Change API from 'Google API Client' to 'Google Cloud Python Client'

[pablomh]

According to the 'Google API Client' GH page:

This library is considered complete and is in maintenance mode. This means
that we will address critical bugs and security issues but will not add any
new features.

This library is officially supported by Google. However, the maintainers of
this repository recommend using Cloud Client Libraries for Python, where
possible, for new code development.

So change the code accordingly to adapt it to 'Google Cloud Python Client'.

[paigerube14]

Not sure if this is part of what you added but at the end of the scenario i got this error

2024-10-23 17:56:45,901 [INFO] Node with instance ID: prubenda-4185-c8qj7-worker-b-mqtcq has been rebooted
2024-10-23 17:56:45,902 [INFO] node_reboot_scenario has been successfuly injected!
2024-10-23 17:56:46,556 [ERROR] 'namespace'

[pablomh]

Not sure if this is part of what you added but at the end of the scenario i got this error

2024-10-23 17:56:45,901 [INFO] Node with instance ID: prubenda-4185-c8qj7-worker-b-mqtcq has been rebooted
2024-10-23 17:56:45,902 [INFO] node_reboot_scenario has been successfuly injected!
2024-10-23 17:56:46,556 [ERROR] 'namespace'

I took a look and this comes from here:

krkn/krkn/utils/functions.py

Line 74 in 462c9ac

namespaces.extend(krkn_lib.utils.deep_get_attribute("namespace", scenario_config))

There's no namespace in any of the example config files (nor in the one I used and probably the one you used as well), so it complains about it.

[paigerube14]

/lgtm

[paigerube14]

For the power shut down scenario to properly call the stop instances we need to have this funtion return both the instance name and the zone. See the previous return line

Getting error:

2024-11-01 13:53:11,827 [INFO] Error on pool multiprocessing: stop_instances() missing 1 required positional argument: 'instance_id'
2024-11-01 13:53:11,828 [ERROR] ShutDownScenarioPlugin scenario scenarios/openshift/cluster_shut_down_scenario.yml failed with exception: wait_until_stopped() missing 1 required positional argument: 'timeout'
2024-11-01 13:53:12,441 [INFO] wating 60 before running the next scenario

[chaitanyaenr]

@pablomh can you review the ^ comment when you get time please. Thanks.

[paigerube14]

Still hitting this issue:

2024-12-09 11:47:43,348 [INFO] Starting cluster_shut_down scenario injection
2024-12-09 11:47:43,420 [INFO] nodes type <class 'str'>
2024-12-09 11:47:43,420 [INFO] pool type<class 'list'>
2024-12-09 11:47:43,421 [INFO] Error on pool multiprocessing: stop_instances() missing 1 required positional argument: 'instance_id'
2024-12-09 11:47:43,422 [ERROR] ShutDownScenarioPlugin scenario scenarios/openshift/cluster_shut_down_scenario.yml failed with exception: wait_until_stopped() missing 1 required positional argument: 'timeout'

[paigerube14]

@pablomh any updates on this PR? I am hitting an issue with the node scenarios on GCP and wonder if this PR will help fix it/get around it since its using a different library

#740

[pablomh]

I'll be able to keep working on it tomorrow, so keep tuned and let me know if I can help you with your issue.