Kubernetes - What are the common failures in a real world k8s cluster?

[Accela Zhao]

Hi,

Since kubernetes are growing increasingly popular and more and more people are run it in production or evaluating, I'm curious what are the common failures / errors that one meets in real world k8s. I found this doc mentioned some (failure modes), https://github.com/kubernetes/kubernetes/blob/37f0368ba26f5f503df2407f3241a3ad62cc8e59/docs/availability.md. But it would be definitely good if there were more details and real cases. Thanks a lot if anyone would like to share some experience!

Best,

Accela

[Accela Zhao]

Found these two may be of some help

    Google ClusterData 2011:

        https://github.com/google/cluster-data/blob/master/ClusterData2011_2.md

    The Computer Failure Data Repository (CFDR) 

        https://www.usenix.org/cfdr