,20 +75,15 @@ remounted and that you should logout and login. Follow its instructions.
 Once configuration is complete, install AOCC by running:
 
 ```shell
-sudo -i bash /var/tmp/install_aocc.sh
+sudo bash /var/tmp/install_aocc.sh
 ```
 
 Spack will prompt you to accept the AOCC End User License Agreement by opening a
 text file containing information about the license. Leave the file unmodified
 and write it to disk by typing `:q` as two characters in sequence
 ([VI help][vihelp]).
 
-Installation of AOCC and OpenMPI will take approximately 15 minutes. Once they
-are  installed, you can install additional packages such as `amdblis`:
-
-```shell
-sudo -i spack -d install -v amdblis %aocc@3.2.0
-```
+Installation of AOCC and OpenMPI will take approximately 15 minutes.
 
 Configure SSH user keys for access between cluster nodes:
 

community/examples/AMD/hpc-cluster-amd-slurmv5.yaml

@@ -65,20 +65,46 @@ deployment_groups:
       - type: shell
         source: modules/startup-script/examples/install_ansible.sh
         destination: install_ansible.sh
+      - $(swfs.install_nfs_client_runner)
+      - $(swfs.mount_runner)
       - $(spack.install_spack_deps_runner)
       - $(spack.install_spack_runner)
+      - type: shell
+        content: "shutdown -h +1"
+        destination: shutdown.sh
+
+  - id: slurm_startup
+    source: modules/scripts/startup-script
+    settings:
+      runners:
+      - type: data
+        destination: /etc/profile.d/spack.sh
+        content: |
+          #!/bin/sh
+          if [ -f /sw/spack/share/spack/setup-env.sh ]; then
+                  . /sw/spack/share/spack/setup-env.sh
+          fi
       # the following installation of AOCC may be automated in the future
       # with a clear direction to the user to read the EULA at
       # https://developer.amd.com/aocc-compiler-eula/
       - type: data
         destination: /var/tmp/install_aocc.sh
         content: |
           #!/bin/bash
+          source /sw/spack/share/spack/setup-env.sh
           spack install aocc@3.2.0 +license-agreed
           spack load aocc@3.2.0
           spack compiler find --scope site
           spack -d install -v openmpi@4.1.3 %aocc@3.2.0 +legacylaunchers +pmi schedulers=slurm
 
+  # must restart vm to re-initiate subsequent installs
+  - id: spack_builder
+    source: modules/compute/vm-instance
+    use: [network1, swfs, spack-startup]
+    settings:
+      name_prefix: spack-builder
+      machine_type: c2d-standard-16
+
   - id: low_cost_partition
     source: community/modules/compute/schedmd-slurm-gcp-v5-partition
     use:
@@ -118,6 +144,6 @@ deployment_groups:
     use:
     - network1
     - slurm_controller
-    - spack-startup
+    - slurm_startup
     settings:
       machine_type: c2d-standard-4

community/examples/cloud-batch.yaml

@@ -29,17 +29,14 @@ deployment_groups:
   modules:
   - id: network1
     source: modules/network/pre-existing-vpc
-    kind: terraform
 
   - id: appfs
     source: modules/file-system/filestore
-    kind: terraform
     use: [network1]
     settings: {local_mount: /sw}
 
   - id: hello-startup-script
     source: modules/scripts/startup-script
-    kind: terraform
     settings:
       runners:
       - type: shell
@@ -55,7 +52,6 @@ deployment_groups:
 
   - id: batch-job
     source: community/modules/scheduler/cloud-batch-job
-    kind: terraform
     use: [network1, appfs, hello-startup-script]
     settings:
       runnable: "cat /sw/hello.txt"
@@ -66,6 +62,5 @@ deployment_groups:
 
   - id: batch-login
     source: community/modules/scheduler/cloud-batch-login-node
-    kind: terraform
     use: [batch-job]
     outputs: [instructions]

community/examples/hpc-cluster-small-sharedvpc.yaml

@@ -43,15 +43,13 @@ deployment_groups:
   modules:
   - id: network1
     source: modules/network/pre-existing-vpc
-    kind: terraform
     settings:
       project_id: $(vars.host_project_id)
       network_name: your-shared-network
       subnetwork_name: your-shared-subnetwork
 
