Merge branch 'test/platform-application-checkpoints' into 'master' (5bcfda5a) · Commits · OOP / deploy / ansible

.gitlab-ci.yml

+214 −2

Changes for .gitlab-ci.yml: 214 added lines, 2 removed lines.
Original line number	Diff line number	Diff line
		workflow:
		rules:
		- if: '$CI_PIPELINE_SOURCE == "merge_request_event"'
		- if: '$CI_PIPELINE_SOURCE == "push" && ($CI_COMMIT_BRANCH == "master" \|\| $CI_COMMIT_BRANCH == "main")'
		- if: '$CI_PIPELINE_SOURCE == "push" && $CI_OPEN_MERGE_REQUESTS'
		when: never
		- when: always

		image: python:3.12-slim

		stages:
		- lint
		- test
		- report

		variables:
		ANSIBLE_CONFIG: ansible/ansible.cfg
		@@ -23,6 +32,10 @@ cache:
		- ansible-galaxy collection install -r ansible/requirements.yml

		secret-scan:
		rules:
		- if: '$RUN_SDK_ACCEPTANCE == "1"'
		when: never
		- when: on_success
		stage: lint
		image: alpine:latest
		before_script:
		@@ -39,18 +52,30 @@ secret-scan:
		fi

		yamllint:
		rules:
		- if: '$RUN_SDK_ACCEPTANCE == "1"'
		when: never
		- when: on_success
		stage: lint
		extends: .ansible_setup
		script:
		- yamllint .

		ansible-lint:
		rules:
		- if: '$RUN_SDK_ACCEPTANCE == "1"'
		when: never
		- when: on_success
		stage: lint
		extends: .ansible_setup
		script:
		- ansible-lint -c .ansible-lint ansible/playbooks ansible/roles ansible/role-template
		- ansible-lint -c .ansible-lint ansible/playbooks ansible/roles ansible/role-template ansible/tests/test_platform_bootstrap.yml

		syntax-check:
		rules:
		- if: '$RUN_SDK_ACCEPTANCE == "1"'
		when: never
		- when: on_success
		stage: test
		extends: .ansible_setup
		script:
		@@ -61,7 +86,194 @@ syntax-check:
		ansible-playbook -i inventory/r2.example.yml --syntax-check "$playbook"
		done

		- ANSIBLE_ROLES_PATH=roles ansible-playbook -i localhost, tests/test_platform_bootstrap.yml

		python-smoke:
		rules:
		- if: '$RUN_SDK_ACCEPTANCE == "1"'
		when: never
		- when: on_success
		stage: test
		script:
		- python3 -m py_compile ansible/roles/oop-platform/files/fm_partners.py ansible/roles/oop-platform/files/tfsdk_bootstrap.py ansible/tests/r2_e2e/.py ci/.py
		- sh -n ci/run_platform_lab.sh
		- python3 -m unittest discover -s ansible/tests -p test_tfsdk_bootstrap.py

		platform-smoke-checks:
		stage: test
		image: python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1
		cache: []
		variables:
		TMPDIR: "$CI_PROJECT_DIR/.cache/smoke-tests"
		PYTHONDONTWRITEBYTECODE: "1"
		script:
		- mkdir -p "$TMPDIR"
		- pip install -r ci/requirements-smoke.txt
		- python -m pytest --confcutdir=ansible/tests/unit -q -p no:cacheprovider ansible/tests/unit --basetemp="$TMPDIR/pytest"
		- python -m pytest --collect-only -q -p no:cacheprovider ansible/tests/r2_e2e

