⇩ Markdown

video-kubernetes-architecture

https://www.youtube.com/watch?v=zeS6OyDoy78

At 8:35 diagram of what runs on master node

At 8:35 master /can/ run containers, but usually doesn't. ^master-containers

At 8:35 fluentd runs on the master node ^fluentd-runs-on-master

At 8:35 supervisord runs docker and kubelet ^supervisord-docker-kubelet

At 8:35 DNS and UI are addons ^addons

At 10:00 the kubelet reports health to the master ^kubelet-reports-health-to-master

At 10:20 supervisord ensures that docker and kublet remain running ^supervisord-ensures-docker-and-kublet-running

At 12:25 the node is also called a minion or a slave (TODO is that only the worker node?) ^minion-slave

At 14:30 pods

At 18:05 - Labels and selectors

At 18:20 "Every pod is completely autonomous. There is no notion of an 'application' that brings multiple pods together" ^pods-are-autonomous

At 22:15 an abstraction to define a logical set of pods, bound by a policy to access them. ^service-abstraction

At 22:15 services are exposed through internal and external endpoints ^service-internal-external-enpoints

At 22:15 can also point to a non-kubernetes endpoint through a Virtual-IP-Bridge ^service-non-kubernetes-endpoints

At 22:15 supports TCP and UDP ^service-tcp-and-udp

Interfaces with kube-proxy to manipulate IP tables ^service-interfaces-with-kube-proxy

At 24:00 - "Services are the endpoints by which one pod can talk to the other" ^services-are-endpoints

At 24:30 - A Cluster IP is available only within the cluster. ^cluster-ip

At 24:40 - A Node Port is an external service. Called node port because every node in the cluster makes the service available on the same port. ^node-port

At 28:24 he has a Docker compose file (TODO is that really needed in Kubernetes?)

docker-compose up -d

kubectl get nodes

kubectl get cs (get component statuses)

^commands

At 30:54 - When you run kubectl get nodes if it says SchedulingDisabled on the master node, it will not run containers ^scheduling-disabled

At 37:50 - The default option is exposing the service port as a cluster IP ^service-port-as-cluster-ip

At 39:40 - there is a default kubernetes service when you do kubectl get svc ^default-kubernetes-service

At 42:50 - kubectl exec -it web /bin/bash gets him into the web container

At 44:20 - from within the web container he does ping redis and it shows that it's at redis.default.svc.cluster.local

At 46:15 - the node ports fall within a specific range ^node-ports-range

At 47:25 - kubectl describe pod web

At 48:10 - If the web pod is running on the third node and we access it at the second node, how is it able to access it? That's where kube-proxy will kick in and bounce the traffic to node 3

At 48:10 - if you access a service at the right port, but wrong node, kube proxy will bounce it to the right node ^kube-proxy-gets-request-to-right-node

At 48:50 - An intranet-only service is called a cluster ip ^cluster-ip-is-intranet-only-service

At 51:50 - If you scale a service by using a replication controller, the service will route to the right nodes using a load balancing scheme ^scale-service

At 52:42 - definition seems a bit like manifest ^definition

At 57:08 - kubectl scale rc web --relicas=50


kubernetes-glossary


tag--video

tag--project-kubernetes

tag--pub-to-kubnotes

Janakiram Msv

tag--kubernetes-architecture