   - id: homefs
     source: modules/file-system/filestore
-    kind: terraform
     use: [network1]
     settings:
       local_mount: /home
@@ -61,7 +59,6 @@ deployment_groups:
   # This debug_partition will work out of the box without requesting additional GCP quota.
   - id: debug_partition
     source: community/modules/compute/SchedMD-slurm-on-gcp-partition
-    kind: terraform
     use:
     - network1
     - homefs
@@ -75,7 +72,6 @@ deployment_groups:
   # This compute_partition is far more performant than debug_partition but may require requesting GCP quotas first.
   - id: compute_partition
     source: community/modules/compute/SchedMD-slurm-on-gcp-partition
-    kind: terraform
     use:
     - network1
     - homefs
@@ -85,7 +81,6 @@ deployment_groups:
 
   - id: slurm_controller
     source: community/modules/scheduler/SchedMD-slurm-on-gcp-controller
-    kind: terraform
     use:
     - network1
     - homefs
@@ -97,7 +92,6 @@ deployment_groups:
 
   - id: slurm_login
     source: community/modules/scheduler/SchedMD-slurm-on-gcp-login-node
-    kind: terraform
     use:
     - network1
     - homefs

community/examples/htcondor-pool.yaml

@@ -29,7 +29,6 @@ deployment_groups:
   modules:
   - id: network1
     source: modules/network/vpc
-    kind: terraform
     settings:
       network_name: htcondor-pool
       subnetwork_name: htcondor-pool-usc1
@@ -38,21 +37,17 @@ deployment_groups:
 
   - id: htcondor_install
     source: community/modules/scripts/htcondor-install
-    kind: terraform
 
   - id: htcondor_services
     source: community/modules/project/service-enablement
-    kind: terraform
     use:
     - htcondor_install
 
   - id: htcondor_configure
     source: community/modules/scheduler/htcondor-configure
-    kind: terraform
 
   - id: htcondor_configure_central_manager
     source: modules/scripts/startup-script
-    kind: terraform
     settings:
       runners:
       - type: shell
@@ -63,7 +58,6 @@ deployment_groups:
 
   - id: htcondor_cm
     source: modules/compute/vm-instance
-    kind: terraform
     use:
     - network1
     - htcondor_configure_central_manager
@@ -80,7 +74,6 @@ deployment_groups:
 
   - id: htcondor_configure_execute_point
     source: modules/scripts/startup-script
-    kind: terraform
     settings:
       runners:
       - type: shell
@@ -91,7 +84,6 @@ deployment_groups:
 
   - id: htcondor_execute_point
     source: community/modules/compute/htcondor-execute-point
-    kind: terraform
     use:
     - network1
     - htcondor_configure_execute_point
@@ -106,7 +98,6 @@ deployment_groups:
 
   - id: htcondor_configure_access_point
     source: modules/scripts/startup-script
-    kind: terraform
     settings:
       runners:
       - type: shell
@@ -130,7 +121,6 @@ deployment_groups:
           queue
   - id: htcondor_access
     source: modules/compute/vm-instance
-    kind: terraform
     use:
     - network1
     - htcondor_configure_access_point

community/examples/intel/daos-cluster.yaml

@@ -30,14 +30,12 @@ deployment_groups:
   modules:
   - id: network1
     source: modules/network/pre-existing-vpc
-    kind: terraform
 
   # This module creates a DAOS server. Server images MUST be created before running this.
   # https://github.com/daos-stack/google-cloud-daos/tree/main/images
   # more info: https://github.com/daos-stack/google-cloud-daos/tree/main/terraform/modules/daos_server
   - id: daos-server
     source: github.com/daos-stack/google-cloud-daos.git//terraform/modules/daos_server?ref=v0.2.1
-    kind: terraform
     use: [network1]
     settings:
       number_of_instances: 2
@@ -48,7 +46,6 @@ deployment_groups:
   # more info: https://github.com/daos-stack/google-cloud-daos/tree/main/terraform/modules/daos_client
   - id: daos-client
     source: github.com/daos-stack/google-cloud-daos.git//terraform/modules/daos_client?ref=v0.2.1
-    kind: terraform
     use: [network1, daos-server]
     settings:
       number_of_instances: 2