		# Read-only opt-in check of an already deployed platform, through its public URLs.
		platform-smoke:
		stage: test
		image: python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1
		cache: []
		timeout: 10 minutes
		variables:
		TMPDIR: "$CI_PROJECT_DIR/.cache/smoke-install"
		PYTHONDONTWRITEBYTECODE: "1"
		rules:
		- if: '$RUN_PLATFORM_SMOKE == "1" && $CI_COMMIT_REF_PROTECTED == "true"'
		- when: never
		script:
		- mkdir -p "$TMPDIR"
		- pip install -r ci/requirements-smoke.txt
		- python ci/platform_smoke.py --inventory "$OOP_SMOKE_INVENTORY"
		artifacts:
		when: always
		paths:
		- .cache/platform-smoke/result.json
		reports:
		junit: .cache/platform-smoke/junit.xml
		expire_in: 1 week

		# Disposable core platform: create -> existing deploy -> existing health -> delete.
		# Requires a trusted DinD-capable runner with shared TLS client certificates.
		platform-lab:
		stage: test
		image: docker:29.5.1-cli@sha256:b40b3737eb3bf588d25bb856d3564dd3f9fdb32ac2fc19ebe85cc58d761692a5
		services:
		- name: docker:29.5.1-dind@sha256:eed377f3ce14afda560e742c5b3ae446df1ce00a16e1beab2ac6c78c1477fe85
		alias: docker
		variables:
		HEALTHCHECK_TCP_PORT: "2376"
		cache: []
		timeout: 40 minutes
		variables:
		OOP_LAB_REPORT_DIR: .cache/platform-lab/container
		FF_NETWORK_PER_BUILD: "true"
		DOCKER_HOST: tcp://docker:2376
		DOCKER_TLS_CERTDIR: /certs
		DOCKER_TLS_VERIFY: "1"
		DOCKER_CERT_PATH: /certs/client
		rules:
		- if: '$CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH \|\| $CI_COMMIT_BRANCH == "test/platform-application-checkpoints" \|\| $CI_MERGE_REQUEST_SOURCE_BRANCH_NAME == "test/platform-application-checkpoints"'
		when: on_success
		allow_failure: false
		variables:
		OOP_LAB_PROFILE: ci/lab/profile.release-2.0.json
		OOP_LAB_REFRESH_CANDIDATE: "1"
		- if: '$CI_COMMIT_BRANCH == "test/platform-lab"'
		when: manual
		allow_failure: false
		- if: '$RUN_PLATFORM_LAB == "1" && $CI_COMMIT_REF_PROTECTED == "true"'
		- when: never
		before_script:
		- apk add --no-cache python3
		script:
		- sh ci/run_platform_lab.sh "${OOP_LAB_PROFILE:-ci/lab/profile.health.json}" "$OOP_LAB_REPORT_DIR"
		artifacts:
		when: always
		paths:
		- $OOP_LAB_REPORT_DIR/result.json
		- $OOP_LAB_REPORT_DIR/junit.xml
		- $OOP_LAB_REPORT_DIR/chart.json
		- $OOP_LAB_REPORT_DIR/candidate.json
		- $OOP_LAB_REPORT_DIR/feature-matrix.json
		- $OOP_LAB_REPORT_DIR/feature-matrix.md
		- $OOP_LAB_REPORT_DIR/wrapper-junit.xml
		- $OOP_LAB_REPORT_DIR/pods.txt
		- $OOP_LAB_REPORT_DIR/events.txt
		- $OOP_LAB_REPORT_DIR/health/result.json
		- $OOP_LAB_REPORT_DIR/health/junit.xml
		- $OOP_LAB_REPORT_DIR/health/acceptance/
		reports:
		junit:
		- $OOP_LAB_REPORT_DIR/junit.xml
		- $OOP_LAB_REPORT_DIR/wrapper-junit.xml
		expire_in: 90 days

