Now checking out an unconference session- "Pod Lifecycle: the good, the bad, and the misunderstood" with @bobbypage!
Pod Lifecycle is complicated and we have the issues to prove it. Many issues are on the kubelet side, but not all of them. With the wide variety of controllers possible, there are a lot of ways for the lifecycle of a pod to go wrong. #kcsna#KubeCon#Kubernetes
Static pods are a particularly weird use case. I remember learning about these while studying for the CKA (Kubernetes Admin certification) & being like, "THAT'S A THING?!" One of those things that feels like an anti-pattern, but essential in certain use cases.
Pod Lifecycle isn't really something I hear talked about clearly often. Documentation is needed to clarify what it is & what the problems are. Greater testing required - new contributors wanted!
The Pod Lifecycle- how your pod/workloads lives & dies in Kubernetes, with the states in between, is foundational knowledge but not currently well-documented as its own subject. But work is in progress. Add it to your Kubernetes Lexicon if you haven't already!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
It's cool to hear the ways folks today can understand Borg, a technology like 15 years old, via papers about it from 6 years ago- by comparing it to their modern experience using its open source descendant, #Kubernetes today.
The goals of borg and the ways they were accomplished are similar enough to k8s to be recognizable, but sometimes accomplished a bit differently. Similar naming conventions (borglets vs kubelets), BCL vs YAML, etc.
A difference is that the separate components of the Kubernetes control plane all seem to be part of a singular component, the Borgmaster. Which means Borgmasters could scale vertically, but not really horizontally - which is allowed more by the k8s component separation.