[07:09:18] <akosiaris>	 Found it
[07:09:39] <akosiaris>	 So, the reason that conflict was added is because kubernetes had cadvisor integrated 
[07:10:00] <akosiaris>	 you could pass --cadvisor-port and it would start a standard cadvisor web ui
[07:10:16] <akosiaris>	 per https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.12.md 
[07:10:21] <akosiaris>	 The formerly publicly-available cAdvisor web UI that the kubelet started using --cadvisor-port has been entirely removed in 1.12. The recommended way to run cAdvisor if you still need it, is via a DaemonSet. (#65707, @dims)
[07:17:04] <akosiaris>	 elukey: So, I replied, but I have a bigger question. Why do we want to run cadvisor on the k8s nodes? We are duplicating effort if we do, all the metrics are already exposed by the kubelet and scraped by prometheus 
[08:25:22] <elukey>	 akosiaris: kalimera, thanks :) So I don't have a lot of context besides https://phabricator.wikimedia.org/T108027, I think that we should probably follow up with Observability and/or ask the question in the task
[12:29:54] <akosiaris>	 cool, will do
[16:08:26] <jayme>	 I think I found all references to old hiera structure for kubernetes clusters in wikitech but in case I missed something: It's all in https://gerrit.wikimedia.org/g/operations/puppet/+/refs/heads/production/hieradata/common/kubernetes.yaml now, please find parameter docs in https://gerrit.wikimedia.org/g/operations/puppet/+/refs/heads/production/modules/k8s/types/clusterconfig.pp