		# Same lab and lifecycle, using a chart archive instead of a container manifest.
		platform-lab-helm:
		extends: platform-lab
		variables:
		OOP_LAB_PACKAGE_TYPE: HELM
		OOP_LAB_REPORT_DIR: .cache/platform-lab/helm
		rules:
		- if: '$CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH \|\| $CI_COMMIT_BRANCH == "test/platform-application-checkpoints" \|\| $CI_MERGE_REQUEST_SOURCE_BRANCH_NAME == "test/platform-application-checkpoints"'
		variables:
		OOP_LAB_PROFILE: ci/lab/profile.release-2.0.json
		OOP_LAB_REFRESH_CANDIDATE: "1"
		- if: '$RUN_PLATFORM_HELM_LAB == "1" && $CI_COMMIT_REF_PROTECTED == "true"'
		- when: never

		# Actual provider compatibility, extending the same container application lab.
		platform-lab-network:
		extends: platform-lab
		variables:
		OOP_LAB_NETWORK_PROVIDER: EURECOM
		OOP_LAB_REPORT_DIR: .cache/platform-lab/network
		rules:
		- if: '$CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH \|\| $CI_COMMIT_BRANCH == "test/platform-application-checkpoints" \|\| $CI_MERGE_REQUEST_SOURCE_BRANCH_NAME == "test/platform-application-checkpoints"'
		variables:
		OOP_LAB_PROFILE: ci/lab/profile.release-2.0.json
		OOP_LAB_REFRESH_CANDIDATE: "1"
		- if: '$RUN_PLATFORM_NETWORK_LAB == "1" && $CI_COMMIT_REF_PROTECTED == "true"'
		- when: never

		# Consolidation consumes real job artifacts; it adds no approval or input form.
		release-feature-report:
		stage: report
		image: python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1
		cache: []
		when: always
		needs:
		- {job: platform-lab, artifacts: true, optional: true}
		- {job: platform-lab-helm, artifacts: true, optional: true}
		- {job: platform-lab-network, artifacts: true, optional: true}
		rules:
		- if: '$CI_COMMIT_BRANCH == $CI_DEFAULT_BRANCH \|\| $CI_COMMIT_BRANCH == "test/platform-application-checkpoints" \|\| $CI_MERGE_REQUEST_SOURCE_BRANCH_NAME == "test/platform-application-checkpoints"'
		- when: never
		script:
		- python ci/merge_feature_reports.py .cache/platform-lab .cache/release-features
		artifacts:
		when: always
		expose_as: Observed release features
		paths:
		- .cache/release-features/feature-matrix.md
		- .cache/release-features/feature-matrix.json
		expire_in: 90 days

		# Explicit live mode leaves all default component tests and deployment jobs alone.
		sdk-live-acceptance:
		stage: test
		image: python:3.12-slim@sha256:a6e34c598f2467ed0e9a8d349809fcd8b5c603269512df273a0bb1784edc11b1
		tags: [sdk-lab]
		interruptible: false
		timeout: 30 minutes
		resource_group: dedicated-sdk-lab
		environment:
		name: sdk-acceptance
		action: verify
		cache: []
		variables:
		TMPDIR: "$CI_PROJECT_DIR/.cache/sdk-tmp"
		PYTHONDONTWRITEBYTECODE: "1"
		rules:
		- if: '$RUN_SDK_ACCEPTANCE == "1" && $CI_COMMIT_REF_PROTECTED == "true"'
		- when: never
		before_script:
		- mkdir -p "$TMPDIR"
		- pip install pytest==8.3.2 requests==2.33.0
		- python ci/install_kubectl.py
		- export PATH="$CI_PROJECT_DIR/.cache/bin:$PATH"
		- kubectl version --client
		script:
		- python3 -m py_compile ansible/roles/oop-platform/files/fm_partners.py ansible/tests/r2_e2e/*.py
		- python ci/sdk_live.py
		artifacts:
		when: always
		paths:
		- .cache/sdk-live/qualification.json
		- .cache/sdk-e2e/*/run.json
		reports:
		junit: .cache/sdk-live/junit.xml
		expire_in: 90 days

README.md

+53 −0

Changes for README.md: 53 added lines, 0 removed lines.
Original line number	Diff line number	Diff line
		@@ -39,3 +39,56 @@ Sergio Giménez <sergio.gimenez@i2cat.net>
		## License

		Apache 2.0. See `LICENSE` and `NOTICE` for attribution requirements.


