check-kube-pods-running.rb - undefined method `status' for nil:NilClass

[jkomara]

Testing scenario

A master with a single k8s node that has 3 pods deployed. All pods are in "Running" state and check-kube-pods-running.rb passes.

Delete a pod and then immediately run check-kube-pods-running.rb. Receive the following error:

Check failed to run: undefined method `status' for nil:NilClass, 
["/opt/sensu/embedded/lib/ruby/gems/2.3.0/gems/sensu-plugins-kubernetes-2.0.0/bin/check-
kube-pods-running.rb:95:in `block in run'", 
"/opt/sensu/embedded/lib/ruby/2.3.0/delegate.rb:341:in `each'", 
"/opt/sensu/embedded/lib/ruby/2.3.0/delegate.rb:341:in `block in delegating_block'", 
"/opt/sensu/embedded/lib/ruby/gems/2.3.0/gems/sensu-plugins-kubernetes-2.0.0/bin/check-
kube-pods-running.rb:89:in `run'", "/opt/sensu/embedded/lib/ruby/gems/2.3.0/gems/sensu-
plugin-1.4.2/lib/sensu-plugin/cli.rb:58:in `block in <class:CLI>'"]

Cause

This was because in my local environment I did not have enough resource to schedule the new pod until the old pod was deleted. The check is looking to see if pod.status.conditions[1].stats == 'False'. Since it was waiting to be scheduled pod.status.conditions[1] did not exist and threw this error.

Expected result

A critical that declares the pod is not scheduled because of insufficient resources. I feel like in the real world, if I had a pod that could not be schedule due to lack of resources until the existing pod was terminated, I would want to know about it.

If you need any more information please let me know.

Thank you.

[replicant0wnz]

This issue just popped up for me as well. Not sure what caused it. All pods are running correctly with plenty of resources. I'm disabling the monitor for now.

[majormoses]

fixed by #42

released: https://rubygems.org/gems/sensu-plugins-kubernetes/versions/3.0.1