Intro to Kubernetes and Rancher Online Training: January 19, 2019

preview_player
Показать описание
Ready to learn about using Kubernetes clusters and workloads?

This free online training is aimed at users new to Kubernetes, or those interested in using Rancher to easily deploy and manage Kubernetes clusters in production or in your home lab.

These sessions demo Rancher, a free, open-source platform for deploying and managing Kubernetes clusters across the cloud or on-prem.

Sign up for upcoming sessions at:

In this session, our engineer walks you through:

-Essential Kubernetes concepts: Pods, Deployments, Services, Config Maps, Ingresses
-The key components of a Kubernetes deployment
-Deploying and scaling a containerized application with Kubernetes, Helm, and Rancher

All demonstrations were done live and in real-time. This demo was done on Rancher.

To learn more about Kubernetes and Rancher:

Рекомендации по теме
Комментарии
Автор

This is a stunningly good video. Thank you for providing it for learning about k8s and rancher. I am evaluating rancher for use and this was absolutely invaluable.

JimBasilio
Автор

Great presentation! The specific examples given at 14:30 really hit home.

Draaack
Автор

Hi,
I like the video. Lots of information.
I know I am getting dumber with age so this is probably something that everyone else understands but not me (I've researched and could not find a clear shot answer to this question hence, I think it's just me not understanding).

I've installed a single node Rancher on a VM provisioned with Docker. Gave it ip 192.168.1.10.
I've created 6 VMs with Docker. Gave them the following IPs. 192.168.1.11 ... 192.168.1.16.
From the Rancher interface, I've created the cluster and ran the commands on my 6 machines. 3 Masters (controlplane + etcd) and 3 Minions (just workers).
All is good... the cluster is up and running.... stunningly easy.

I then deploy a workload from Rancher's UI. Simple one... just an NGINX hello world page. I set it to run 1 instance on each worker. So far so good.. all instances running.

Now, here's my confusion(s).

I create an ingress to point to that workload, I give it a hostname "test.mycrazydomain.com". I create a DNS entry to resolve that hostname to the Rancher IP (192.168.1.10). I create a dynamic NodePort and expose port 80 on the ingress.
If I access any minion IP with the nodeport, nginx greets me with a beautiful page.
However, if I try to access test.mycrazydomain.com, I get the Rancher login screen.

My understanding was that the ingress exposes itself at the rancher IP and that the hostname is used to redirect to the workload.
Even, when I look at the conf of the nginx container in rancher, I can see that the hostname has been used to create a new virtual server.


What did I not understand?

Thanks!

maxbonfield
Автор

Discovered this gem very late. Do you have any new upcoming training that I can signup for? The one I signed up for last week got cancelled I guess :-(

kumardeepam
Автор

can you share what is the auto history complete tool you are using?
its look like fish

hagaluly
Автор

That was awesome. Everything resonated so well and really appreciate the amount of time that has gone into this...Keep it up guys

odyseykg
Автор

Context appreciated. Would watch longer sessions.

marketreactor
Автор

Thanks for posting.
Great value, helped me a lot.

LGGlReplay
Автор

17:38 what do you mean by cloning it? I think of it as more of referencing it. So if a pod needs an SQL pod, it's simply referencing the SQL pod, not cloning another instance of that SQL pod.

Darkyahweh
Автор

Hi, can we conver VM into a docker container, that VM has a legacy application running in it.

pranavkoritala
Автор

Hello Sir,


Thank you for nice explanation. How can i deploy ELK stack using rancher 2 ?


Any help would be appreciated! Thank you in advance.

mohamedbenighil
Автор

Hello,


Thank you for nice explanation !




Any help would be really appriciated !

mohamedbenighil
Автор

Has anyone sued Rancher for a larger project and had any success?

King_
Автор

Man, thanks for this, but I'm really disappointed that you've skipped lots of important info! Just make a longer training, like 4 hours or something, with a pause like for long movies at the cinema!

mysticaltech
Автор

does anyone want to buy a stake? not 100% but pefer about 1%, could be less. pls add him on skype to discuss

sf