Red Hat OpenShift Managed Services | Red Hat OpenShift, 31 Mar 2021

Previous Meeting

⏯

youtube image

►

From YouTube: Hands-on demo of Red Hat OpenShift Service on AWS

Description

Get a hands-on demonstration of Red Hat OpenShift Service on AWS. We'll cover setting up your first cluster, adding and deleting clusters, and much more.

Learn more at https://www.openshift.com/products/amazon-openshift

A

Hi, my name is rhys oxnam and I manage the field product management team within the cloud platforms business unit here at red hat today, we're going to look at red hat openshift service on aws or rosa for short, and we're going to demonstrate how customers can easily deploy fully managed openshift clusters in the public cloud.

A

Red hat has worked closely with amazon to integrate this service by the amazon web services management console allowing existing amazon customers to onboard openshift capabilities via a simple subscription with all billing centralized just as it is today. Here you can see the landing page for rosa and I've already got.

A

This service enabled there's extensive documentation for the rosa service that walks customers through all of the prerequisites, including requirements for setting up the aws account and how to instantiate clusters after they've subscribed to the service, I'm going to walk you through a deployment end to end today.

A

First things: first, we need to set up an aws account with identity and access management. Permissions. The documentation describes how to create this user and the policies that need to be in place. This is easy to set up either with a standard, aws user or if your user is part of an aws organization.

A

Here, you can see that I've pre-created an iam user called audiorosa and that it has administrator access and hence has full control.

A

Next, we need to configure the aws command line interface with the access key and the secret that reflects this iam user, and this can be done with aws configure here. You can see that I've already set the credentials. You can also specify your default region where I've selected london.

A

Now we can verify that the aws credentials are correctly configured by requesting the caller identity, verifying that it's an iam account and that it's rdo rosa as expected.

A

We can also further verify access to ec2 by requesting a list of instances, here's showing zero. Now that we have aws configured, we need to download the rosa client. This is a dedicated command line utility that controls all aspects of deploying a fully managed openshift cluster on aws fire, this service from the aws management console. We have an option for downloading the cli and there are options for linux, mac, os and windows.

A

I've already downloaded and installed the linux client, let's go ahead and ensure that the rosa client can utilize the aws cli that we just set up and ensure that our account is properly configured so that it has the correct permissions and enough quota to satisfy deployment in our region of choice.

A

The next step is to log into our red hat account via the rosa cli. This ensures that resulting clusters are linked to the openshift cluster manager and associates the cluster to us for any future support requirements. For this we need to grab a token from cloud.redder.com via the link.

A

A

We'll copy this back into rosa and you can see that we're logged in properly a further rosa who am, I will expand on that and will give further clarity on how things are set up with my aws account id the region. I want to deploy into the iam user, we're going to leverage and the link to the red hat account.

A

Now that we've verified that we're ready to start deploying our rosa cluster, the first step is to run rosa init. This will run a short cloudformations template that configures, the aws account ready for the deployment and management of our cluster. This usually takes a minute or two so we'll speed this section up all being well. We're now ready to go ahead and actually deploy the rosa cluster.

A

For this, we simply request the cluster to be created via the rosa cli and we'll give it a name here, I'll use, rdo rosa.

A

This will now start the deployment behind the scenes, and some configuration details will be printed to the console.

A

The rosa cli also allows us to optionally watch the deployment logs where we get the output of the openshift installer. So we can closely monitor progress.

A

The deployment of the cluster takes approximately 40 minutes, so we'll jump forward pretty quickly.

A

A

So there we go install complete. This can be verified by listing the clusters with the rosa cli, and we can ask you to further describe the cluster.

A

Here showing the api url, the console url and the link to the openshift cluster manager page, which we'll dive into first the openshift cluster manager page, gives us deep dive insights into the cluster, including the cluster specifications versions, used number of nodes, networking configuration and much more from here. We can also go ahead and configure access control.

A

Any changes invoked through openshift cluster manager will be enacted on our rosa cluster. So if we want to set up identity management, this can be done here or it can absolutely be done in the rosa cli2.

A

It's possible to set up a wide variety of identity providers from github google to ldap, but in this case to keep things simple for the video we're just going to set up a standard administrator user that exists within the cluster. Only for that we'll need to ask rosa to create this user. For us.

A

Once complete, we'll get the openshift command line parameters that can be used to log directly into the cluster, and we can verify that it's working.

A

We can also verify that we are the cluster admin that we're using the 472 version of openshift and that we have all of our nodes in a ready state.

A

Given that we've now set up a user, let's go ahead and connect the openshift web console, there's a handy link in the openshift cluster manager page and we can grab the password from the previous admin creation command from the rosa cli.

A

As we can see, the openshift cluster is fully functional. Let's take a look at some of the other capabilities of the openshift cluster manager, as it relates to a rosa-based cluster, we're able to view networking configurations.

A

We can then go ahead and see and configure machine pools, and here you can see how easy it is to scale the number of worker nodes from two nodes to three and behind the scenes. The machine set will be scaled and a new aws instance will be provisioned automatically for.

A

A

It's possible to also raise support cases for the specific cluster right from the openshift cluster manager automatically associating it to the cluster in question. It will then take us directly through to the support portal to hugely simplify the creation of a support case.

A

Finally, it's also possible to change the update schedule by default. Cluster updates are manual and left to the cluster operator to invoke, alternatively, you can specify an automatic update schedule where the latest version will be applied at a date and time of your choosing.

A

There's also the ability to enable cluster logging, where logs from the cluster will be automatically shipped to aws, cloudwatch, so they're easily visible from the aws console.

A

This functionality is defined as an add-on and can be enabled by the openshift cluster manager or via the rosa cli here, we'll enable it, via the openshift cluster manager. Turning all options on we'll then watch for the logging pods to be started automatically on our.

A

A

Once they're up and running, we can then go into aws cloudwatch and validate that it's starting to receive log files from our cluster, making this openshift cluster just another part of a customer's infrastructure within.

A

Aws, lastly, we can verify that the rosa cli is also showing the installation of the logging add-on as ready, and that concludes the demonstration of rosa. The only thing left to do now is to delete the cluster which we'll do via the rosa cli.

A

This will then take care of the tear down of the cluster with an aws and the deletion within openshift cluster manager. This step typically takes some time and also allows you to watch the output when available with the watch option.

A

Thank you for your time. I hope that this was useful for you. Please don't hesitate to get in touch if you have any.

A

A