		## Single SRM / TF-SDK bootstrap

		The existing `oop-platform` role accepts `oop_platform_bootstrap` as one
		operator declaration. It emits the complete bootstrap and matching read-only
		credential mounts through the Release 2 chart:

		```yaml
		oop_platform_bootstrap:
		version: openop.srm.bootstrap/v1
		control_paths:
		- ref: athens
		adapter_type: kubernetes
		kubeconfig_secret:
		name: athens-credentials
		key: config
		infrastructure:
		kube_context: athens
		namespace: applications
		- ref: local-workloads
		adapter_type: kubernetes
		infrastructure:
		in_cluster: true
		namespace: applications
		zones: [] # Add the real declared topology and bindings.
		```

		For in-cluster workloads, configure the existing `workloads.namespaces` RBAC
		setting explicitly. Declaring a control path does not grant permissions.

		Create the referenced Secrets in the SRM namespace through the existing
		operator credential process. The role never reads their values or implements
		a secret resolver. It derives file paths from refs, keeping each control path
		separate, and strips deployment-only Secret metadata from emitted bootstrap.
		The SDK identity is derived by SRM from each ref, not repeated in infrastructure.
		The deployment controller's kubeconfig remains separate from SDK credentials.

		Remove `controlPaths` from extra chart values. When using
		`oop_platform_bootstrap`, also remove manually supplied `topology` and
		`controlPathCredentials`; competing sources fail before deployment. Existing
		empty defaults preserve health-only deployments. The paired chart provides
		`/var/run/openop-tfsdk` for private execution files. Use matching SDK 2.0.0rc3,
		SRM and chart revisions; upgrade or roll back the set together. Bootstrap
		generation is not a backend-health or credential-validity check.

		Offline producer tests run in the existing Python CI job. The real-role check
		also runs locally and in the existing syntax-check job, without cluster access:

		```sh
		python -m unittest discover -s ansible/tests -p test_tfsdk_bootstrap.py
		ANSIBLE_ROLES_PATH=ansible/roles ansible-playbook -i localhost, ansible/tests/test_platform_bootstrap.yml
		```

ansible/inventory/r2-sdk-lab.example.yml

0 → 100644

+62 −0

Changes for ansible/inventory/r2-sdk-lab.example.yml: 62 added lines, 0 removed lines.
Original line number	Diff line number	Diff line
		---
		# Separate, single-operator lab. Copy to r2-sdk-lab.local.yml and fill the pins.
		# No production inventory, federation, shared workload namespace or floating SDK install.
		all:
		children:
		oop_platform_operators:
		hosts:
		sdk_lab:
		ansible_connection: local
		ansible_python_interpreter: "{{ ansible_playbook_python }}"
		oop_platform_kubeconfig: ~/.kube/oop-sdk-lab.yaml
		oop_platform_kube_context: oop-sdk-lab
		oop_platform_namespace: oop-e2e-platform
		oop_platform_public_host: sdk-lab.example.test
		oop_platform_federation_id: sdk-lab
		oop_platform_country_code: GR
		oop_platform_mcc: "202"
		oop_platform_mncs: ["01"]
		oop_platform_federation_clients: []
		oop_platform_portal_enabled: false
		oop_platform_chart_version: 52a7b0d942af88d004a0781e365b0a6db7c66e2c
		oop_platform_fm_image_tag: REPLACE_WITH_REVIEWED_FM_TAG
		oop_platform_oeg_image_tag: REPLACE_WITH_REVIEWED_OEG_TAG
		oop_platform_srm_image_tag: REPLACE_WITH_REVIEWED_SRM_TAG
		e2e_temp_dir: "{{ playbook_dir }}/../../../.cache/sdk-e2e-tmp"
		oop_platform_bootstrap:
		version: openop.srm.bootstrap/v1
		platform_instance_id: sdk-lab
		control_paths:
		- ref: sdk-lab-k8s
		adapter_family: compute
		adapter_type: kubernetes
		adapter_contract_version: v1
		infrastructure:
		namespace: oop-e2e-workloads
		in_cluster: true
		endpoint_host: sdk-lab.example.test
		zones:
		- ref: sdk-lab-zone
		name: SDK lab
		kind: resource
		state: active
		domains:
		- ref: sdk-lab-compute
		name: SDK lab compute
		kind: compute
		state: active
		capabilities:
		- ref: sdk-lab-workload
		name: SDK lab workload
		kind: workload
		state: active
		control_path_bindings:
		- ref: sdk-lab-workload-binding
		control_path_ref: sdk-lab-k8s
		is_default: true
		state: active
		oop_platform_extra_values:
		srm:
		workloads:
		namespaces:
		- oop-e2e-workloads

