k8s lingo

k8s basic objects

1. pods:  encapsulates an app’s container(s), storage, IP, and options
2. service:  abstract way to expose an app running on a set of Pods as a network service (could be a deployment  or stuff)  as part of  the 'spec' there  is a 'selector' which targets pods.
1. clusterIP:  only available intra cluster
2. nodePort:  Exposes the Service on each Node’s IP at a static port
   (implies clusterIP too)
3. loadBalancer:  Exposes the Service externally using a cloud provider’s load balancer.
   (implies nodePort+clusterIP too)
4. externalName:  Maps the Service to the contents of the externalName.   (returns the CNAME instead of a the  clusterIP of the service.)
3. volume:    offers some persistence  across restarts and sharing across containers in a pod. lots  of volume type choices.
4. namespace

k8s higher-level  objects

1. deployment:  kinda covers  pods+replicasets.   like 'what to run and  how many'
2. daemonSet:  ensures that all (or some) Nodes run a copy of a Pod.  think logging/metrics  daemons per pod.
3. statefulSet:  for managing  stateful apps.  provides guarantees about the ordering and uniqueness of these Pods.  sticky identifier.
4. replicaSet:  maintain a stable set of replica Pods running at any given time.  you may never need this....its kinda baked into a Deployment.
5. job:  creates one or more Pods and ensures that a specified number of them successfully terminate.
6. cronJob:  creates Jobs on a time-based schedule.

k8s Control Plane =  k8s Master + kubelet processes

Node -  is a worker machine. NOT inherently created by k8s!
Services on a node include [container runtime, kubelet, kube-proxy]


kubectl get nodes
kubectl describe node minikube
(Addresses (internal/external  ip),  Conditions(Resource Pressures), Capacity+Available,  Info(versions) )


k8s  master ie  API  server  runs  a bunch  of controllers ex:  [nodeController, deploymentController, jobController, etc, etc]