rename arcus environment to .stackhpc

Author: sjpb

.github/workflows/fatimage.yml

@@ -4,32 +4,26 @@ on:
   workflow_dispatch:
 jobs:
   openstack:
-    name: openstack-ci-${{ matrix.cloud }}
-    strategy:
-      matrix:
-        cloud:
-          - "arcus"   # Arcus OpenStack in rcp-cloud-portal-demo project, with RoCE
-      fail-fast: false # as want clouds to continue independently
+    name: openstack-build-arcus
     concurrency: ${{ github.ref }} # to branch/PR
     runs-on: ubuntu-20.04
     env:
       ANSIBLE_FORCE_COLOR: True
       OS_CLOUD: openstack
-      TF_VAR_cluster_name: ci${{ github.run_id }}
     steps:
       - uses: actions/checkout@v2
 
       - name: Setup ssh
         run: |
           set -x
           mkdir ~/.ssh
-          echo "${${{ matrix.cloud }}_SSH_KEY}" > ~/.ssh/id_rsa
+          echo "${arcus_SSH_KEY}" > ~/.ssh/id_rsa
           chmod 0600 ~/.ssh/id_rsa
         env:
           arcus_SSH_KEY: ${{ secrets.ARCUS_SSH_KEY }}
 
       - name: Add bastion's ssh key to known_hosts
-        run: cat environments/${{ matrix.cloud }}/bastion_fingerprint >> ~/.ssh/known_hosts
+        run: cat environments/.stackhpc/bastion_fingerprint >> ~/.ssh/known_hosts
         shell: bash
 
       - name: Install ansible etc
@@ -38,21 +32,21 @@ jobs:
       - name: Write clouds.yaml
         run: |
           mkdir -p ~/.config/openstack/
-          echo "${${{ matrix.cloud }}_CLOUDS_YAML}" > ~/.config/openstack/clouds.yaml
+          echo "${arcus_CLOUDS_YAML}" > ~/.config/openstack/clouds.yaml
         shell: bash
         env:
           arcus_CLOUDS_YAML: ${{ secrets.ARCUS_CLOUDS_YAML }}
 
       - name: Setup environment
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
         
       - name: Build fat image with packer
         id: packer_build
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           cd packer/
           packer init
           PACKER_LOG=1 packer build -only openstack.openhpc -on-error=ask -var-file=$PKR_VAR_environment_root/builder.pkrvars.hcl openstack.pkr.hcl
@@ -68,5 +62,3 @@ jobs:
           IMAGE_NAME=$(openstack image show -f value -c name $IMAGE_ID)
           echo "::set-output name=IMAGE_ID::$IMAGE_ID"
           echo "::set-output name=IMAGE_NAME::$IMAGE_NAME"
-        env:
-          OS_CLOUD: openstack

.github/workflows/stackhpc.yml

@@ -1,5 +1,5 @@
 
-name: Test deployment and image build on OpenStack
+name: Test deployment and reimage on OpenStack
 on:
   workflow_dispatch:
   push:
@@ -8,12 +8,7 @@ on:
   pull_request:
 jobs:
   openstack:
-    name: openstack-ci-${{ matrix.cloud }}
-    strategy:
-      matrix:
-        cloud:
-          - "arcus"   # Arcus OpenStack in rcp-cloud-portal-demo project, with RoCE
-      fail-fast: false # as want clouds to continue independently
+    name: openstack-ci-arcus # Arcus OpenStack in rcp-cloud-portal-demo project, with RoCE
     concurrency: ${{ github.ref }} # to branch/PR
     runs-on: ubuntu-20.04
     env:
@@ -27,13 +22,13 @@ jobs:
         run: |
           set -x
           mkdir ~/.ssh
-          echo "${${{ matrix.cloud }}_SSH_KEY}" > ~/.ssh/id_rsa
+          echo "${arcus_SSH_KEY}" > ~/.ssh/id_rsa
           chmod 0600 ~/.ssh/id_rsa
         env:
           arcus_SSH_KEY: ${{ secrets.ARCUS_SSH_KEY }}
 
       - name: Add bastion's ssh key to known_hosts
-        run: cat environments/${{ matrix.cloud }}/bastion_fingerprint >> ~/.ssh/known_hosts
+        run: cat environments/.stackhpc/bastion_fingerprint >> ~/.ssh/known_hosts
         shell: bash
 
       - name: Install ansible etc
@@ -44,20 +39,20 @@ jobs:
 
       - name: Initialise terraform
         run: terraform init
-        working-directory: ${{ github.workspace }}/environments/${{ matrix.cloud }}/terraform
+        working-directory: ${{ github.workspace }}/environments/.stackhpc/terraform
 
