Overview

Cloud computing offers significant benefits from an agility perspective. By leveraging the cloud, businesses can rapidly provision and scale resources based on their changing needs, enabling quick responses to market demands and opportunities. The cloud’s self-service capabilities empower teams to easily access and deploy applications and services, reducing time-to-market and enabling faster innovation cycles.

But what if AWS suddenly decides to deprecate either a Service or a feature?

Historically, AWS didn’t have a large track record of deprecating services. However, in the last six months, AWS has announced more than 10 deprecations across multiple services. This seems to be aligned with the new company’s focus on optimization and cost reduction. We believe that the pace of service deprecation will increase even more in the next 6-12 months.

 

The  AWS Customer Agreement highlights in section 1.5 that services can be discontinued or changed:

Notice of Changes to the Services. We may change or discontinue any of the Services from time to time. We will provide you at least 12 months’ prior notice before discontinuing a material functionality of a Service that you are using, or materially altering a customer-facing API that you are using in a backwards-incompatible fashion, except that this notice will not be required if the 12 month notice period (a) would pose a security or intellectual property issue to us or the Services, (b) is economically or technically burdensome, or (c) would cause us to violate legal requirements.

 

Deprecation Categories

There are mainly two deprecation categories as follows:

Planned End of Life

For some services, the end-of-life of software versions is communicated upfront in the AWS documentation. This applies to:

  • Software provided by AWS such as Amazon Linux 2. If we take a recent deprecation, the End of Life was scheduled for Jun-2023 but got extended until Jun-2025 as mentioned in the Amazon Linux 2 FAQ section
  • Third-party commercial software such as Databases, has a direct impact on Amazon’s Relational Database Service (RDS). If they take an RDS Oracle Engine for example, customers can plan in advance, as Oracle publishes publicly their End of Life and Release Schedule. AWS aligns with the Oracle schedule and highlights the RDS Upgrade path, which means that customers will have no choice but upgrade the database engine to a newer version at some stage. An upgrade to a major version might have a big impact. The customer will have to do a full regression test of their application, and potentially change the application code and in some cases for legacy applications it might not even be possible. Some alternatives in this case are provided by AWS to run the database either in EC2 or the RDS Custom for Oracle. These also come with additional migration efforts.
  • Open Source Software such as Python and NodeJS, have their own release and deprecation schedule. In these two cases, AWS has a specific Lambda Runtime Deprecation Policy as an example which determines what is the impact if not upgraded.

Ad-hoc Service Deprecation

The ad-hoc service deprecation is the most concerning from a customer perspective. It’s at AWS’s discretion to decide when they will deprecate a service, without the customer’s consultation. Customers have typically 12 months to migrate their solution. Migration effort can vary from weeks to months of work depending on the service and how much these services are embedded into the customer’s solution.

Deprecated Services and Features – A few examples

We have consolidated below a few deprecation notices which AWS has announced in the last few months. This list is not complete. These are the services that we are aware have been deprecated. Unfortunately, AWS does not make it easy to find this information in a centralized place. If you are aware of other services or features being deprecated, please let us know.

AWS OpsWorks Stacks – May 2024

    • You are receiving this notification because you have one or more AWS OpsWorks Stacks resources. We are contacting you to inform you that we will discontinue AWS OpsWorks Stacks on May 26, 2024. Starting May 26, 2023, new customers will be unable to onboard to the service. Existing accounts can continue to use the service as normal until May 26, 2024. From May 26, 2024, onwards, customers will be unable to use the OpsWorks Console, API, CLI, and CloudFormation resources. We recommend customers prepare for this by transitioning their applications running on AWS OpsWorks Stacks to AWS Systems Manager as soon as possible.

AWS OpsWorks for Puppet Enterprise – Mar 2024

    • AWS OpsWorks for Puppet Enterprise is not accepting new customers. Existing customers will be unaffected until March 31, 2024, at which time the service will become unavailable. We recommend that existing customers migrate to other solutions as soon as possible.

AWS DeepLens – Jan 2024

    • Starting January 31, 2024, you can no longer access AWS DeepLens through the AWS management console, manage AWS DeepLens devices, or access any projects you have created.

AWS API Endpoints Deprecate TLS 1.0 and 1.1 – Jun 2023

    • After June 28, 2023, AWS will update our API endpoint configuration to remove TLS 1.0 and TLS 1.1, even if you still have connections using these versions.

Amazon Linux – Jun 2023

    • The Amazon Linux AMI ended its standard support on December 31, 2020, and has entered a maintenance support phase. 

AWS CodePipeline Polling for pipelines not ran in the last 30 days – May 2023

    • You are receiving this message because you have one or more polling pipelines in your account. A polling pipeline is defined as a pipeline that has at least one source action configured to poll for changes. The AWS CodePipeline team will be disabling polling in inactive pipelines effective May 25, 2023. An inactive pipeline is defined as one that has not had a pipeline execution started in the last 30 days. Inactive pipelines that only use Amazon EventBridge rule, or AWS CodeStar Connections, or webhooks to trigger the pipeline will not be affected. Additionally, any active pipeline will also not be affected.

