playbook in error when using openshift lib 0.12

[nergalex]

SUMMARY

Error encounter when using openshift 0.12 package and community.kubernetes.k8s. See attached file: error_with_openshift_0.12.json.txt
community.kubernetes.k8s works well with openshift 0.11.2 pakage. See attached file: ok_with_openshift_0.11.2.json.txt
error_with_openshift_0.12.json.txt
ok_with_openshift_0.11.2.json.txt

ISSUE TYPE

• Bug Report

COMPONENT NAME

community.kubernetes.k8s

ANSIBLE VERSION

Virtual Env with openshift 0.11.2 and 0.12.0 run both:

ansible 2.9.18
  config file = /etc/ansible/ansible.cfg
  configured module search path = [u'/root/.ansible/plugins/modules', u'/usr/share/ansible/plugins/modules']
  ansible python module location = /opt/custom-envs/cloudbuilder-collection2/lib/python2.7/site-packages/ansible
  executable location = /opt/custom-envs/cloudbuilder-collection2/bin/ansible
  python version = 2.7.5 (default, Aug 13 2020, 02:51:10) [GCC 4.8.5 20150623 (Red Hat 4.8.5-39)]

CONFIGURATION

# ansible-config dump --only-changed
/opt/custom-envs/cloudbuilder-collection2/lib/python2.7/site-packages/ansible/parsing/vault/__init__.py:44: CryptographyDeprecationWarning: Python 2 is no longer supported by the Python core team. Support for it is now deprecated in cryptography, and will be removed in the next release.
  from cryptography.exceptions import InvalidSignature

OS / ENVIRONMENT

# cat /etc/redhat-release 
Red Hat Enterprise Linux Server release 7.9 (Maipo)

STEPS TO REPRODUCE

https://github.com/nergalex/f5-aks-nginx_ingress_app_protect/blob/master/playbooks/roles/poc-k8s/tasks/create_nap_log_format.yaml

EXPECTED RESULTS

See attached file: ok_with_openshift_0.11.2.json.txt

ACTUAL RESULTS

See attached file: ok_with_openshift_0.11.2.json.txt

[tima]

@nergalex What version of this collection are you using? Is it the latest stable, v1.2?

[nergalex]

Sorry, please find attached error ouput with community.kubernetes version: 1.2.0 (and not 1.1.1 in the previous message)

"msg": "Failed to load kubeconfig due to Invalid kube-config file. No configuration found."

error_with_openshift_0.12_and_collection_1.2.0.json.txt

[gravesm]

@nergalex I am unable to reproduce this. Specifying the path to the kubeconfig file works for me.

[belenot]

I have this issue too.
It's solved when I patched common.py in /usr/local/lib/python3.8/dist-packages/ansible_collections/community/kubernetes/plugins/module_utils/common.py on line 263.

            #return DynamicClient(kubernetes.client.ApiClient(configuration))
            return DynamicClient(kubernetes.config.new_client_from_config())

I don't know how to watch version, and have not time for this anymore.
I can give more info, if needed and if someone tell me how to obtain it )

[nergalex]

thx @belenot . I would not prefer to modify kubernetes ansible collection because it's loaded for each job by Tower (collections/requirements.yml is used).

@gravesm
using a config.yaml file located in a AWX project, the module invocation differs:

• kubernetes collection 1.2.0 : invocation.module_args.kubeconfig: "config.yaml" --> error_with_openshift_0.12_and_collection_1.2.0.json.txt line 12
• kubernetes collection 1.1.1 : invocation.module_args.kubeconfig: "/tmp/bwrap_24858_brzydgui/awx_24858_bs7aho2c/project/playbooks/roles/poc-k8s/files/config.yaml" --> ok_with_openshift_0.11.2.json.txt line 63

Did you reproduced this behavior?

[belenot]

thx @belenot . I would not prefer to modify kubernetes ansible collection because it's loaded for each job by Tower (collections/requirements.yml is used).

I know @nergalex , that's why I just specified previous version of package for pip3 installation on remote host, namely openshift==0.11.0. This fixed my issue.

[nergalex]

thx @belenot, same workaround :)