       - name: Write clouds.yaml
         run: |
           mkdir -p ~/.config/openstack/
-          echo "${${{ matrix.cloud }}_CLOUDS_YAML}" > ~/.config/openstack/clouds.yaml
+          echo "${arcus_CLOUDS_YAML}" > ~/.config/openstack/clouds.yaml
         shell: bash
         env:
           arcus_CLOUDS_YAML: ${{ secrets.ARCUS_CLOUDS_YAML }}
 
       - name: Setup environment-specific inventory/terraform inputs
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           ansible-playbook ansible/adhoc/generate-passwords.yml
           echo vault_testuser_password: "$TESTUSER_PASSWORD" > $APPLIANCES_ENVIRONMENT_ROOT/inventory/group_vars/all/test_user.yml
         env:
@@ -67,15 +62,15 @@ jobs:
         id: provision_servers
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           cd $APPLIANCES_ENVIRONMENT_ROOT/terraform
           terraform apply -auto-approve
 
       - name: Get server provisioning failure messages
         id: provision_failure
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           cd $APPLIANCES_ENVIRONMENT_ROOT/terraform
           TF_FAIL_MSGS="$(../../skeleton/\{\{cookiecutter.environment\}\}/terraform/getfaults.py  $PWD)"
           echo TF failure messages: $TF_FAIL_MSGS
@@ -85,29 +80,29 @@ jobs:
       - name: Delete infrastructure if failed due to lack of hosts
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           cd $APPLIANCES_ENVIRONMENT_ROOT/terraform
           terraform destroy -auto-approve
         if: ${{ always() && steps.provision_servers.outcome == 'failure' && contains(steps.provision_failure.messages, 'not enough hosts available') }}
 
       - name: Configure cluster
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           ansible all -m wait_for_connection
           ansible-playbook -v ansible/site.yml
           ansible-playbook -v ansible/ci/check_slurm.yml
 
       - name: Run MPI-based tests
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           ansible-playbook -vv ansible/adhoc/hpctests.yml
 
       - name: Confirm Open Ondemand is up (via SOCKS proxy)
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           
           # load ansible variables into shell:
           ansible-playbook ansible/ci/output_vars.yml \
@@ -139,7 +134,7 @@ jobs:
       #   id: packer_build
       #   run: |
       #     . venv/bin/activate
-      #     . environments/${{ matrix.cloud }}/activate
+      #     . environments/.stackhpc/activate
       #     cd packer/
       #     packer init
       #     PACKER_LOG=1 packer build -except openstack.fatimage -on-error=ask -var-file=$PKR_VAR_environment_root/builder.pkrvars.hcl openstack.pkr.hcl
@@ -148,15 +143,15 @@ jobs:
       # - name: Test reimage of compute nodes to new environment-specific image (via slurm)
       #   run: |
       #     . venv/bin/activate
-      #     . environments/${{ matrix.cloud }}/activate
+      #     . environments/.stackhpc/activate
       #     ansible login -v -a "sudo scontrol reboot ASAP nextstate=RESUME reason='rebuild image:${{ steps.packer_build.outputs.NEW_COMPUTE_IMAGE_ID }}' ${TF_VAR_cluster_name}-compute-[0-3]"
       #     ansible compute -m wait_for_connection -a 'delay=60 timeout=600' # delay allows node to go down
       #     ansible-playbook -v ansible/ci/check_slurm.yml
 
       - name: Test reimage of all nodes (via rebuild adhoc)
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           ansible-playbook -v --limit control,login ansible/adhoc/rebuild.yml
           ansible all -m wait_for_connection -a 'delay=60 timeout=600' # delay allows node to go down
           ansible-playbook -v ansible/site.yml
@@ -165,25 +160,25 @@ jobs:
       - name: Check sacct state survived reimage
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           ansible-playbook -vv ansible/ci/check_sacct_hpctests.yml
 
       - name: Check MPI-based tests are shown in Grafana
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           ansible-playbook -vv ansible/ci/check_grafana.yml
 
       - name: Delete infrastructure
         run: |
           . venv/bin/activate
-          . environments/${{ matrix.cloud }}/activate
+          . environments/.stackhpc/activate
           cd $APPLIANCES_ENVIRONMENT_ROOT/terraform
           terraform destroy -auto-approve
         if: ${{ success() || cancelled() }}
 
       # - name: Delete images
       #   run: |
       #     . venv/bin/activate
-      #     . environments/${{ matrix.cloud }}/activate
+      #     . environments/.stackhpc/activate
       #     ansible-playbook -vv ansible/ci/delete_images.yml