AWS Data Pipeline Console Access – Apr 2023

    • Console access to the AWS Data Pipeline service will be removed on April 30, 2023. On this date, you will no longer be able to access AWS Data Pipeline through the console. You will continue to have access to AWS Data Pipeline through the command line interface and API. Please note that the AWS Data Pipeline service is in maintenance mode and we are not planning to expand the service to new regions.

AWS CodeBuild update on AL2x86_64 Standard 3.0 Deprecation – March 2023

    • AWS CodeBuild notified customers on March 2, 2023, that the Amazon Linux 2 standard 3.0 image [1] will be deprecated on March 30, 2023. We have since revised the decision and extended the deprecation date for the image to April 30, 2023. You may continue using the image for your builds after April 30, 2023, but you could see increased build latency. However, this image will not get any further updates. We recommend you update your Build Projects to use the latest build images in order to get the latest language runtimes and tools. For a list of Docker images provided by CodeBuild, please refer to the CodeBuild documentation [2].

Amazon Sumerian – Feb 2023

    • The Amazon Sumerian service is no longer accepting new customers. Existing customer scenes will not be available after February 21, 2023. We, therefore, recommend that you move any existing scenes to the new Babylon.js-AWS experience and publish your web app using a service like AWS Amplify Hosting.

Amazon EC2 Auto Scaling will no longer add support for new EC2 features to Launch Configurations – End of 2022

    • You can continue using launch configurations, and AWS is committed to supporting applications you have already built using them, but in order for you to take advantage of our most recent and upcoming releases, a migration to launch templates is recommended. Additionally, we plan to no longer support new instance types with launch configurations by the end of 2022.

AWS IoT GreenGrass V1 – TBD

    • The AWS IoT Greengrass V1 endpoints and API operations will remain available until further notice. After the maintenance phase ends on June 30, 2023, the AWS IoT Greengrass Core software v1.11.6 will enter the extended life phase until further notice. During this phase, devices that run the AWS IoT Greengrass Core software v1.x can continue to connect to the AWS IoT Greengrass V1 service to operate until further notice.If AWS IoT Greengrass V1 stops being supported in the future, AWS IoT Greengrass will provide 12 months advance notice before this happens. This will help you plan the update of your applications to use AWS IoT Greengrass V2 and the AWS IoT Greengrass Core software v2.x. For more information about how to update your applications to V2

EC2 Classic – 2022

    • At the end of October 2021, Amazon disabled EC2-Classic on any account that had no active EC2-Classic resources and stopped selling one-year and three-year reserved instances for EC2-Classic. AWS ex all migrations to be complete by August 2022, with no remaining EC2-Classic resources present in any deployment and mandatory VPC for all instances.

Impact on Customers

As discussed at the beginning of this post, there are many befits of leveraging Cloud Computing and in particular AWS.

However, we’ve seen the deprecation of services impacting some of our customers. Quite often customers are busy working on strategic projects and are under a lot of pressure to deliver. Then suddenly they receive an email from AWS with a deprecation notice. At first, because these teams are working hard to deliver business value they defer doing a deeper impact analysis to a later stage. More often than not they start the impact analysis a few months before the deprecation, which leads to distraction, increase in stress, and increased risk to the business.

The way forward

AWS automatically sends email notifications regarding deprecated services or features, making it vital to keep your email address updated in your AWS Account. Ensure these notifications don’t end up in spam and that your Cloud team receives and takes action on them promptly. To maintain visibility even during team member absences, consider using an email distribution list, enabling comprehensive awareness and collaboration.

Whether you are a significant strategic AWS customer or have AWS Enterprise Support, it is highly likely that the AWS Account team will proactively reach out to ensure your awareness of upcoming deprecations, especially if you have deployed those services within your AWS accounts.

The next step then is to identify the resources impacted by the deprecation. Some important aspects to take into consideration:

  • Identify how many resources are impacted. This will help you to estimate effort.
  • Make sure that you inspect multiple AWS accounts and regions
  • Are the impacted AWS resources being accessed by internal or external teams?
  • Consider the migration path options. Usually AWS suggests a migration path based on your use case. Whenever possible, AWS also provide scripts to help you identify impacted resources.
  • Make sure that you validate and test migration in a dev/test environment and automate the deployment using IaC (e.g: Terraform, CloudFormation, CDK)

 

 AWS_Advanced_tier_servicesIf your team is busy or requires assistance from AWS experts to assess the impact of a service deprecation on your workloads, we’re here to support you. We can collaborate with your teams, providing guidance on architecture, service design, and assisting with implementation and migration efforts. Don’t hesitate to reach out to us for the help you need in navigating this transition smoothly.

Francioni is an accredited provider of cloud professional services and solutions including DevOps, Cloud Native Development, Cloud Security, Data Migration and Analytics, and Digital transformation. We are an Advanced AWS Partner and can help you in your journey.