ansible/inventory/r2-smoke.example.yml

0 → 100644

+13 −0

Changes for ansible/inventory/r2-smoke.example.yml: 13 added lines, 0 removed lines.
Original line number	Diff line number	Diff line
		---
		# HTTP-only view of an already deployed platform. Copy to r2-smoke.local.yml.
		# Public URLs must be reachable from the machine/container running the command.
		# No SSH credentials, kubeconfig or application image is needed for this scenario.
		all:
		children:
		oop_platform_operators:
		hosts:
		local_lab:
		oop_platform_public_host: 127.0.0.1
		oop_platform_scheme: http
		oop_platform_oeg_node_port: 32263
		oop_platform_fm_node_port: 30989

ansible/inventory/r2.example.yml

+9 −14

Changes for ansible/inventory/r2.example.yml: 9 added lines, 14 removed lines.
Original line number	Diff line number	Diff line
		@@ -42,8 +42,8 @@ all:
		# No Release 2 portal image is published yet: build one, or turn the portal off.
		oop_platform_portal_enabled: false
		# The Release 2 charts are on this branch until oop/deploy/helm!15 merges;
		# helm develop still has the Release 1 (MongoDB) charts.
		oop_platform_chart_version: refactor/helm-2.0
		# Pin the reviewed Release 2 chart, including single-bootstrap mounts.
		oop_platform_chart_version: 52a7b0d942af88d004a0781e365b0a6db7c66e2c
		# Plain HTTP between the operators: test networks only.
		oop_platform_allow_insecure_partner_endpoints: true

		@@ -68,32 +68,27 @@ all:
		# deployed by Helm into namespace oop-apps of the operator's own cluster.
		oop_platform_extra_values:
		srm:
		srmcontroller:
		workloads:
		namespaces: [oop-apps]
		controlPaths:
		- control_path_ref: k8s-local
		adapter: kubernetes
		config:
		namespace: oop-apps
		topology: \|
		oop_platform_bootstrap:
		version: openop.srm.bootstrap/v1
		platform_instance_id: {{ oop_platform_federation_id }}
		platform_instance_id: "{{ oop_platform_federation_id }}"
		control_paths:
		- ref: k8s-local
		adapter_family: compute
		adapter_type: kubernetes
		adapter_contract_version: v1
		infrastructure:
		in_cluster: true
		namespace: oop-apps
		zones:
		- ref: {{ oop_platform_federation_id }}-zone
		name: {{ srm_zone_name }}
		- ref: "{{ oop_platform_federation_id }}-zone"
		name: "{{ srm_zone_name }}"
		kind: resource
		state: active
		domains:
		- ref: {{ oop_platform_federation_id }}-compute
		name: {{ srm_zone_name }} compute
		- ref: "{{ oop_platform_federation_id }}-compute"
		name: "{{ srm_zone_name }} compute"
		kind: compute
		state: active
		capabilities: