Investing in Kubernetes helm charts and the end of support for AWS and Azure templates

[#1]

Hello, Data Center Community!

We’re excited to share our plans for continued investments in deployment on Kubernetes environments per your feedback. This way, you can select any cloud provider or self-hosted platform that fits your instance’s needs best. After releasing the Data Center on Kubernetes feature in 2021, we've seen more and more customers successfully deploy their instances on Kubernetes environments. Furthermore, we believe our Helm charts (which enable you to deploy all Data Center products on cloud providers that support Kubernetes) are an easier, faster, and more flexible way of deployment.

We continue to be hyper-focused on providing you with the features and capabilities you need to modernize your environments. To appropriately and effectively concentrate on further Kubernetes investments, we've decided to end our support and maintenance of the Atlassian Azure and AWS deployment templates dedicated for Jira Software, Jira Service Management, Confluence, Bitbucket, and Crowd. After their end-of-support dates, we’ll no longer update the templates, accept support requests, or collect improvement suggestions.

The planned end-of-support dates are as follows:

• The Azure template will stop receiving support on 31 May 2023.

• The AWS template will stop receiving support on 29 Feb 2024.

What if you’re currently using the templates?

If you’re currently using the Azure and AWS templates, you can still take advantage of our support and maintenance until the end-of-support dates. After those dates, we’ll keep Atlassian templates read-only and open-source if you still want to make a copy, use, maintain, or customize them by yourselves.

What is Atlassian’s recommendation for cloud deployments?

If you haven’t used the Azure and AWS templates, you don’t need to take any actions. If you're thinking about deploying Atlassian Data Center on a cloud provider, our recommended deployment mechanism is a Kubernetes cluster which offers a more efficient and robust Data Center infrastructure and operation setup. To learn more about the benefits of Kubernetes, check out our documentation.

Thank you,

Atlassian Data Center Team

[Steffen Opel _Utoolity_]

Thanks for the heads-up @Mike Ni, and for keeping the OSS template repositories public after end-of-support.

The current Data Center on AWS approach via the Atlassian Standard Infrastructure on AWS and related product quick starts allow to "Seamlessly deploy Data Center in a cluster-ready architecture" via a semantically one click solution, whereas the officially supported Helm charts don't cover the prerequisites (database, filesystem, ...) and deployment lifecycle (infrastructure as code).

Are you going to also officially support the Deployment Automation for Atlassian DC on K8s before the end-of-support for the templates at hand?

[Valentina Mannara]

Thanks for sharing! Regarding the official documentation for Atlassian deployment with Kubernetes https://atlassian.github.io/data-center-helm-charts/userguide/INSTALLATION/ Will it be extended with more details and configuration examples in the next weeks?

Thanks

[Mike Ni]

Thanks for the feedback @Steffen Opel _Utoolity_ .

So just making sure it is clear to other readers, and as you mentioned, Atlassian does officially support our Helm charts. And please read more details about our support boundaries

And Deployment Automation for Atlassian DC on K8s is now only for Atlassian vendors to run the DC App performance toolkit. And the support limits to this single usage scenario and only to DC App vendors. 

In the meantime, we will definitely consider your request to build it and officially support it for broader customer deployment scenarios.  That said if you want to try and get a chance to customise it, let us know your suggestions on how to improve it.

Cheers

Mike Ni

[Mike Ni]

@Valentina Mannara , Thanks for your question.

First, there is some reference for deploying using Helm Charts on different platforms. Check Available examples 

And we will consider extending it with more details and configuration examples in the coming months. Feel free to let us know which kind of detail and platform extension will be helpful to you and your team. 

Cheers

MIke Ni

[Valentina Mannara]

Hi @Mike Ni 

it might be useful to have some info on the cluster, let me explain better for example for deployment on vm and in particular on aws you can find indications on which ec2 size is ideal based on the type of instance, similarly it would be useful to have info on the cluster nodes k8s so that the tools have certain performances. Or the configuration of the pods, what characteristics they must have,

Cheers

Valentina

[Yevhen]

@Valentina Mannara sorry for a very delayed response. You should follow sizing guides for the conventional installations and make K8s requests similar to EC2 instance sizes, i.e. if you had VMs with 4 cores and 16Gb memory, you should aim at having the same pods.