Files
krkn/CI/tests_v2/lib/deploy.py
T
cef0d4025d test: migrate test_namespace.sh to CI/tests_v2 namespace_deletion (#1406)
* test: migrate test_namespace.sh to CI/tests_v2 namespace_deletion

Migrate the legacy CI/tests/test_namespace.sh functional test to the pytest-based
v2 framework under CI/tests_v2/scenarios/namespace_deletion/.

Covers the service_disruption (namespace deletion) scenario with happy-path cases
(single-namespace object deletion, multi-namespace delete_count, multiple runs,
wait_time handling, label-selector targeting) and negative cases (no-match regex,
namespace/label mutual exclusion, delete_count exceeding available namespaces).

- Add scenario assets (resource.yaml with Deployment+Service, scenario_base.yaml)
- Register namespace_deletion marker in pytest.ini
- Add namespace_deletion execution-evidence marker in lib/utils.py

Closes #1405

* test: move namespace_deletion helpers into reusable lib modules

Address review feedback: extract the per-test helpers into shared lib
modules so other scenarios can reuse them.

- lib/namespace.py: add POD_SECURITY_PRIVILEGED_LABELS, create_labeled_namespace,
  delete_namespace_quietly, and a make_namespace factory fixture (auto-cleanup)
- lib/deploy.py: add deployment_exists, wait_for_no_deployment,
  wait_for_present_deployment_count, deploy_manifest_to_namespace
- conftest.py: re-export make_namespace fixture
- test_namespace_deletion.py: drop local helpers, use the lib functions/fixture

* test: honor --keep-ns-on-fail in make_namespace factory

The make_namespace finalizer always deleted ad-hoc namespaces, ignoring
--keep-ns-on-fail on failure unlike test_namespace. Extract the keep-on-fail
decision into a shared helper and use it in both fixtures so the documented
debugging workflow works for scenarios that create extra namespaces.

* test: surface FailToCreateError in deploy_manifest_to_namespace

Mirror deploy_workload's error handling so manifest-apply failures in
multi-namespace tests raise a formatted RuntimeError listing the underlying
API exceptions instead of an opaque FailToCreateError.

* test: clarify why label-selector test bypasses run_scenario

The inline comment claimed run_scenario was bypassed because **overrides
would collide with the positional namespace arg. The real reason is that
NAMESPACE_IS_REGEX=True wraps an empty namespace as '^$', whereas
label-selector mode needs a literal empty string.

* test: use actual scenario inputs in negative-test failure context

The no-match and mutual-exclusion tests reported context=namespace=self.ns
(the ephemeral namespace), not the inputs actually under test. Reference the
real namespace (and label_selector) so unexpected-success diagnostics are clear.

* test: use non-default wait_time so override patching is exercised

wait_time=30 matched the scenario_base.yaml default, so the test passed even
if override patching regressed. Use wait_time=5 (non-default) so the override
path is actually validated.

* test: guard cluster post-checks under KRKN_TEST_DRY_RUN

Three namespace_deletion tests asserted cluster side effects (workload
deletion) after the Kraken run. Under KRKN_TEST_DRY_RUN=1 Kraken is
skipped, so the seeded workload is never deleted and wait_for_no_deployment
/ wait_for_present_deployment_count would time out and fail. Guard those
post-checks, and in test_label_selector_targeting (which bypasses
run_scenario and calls run_kraken directly) honor dry-run explicitly by
skipping the invocation and post-check.

* test: tighten no-match assertion, clarify runs-loop test, dry-run-safe negatives

Addresses Deep Code Review feedback on the namespace_deletion suite:

- test_no_match_namespace_fails: drop the dead 'no namespaces matching' OR
  branch; the service_disruption plugin only ever logs 'not enough namespaces
  matching ...', so the assertion now checks that string directly.
- test_multiple_runs_repeat_deletion -> test_multiple_runs_repeat_disruption_loop:
  rename + docstring make explicit that it verifies the outer runs loop iterates
  twice, not that object deletion recurs (Krkn does not redeploy between runs, so
  run 2 re-selects an already-empty namespace). Object removal is asserted in
  test_single_namespace_object_deletion.
- Negative tests now return early under KRKN_TEST_DRY_RUN=1, since run_scenario
  returns a fake rc=0 and the failure path cannot be exercised; this makes the
  whole class consistent under make test-dry-run.

* test: use UUID-based namespace in no-match test to avoid accidental matches

* test: assert correct zero-match error in no-match namespace test

A regex matching zero namespaces makes krkn_lib's check_namespaces raise
'there exists no namespaces matching' before the plugin's delete loop, so
the 'not enough namespaces matching' branch is never reached for this case.
Assert the actual zero-match message instead.

* test: poll for async Service deletion in namespace_deletion test

Kubernetes deletions are asynchronous, so checking the Service immediately
after the scenario run could be flaky. Add wait_for_no_service (mirroring
wait_for_no_deployment) and poll until the Service is actually gone before
asserting.

