[BUG]: OpenShift auto-detection crashes on non-OpenShift Kubernetes clusters

[Jatinbhardwaj-093]

Description

In run_kraken.py, OpenShift availability is checked unconditionally at

https://github.com/krkn-chaos/krkn/blob/f7e674d5ad4c86b3d0e2915cfa44a7e73f0ee91f/run_kraken.py#L168-L180

This invokes ocpcli.is_openshift(), which relies on _get_clusterversion_string() implemented in krkn-lib:

krkn-lib/src/krkn_lib/k8s/krkn_kubernetes.py

Lines 188 to 216 in 4bff010

	def _get_clusterversion_string(self) -> str:
	"""
	Return clusterversion status text on OpenShift, empty string
	on other distributions
	*** IT MUST BE CONSIDERED A PRIVATE METHOD (CANNOT USE
	*** DOUBLE UNDERSCORE TO MAINTAIN IT VISIBLE ON SUB-CLASSES)
	*** USED IN KrknKubernetes and KrknOpenshift TO AUTODETECT
	*** THE CLUSTER TYPE
	:return: clusterversion status
	"""
	try:
	cvs = self.custom_object_client.list_cluster_custom_object(
	"config.openshift.io",
	"v1",
	"clusterversions",
	)
	for cv in cvs["items"]:
	for condition in cv["status"]["conditions"]:
	if condition["type"] == "Available":
	return condition["message"].split(" ")[-1]
	return ""
	except client.exceptions.ApiException as e:
	if e.status == 404:
	return ""
	else:
	raise e

The detection logic assumes non-OpenShift clusters fail with an HTTP 404.
However, in practice the call fails with connection-level errors (ConnectionRefusedError, NewConnectionError, MaxRetryError), which are not caught.

As a result:
• is_openshift() throws instead of returning False
• run_kraken exits early with a misconfiguration error
• Kubernetes-only clusters never reach scenario execution

This appears to be the root cause of the observed failures, though further validation may be needed.

LOGS

krkn-ai ➜ uv run krkn_ai run \
    -r krknhub \
    -c ./krkn-ai.yaml \
    -o ./results \
    -p HOST=$HOST
[INFO] Initialized config: ./krkn-ai.yaml
[INFO] Random seed: None (non-reproducible mode)
[INFO] Elasticsearch indexing is disabled
[INFO] Krkn-AI run UUID: a247bd5d-9d0b-4df3-8e30-aab57be5b4d5
[INFO] Saving config file to config.yaml
[INFO] Creating population of size 2
[INFO] | Population |
[INFO] --------------------------------------------------------
[INFO] container-scenarios(robot-shop, service=ratings, 1, .*, 1, 60), 
[INFO] pod-scenarios(robot-shop, service=user, .*, 1, 60, 60), 
[INFO] --------------------------------------------------------
[INFO] | Generation 1 |
[INFO] --------------------------------------------------------
[INFO] Running scenario: container-scenarios(robot-shop, service=ratings, 1, .*, 1, 60)
[WARNING] Could not find 'Chaos data:' in log
[INFO] Krkn scenario return code: 1
[WARNING] Krkn scenario failed with return code 1 (misconfiguration). Skipping fitness calculation to avoid data pollution.
[ERROR] Command output:
WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested
10-02-2026T10:53:48 Checking if OpenShift client is installed
/usr/bin/oc
2026-02-10 10:53:51,281 [INFO] Starting kraken
2026-02-10 10:53:51,292 [INFO] Initializing client to talk to the Kubernetes cluster
2026-02-10 10:53:51,293 [INFO] Generated a uuid for the run: f93d082a-2279-425a-83b4-a71975dcab4d
2026-02-10 10:53:51,330 [WARNING] Retrying (Retry(total=2, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fffebd14f90>: Failed to establish a new connection: [Errno 111] Connection refused')': /apis/config.openshift.io/v1/clusterversions
2026-02-10 10:53:51,331 [WARNING] Retrying (Retry(total=1, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fffeb7cbdd0>: Failed to establish a new connection: [Errno 111] Connection refused')': /apis/config.openshift.io/v1/clusterversions
2026-02-10 10:53:51,331 [WARNING] Retrying (Retry(total=0, connect=None, read=None, redirect=None, status=None)) after connection broken by 'NewConnectionError('<urllib3.connection.HTTPSConnection object at 0x7fffeb7c8f10>: Failed to establish a new connection: [Errno 111] Connection refused')': /apis/config.openshift.io/v1/clusterversions

[paigerube14]

@Jatinbhardwaj-093 thanks so much for this issue! Feel free if you have the capacity to take a shot at fixing this, would be greatly appreciated

[github-actions]

This issue has been automatically marked as stale because it has not had any activity in the last 60 days.
It will be closed in 14 days if no further activity occurs.
If this issue is still relevant, please leave a comment or remove the stale label.
Thank you for your contributions to krkn!