How to Make Identity and Config Operations Boring in Kubernetes – InApps is an article under the topic Devops Many of you are most interested in today !! Today, let’s InApps.net learn How to Make Identity and Config Operations Boring in Kubernetes – InApps in today’s post !
Read more about How to Make Identity and Config Operations Boring in Kubernetes – InApps at Wikipedia
You can find content about How to Make Identity and Config Operations Boring in Kubernetes – InApps from the Wikipedia website
Cloud Native Computing Foundation and VMware sponsored this post, in anticipation of KubeCon + CloudNativeCon North America 2020 – Virtual, Nov. 17-20.
Almost any software engineer will tell you: if they have to do the same thing more than once, they look for a way to automate it. It’s no different when managing or deploying to Kubernetes. No one wants to have to login to a system more than once a day. No one wants to reinvent the wheel when it comes to a configuration template. No one wants to manually update image references in a config file, or make sure images are built in the correct order of operations. And no one wants to manually track the changes they’ve made to their config files.
The good news is that there are tools across a couple of different projects that help with this.
Nikhita is a Senior Member of Technical Staff at VMware, a member of the Kubernetes Steering Committee, and is responsible for the overall governance of the Kubernetes project. She is also a Cloud Native Computing Foundation (CNCF) Ambassador and an international public speaker.
Let’s say you want to deploy an app to a Kubernetes cluster at work. To deploy the app, you first need access to the cluster. Since the cluster is owned by your employer, it is likely that accessing the cluster requires you to enter your credentials, and possibly perform multifactor authentication. However, you don’t want to enter your credentials each time you run
kubectl get pods. That would be very cumbersome! So how do you avoid doing that?
Pinniped, an open source project from VMware, makes this easy by allowing cluster administrators to plug in external identity providers (IDPs) into Kubernetes clusters. It provides a simplified login flow across all clusters. Once you have Pinniped installed on your clusters, the first time that you run a kubectl command it will prompt you to click on a URL. That URL in your browser will redirect you to interactively log in to your upstream IDP and complete authentication. Now, if you run another kubectl command, it will “just work” until your session expires. Even if you access multiple clusters across multiple regions, providers or environments, you only need to login once.
Pinniped makes it simple to consume identities from an external IDP and use those identities across many clusters. Administrators can configure external IDPs via Kubernetes custom resources. This means admins can manage Pinniped using GitOps and standard Kubernetes tools. Currently, the only IDPs supported are webhooks that implement the Kubernetes TokenReview API. Support for more IDP types is coming soon and the complete roadmap for the project is public.
Okay, now you can access your Kubernetes cluster. But you still need to deploy your app. Carvel is a set of open source tools — ytt (YAML Templating Tool), kbld and kapp — that make this process effortless. The beauty of Carvel lies in the fact that each tool has a single purpose and a clear boundary. You can choose which ones to use; and there are no dependencies between the tools. So let’s see how Carvel helps in easing the deployment workflow.
As a first step, you have a config.yaml configuration file that contains Deployment and Service configurations. To ensure that your app is extensible and can take multiple data values, let’s say you decide to customize the configs. Here’s where ytt proves useful. Ytt allows templating through annotated YAML files. Annotations are made using comments against YAML structures — like maps, arrays, etc. If you want to customize imperatively using conditionals and loops, ytt allows you to specify this in a python-like language called Starlark. You can also validate the input data by setting up assertions in Starlark. Since ytt uses structure-aware templating instead of text templating, you no longer need to scratch your head about challenges related to indentation or escaping.
While working on your app, let’s say you come across a Kubernetes Deployment configuration on the internet that you’d like to use. To make sure that the config works seamlessly with your app, you probably need to tweak a few knobs (e.g. replicas) in this third-party config file. But, you don’t want to lose track of this underlying third-party configuration. Ytt’s powerful overlay features make this very easy, by patching the config file with arbitrary changes. Ytt allows the copy of the third-party config file to remain pristine and unmodified. If you aren’t sure of the changes you are making (templating and patching are hard!), you can also try them out in an interactive playground.
If there are any images specified in your config file, they probably need to be built and published. You can use kbld to achieve this. kbld looks for images within your config file, builds the images via Docker (you can also plug in other builders) and pushes it to the registry of your choice. It even updates image references in the config file, resolving those references to an immutable digest. To make debugging easier in the future, it also annotates the Kubernetes resources with image metadata.
Now that your config file is ready, you need to deploy your app. To deploy your app, you can use
kubectl apply directly — but kubectl doesn’t tell you what changes would be applied to the resources in your cluster. To get more transparency into the deploy process, you can use kapp. Kapp is a lightweight CLI tool that calculates changes between your configuration and the live cluster state; and only applies the changes you approve. Kapp is also dependency-aware. If you messed up the order of resources in your config file, kapp orders them correctly (e.g. CRDs before custom resources that need them). Kapp also tracks resources based on a uniquely generated label, so that you can delete all resources in one go with a
kapp delete command.
Find out More and Join Us!
If you want to learn more about Pinniped and Carvel (ytt, kbld and kapp), we’d love to see you at VMware’s booth at the virtual KubeCon+CloudNativeCon North America 2020 conference.
You can follow the work we do and be part of ongoing discussions by participating in Pinniped and Carvel’s GitHub issues and discussions pages. To contribute, please check out the good first issues. If you or your team would like a demo, create an issue or connect with us in #pinniped and #carvel on the Kubernetes Slack. We are very friendly and would love to hear from you!
To learn more about Kubernetes and other cloud native technologies, consider coming to KubeCon + CloudNativeCon North America 2020, Nov. 17-20, virtually.
The Cloud Native Computing Foundation is a sponsor of InApps.
Feature image via Pixabay.
InApps is a wholly owned subsidiary of Insight Partners, an investor in the following companies mentioned in this article: Docker.
List of Keywords users find our article on Google:
|saas steering wheel|
|motivational speaker website template|
|spec home builder boring|
|table extensible made|
|vmware identity manager|
|carvel focus lab menu|
|idp vietnam ho chi minh|
|vmware kubernetes cluster|
|kubernetes virtual clusters|
|virtual cluster kubernetes|
Let’s create the next big thing together!
Coming together is a beginning. Keeping together is progress. Working together is success.