* test: assert Service deletion in label-selector namespace_deletion test

test_label_selector_targeting deploys both a Deployment and a Service but
only asserted the Deployment was removed. Add wait_for_no_service so the
test fully validates that label-selector mode deletes all objects
(including Services), matching test_single_namespace_object_deletion.

Signed-off-by: augmentcode[bot] <185243770+augmentcode[bot]@users.noreply.github.com>

* test: snapshot namespaces in wait_for_present_deployment_count

list(namespaces) consumed the iterable before the poll loop re-iterated
over the original, so a one-shot iterable (e.g. generator) would be empty
on every poll. Snapshot to a list once and iterate over that snapshot.

Signed-off-by: augmentcode[bot] <185243770+augmentcode[bot]@users.noreply.github.com>

* ci: free up runner disk space before tests_v2 KinD run

The Tests v2 (pytest functional) job intermittently fails with
'System.IO.IOException: No space left on device' on ubuntu-latest runners,
which ship with only ~14GB free. Creating the KinD cluster plus pulling and
kind-loading nginx:alpine and krkn:tools exhausts the disk, failing even the
runner's own diagnostic logging.

Reclaim ~20-30GB by removing the bundled .NET/Android/GHC SDKs and pruning
docker images before the cluster is created.

Signed-off-by: augmentcode[bot] <185243770+augmentcode[bot]@users.noreply.github.com>

---------

Signed-off-by: augmentcode[bot] <185243770+augmentcode[bot]@users.noreply.github.com>
Co-authored-by: augmentcode[bot] <185243770+augmentcode[bot]@users.noreply.github.com>
Co-authored-by: Darshan Jain <darjain@redhat.com>
2026-06-25 18:39:18 +05:30

