k8s Flashcards
create a pod with the image nginx
create pod with 2 containers, nginx and busybox
create a replicaset
create a deployment
create a namespace and set context to it
create 2 pods, curl one side from the other with dns name
create a clusterip & nodeport for a deployment
manually schedule a pod on a node by using the node’s name
schedule a deploy on a set of nodes based on the node’s tier (labels)
what if you want the deploy to run on nodes that satisfy either tier large or tier medium?
what if it should meet any nodes that has the label ‘supernode’ set?
what if the requirement would kick deployment off the node?
apply & remove a taint to a node
if the node is already running an app without the toleration, make sure its rescheduled on another node after a grace period of 10 seconds
apply tolerations to a deployment
specify resource request and limit for cpu and ram
create a static pod, restart this static pod
get list of pods by label
get list of deployment by label
get list of nodes by label
whether using json or the label selector
taint a node with key=value and the effect of NoScheduled
remove taint from a node
remove toleration from a deployment
set a label on a node
remove a label from a nod
play with nodeAffiniity, podAffinity, podAntiAffinity
set pod nodeaffinity to a node with label color=blue,
requiredDuringSchedulingIgnoredDuringExecution
the nodeaffinity preferredDuringSchedulingIgnoredDuringExecution seems very hard to setup, please try set it again..?
create a pod, set the limit to 1 cpu
0.1% of cpu
1 % of cpu
memory and all that stuff
set the limit 10m memory
find kubelet configurations
create a service account
create a custom scheduler
use the custom scheduler to run a pod