245 lines
9.1 KiB
Python

"""
Workload deploy and pod/deployment readiness fixtures for CI/tests_v2.
"""
import logging
import time
from pathlib import Path
import pytest
import yaml
from kubernetes import utils as k8s_utils
from kubernetes.client.rest import ApiException
from lib.base import READINESS_TIMEOUT
from lib.utils import patch_namespace_in_docs
logger = logging.getLogger(__name__)
def wait_for_deployment_replicas(k8s_apps, namespace: str, name: str, timeout: int = 120) -> None:
"""
Poll until the deployment has ready_replicas >= spec.replicas.
Raises TimeoutError with diagnostic details on failure.
"""
deadline = time.monotonic() + timeout
last_dep = None
attempts = 0
while time.monotonic() < deadline:
try:
dep = k8s_apps.read_namespaced_deployment(name=name, namespace=namespace)
except Exception as e:
logger.debug("Deployment %s/%s poll attempt %s failed: %s", namespace, name, attempts, e)
time.sleep(2)
attempts += 1
continue
last_dep = dep
ready = dep.status.ready_replicas or 0
desired = dep.spec.replicas or 1
if ready >= desired:
logger.debug("Deployment %s/%s ready (%s/%s)", namespace, name, ready, desired)
return
logger.debug("Deployment %s/%s not ready yet: %s/%s", namespace, name, ready, desired)
time.sleep(2)
attempts += 1
diag = ""
if last_dep is not None and last_dep.status:
diag = f" ready_replicas={last_dep.status.ready_replicas}, desired={last_dep.spec.replicas}"
raise TimeoutError(
f"Deployment {namespace}/{name} did not become ready within {timeout}s.{diag}"
)
def deployment_exists(k8s_apps, namespace: str, name: str) -> bool:
"""Return True if the named deployment exists in the namespace, False on 404."""
try:
k8s_apps.read_namespaced_deployment(name=name, namespace=namespace)
return True
except ApiException as e:
if e.status == 404:
return False
raise
def wait_for_no_deployment(k8s_apps, namespace: str, name: str, timeout: int = 45) -> None:
"""Poll until the named deployment is gone from the namespace; raise if it persists."""
deadline = time.monotonic() + timeout
while time.monotonic() < deadline:
if not deployment_exists(k8s_apps, namespace, name):
return
time.sleep(1)
raise AssertionError(
f"Deployment {name} still present in namespace={namespace} after {timeout}s"
)
def service_exists(k8s_core, namespace: str, name: str) -> bool:
"""Return True if the named service exists in the namespace, False on 404."""
try:
k8s_core.read_namespaced_service(name=name, namespace=namespace)
return True
except ApiException as e:
if e.status == 404:
return False
raise
def wait_for_no_service(k8s_core, namespace: str, name: str, timeout: int = 45) -> None:
"""Poll until the named service is gone from the namespace; raise if it persists."""
deadline = time.monotonic() + timeout
while time.monotonic() < deadline:
if not service_exists(k8s_core, namespace, name):
return
time.sleep(1)
raise AssertionError(
f"Service {name} still present in namespace={namespace} after {timeout}s"
)
def wait_for_present_deployment_count(
k8s_apps, namespaces, name: str, expected: int, timeout: int = 45
) -> list:
"""
Poll until exactly `expected` of the given namespaces still contain the named deployment.
Returns the namespaces where it is still present. Useful for delete_count assertions.
"""
deadline = time.monotonic() + timeout
# Snapshot once so a one-shot iterable (e.g. generator) is not exhausted by the
# first poll, which would make every subsequent iteration see an empty list.
ns_list = list(namespaces)
present = list(ns_list)
while time.monotonic() < deadline:
present = [n for n in ns_list if deployment_exists(k8s_apps, n, name)]
if len(present) == expected:
return present
time.sleep(1)
raise AssertionError(
f"Expected {expected} namespace(s) with {name} present, "
f"found {len(present)}: {present}"
)
def deploy_manifest_to_namespace(
k8s_client,
k8s_apps,
manifest,
namespace: str,
deployment_name: str,
*,
repo_root=None,
timeout: int = READINESS_TIMEOUT,
) -> None:
"""
Apply a manifest file into `namespace` (overriding each doc's namespace) and wait for the
named deployment to become ready. `manifest` may be absolute or relative to repo_root.
Reusable by any scenario that needs to seed a namespace with a target workload.
"""
path = Path(manifest)
if not path.is_absolute() and repo_root is not None:
path = Path(repo_root) / path
docs = patch_namespace_in_docs(list(yaml.safe_load_all(path.read_text())), namespace)
try:
k8s_utils.create_from_yaml(k8s_client, yaml_objects=docs, namespace=namespace)
except k8s_utils.FailToCreateError as e:
msgs = [str(exc) for exc in e.api_exceptions]
raise RuntimeError(
f"Failed to create resources in namespace={namespace}: {'; '.join(msgs)}"
) from e
wait_for_deployment_replicas(k8s_apps, namespace, deployment_name, timeout=timeout)
@pytest.fixture
def wait_for_pods_running(k8s_core):
"""
Poll until all matching pods are Running and all containers ready.
Uses exponential backoff: 1s, 2s, 4s, ... capped at 10s.
Raises TimeoutError with diagnostic details on failure.
"""
def _wait(namespace: str, label_selector: str, timeout: int = READINESS_TIMEOUT):
deadline = time.monotonic() + timeout
interval = 1.0
max_interval = 10.0
last_list = None
while time.monotonic() < deadline:
try:
pod_list = k8s_core.list_namespaced_pod(
namespace=namespace,
label_selector=label_selector,
)
except Exception:
time.sleep(min(interval, max_interval))
interval = min(interval * 2, max_interval)
continue
last_list = pod_list
items = pod_list.items or []
if not items:
time.sleep(min(interval, max_interval))
interval = min(interval * 2, max_interval)
continue
all_running = all(
(p.status and p.status.phase == "Running") for p in items
)
if not all_running:
time.sleep(min(interval, max_interval))
interval = min(interval * 2, max_interval)
continue
all_ready = True
for p in items:
if not p.status or not p.status.container_statuses:
all_ready = False
break
for cs in p.status.container_statuses:
if not getattr(cs, "ready", False):
all_ready = False
break
if all_ready:
return
time.sleep(min(interval, max_interval))
interval = min(interval * 2, max_interval)
diag = ""
if last_list and last_list.items:
p = last_list.items[0]
diag = f" e.g. pod {p.metadata.name}: phase={getattr(p.status, 'phase', None)}"
raise TimeoutError(
f"Pods in {namespace} with label {label_selector} did not become ready within {timeout}s.{diag}"
)
return _wait
@pytest.fixture(scope="function")
def deploy_workload(test_namespace, k8s_client, wait_for_pods_running, repo_root, tmp_path):
"""
Helper that applies a manifest into the test namespace and waits for pods.
Yields a callable: deploy(manifest_path_or_content, label_selector, *, is_path=True)
which applies the manifest, waits for readiness, and returns the namespace name.
"""
def _deploy(manifest_path_or_content, label_selector, *, is_path=True, timeout=READINESS_TIMEOUT):
try:
if is_path:
path = Path(manifest_path_or_content)
if not path.is_absolute():
path = repo_root / path
with open(path) as f:
docs = list(yaml.safe_load_all(f))
else:
docs = list(yaml.safe_load_all(manifest_path_or_content))
docs = patch_namespace_in_docs(docs, test_namespace)
k8s_utils.create_from_yaml(
k8s_client,
yaml_objects=docs,
namespace=test_namespace,
)
except k8s_utils.FailToCreateError as e:
msgs = [str(exc) for exc in e.api_exceptions]
raise RuntimeError(f"Failed to create resources: {'; '.join(msgs)}") from e
logger.info("Workload applied in namespace=%s, waiting for pods with selector=%s", test_namespace, label_selector)
wait_for_pods_running(test_namespace, label_selector, timeout=timeout)
logger.info("Pods ready in namespace=%s", test_namespace)
return test_namespace
return _deploy