Yandex Cloud
Search
Contact UsTry it for free
  • Customer Stories
  • Documentation
  • Blog
  • All Services
  • System Status
  • Marketplace
    • Featured
    • Infrastructure & Network
    • Data Platform
    • AI for business
    • Security
    • DevOps tools
    • Serverless
    • Monitoring & Resources
  • All Solutions
    • By industry
    • By use case
    • Economics and Pricing
    • Security
    • Technical Support
    • Start testing with double trial credits
    • Cloud credits to scale your IT product
    • Gateway to Russia
    • Cloud for Startups
    • Center for Technologies and Society
    • Yandex Cloud Partner program
    • Price calculator
    • Pricing plans
  • Customer Stories
  • Documentation
  • Blog
© 2026 Direct Cursus Technology L.L.C.
Yandex Managed Service for Apache Airflow™
  • Getting started
    • Resource relationships
    • Networking in Managed Service for Apache Airflow™
    • Available Apache Airflow™ versions
    • Apache Airflow™ versioning policy
    • Quotas and limits
    • Impersonation
    • Host classes
    • Maintenance
  • Access management
  • Pricing policy
  • Terraform reference
  • Yandex Monitoring metrics
  • Audit Trails events
  • Release notes
  • FAQ

In this article:

  • Version lifecycle stages
  • Versions 2.X
  • Versions 3.X
  • Version support schedule
  • Versions 2.X
  • Versions 3.X
  • Update policy
  • Specifics of switching from versions 2.X to 3.X
  1. Concepts
  2. Apache Airflow™ versioning policy

Apache Airflow™ versioning policy

Written by
Yandex Cloud
Updated at February 4, 2026
  • Version lifecycle stages
    • Versions 2.X
    • Versions 3.X
  • Version support schedule
    • Versions 2.X
    • Versions 3.X
  • Update policy
    • Specifics of switching from versions 2.X to 3.X

Each version in the Yandex Cloud infrastructure has lifecycle stages assigned depending on its kernel release date. If the Apache Airflow™ version used in a Managed Service for Apache Airflow™ cluster becomes deprecated, it gets updated.

Upgrading to current versions allows you to get new features, patches, and security fixes, which improves the overall stability of the cluster.

Warning

In Managed Service for Apache Airflow™,the lifecycle rules for major Apache Airflow™ versions (2.x and 3.x) are different. This is due to major differences in the version support periods by the service kernel developers.

Managed Service for Apache Airflow™ uses two Apache Airflow™ version support strategies:

  • LTS (Long-Term Support) versions.
  • Versions with regular updates.

Version lifecycle stagesVersion lifecycle stages

Versions 2.XVersions 2.X

For versions 2.X, Apache Airflow™ adopts the LTS support strategy, as per its official version support policy:

  • Current versions: 2.8, 2.10.
  • Target version: 2.11 (LTS).
  • Python version: 3.10, 3.12.

Update procedureUpdate procedure

All clusters versions 2.8 and 2.10 will be updated to version 2.11. The forced update will not begin until February 2026.

Version 2.11 will be supported until the end of 2026.

Versions 3.XVersions 3.X

Versions 3.X follow the regular updates strategy. Kernel developers release new feature versions (3.X) approximately once every six months. Yandex Cloud makes efforts to make new versions promptly available for its customers.

Python version: currently, versions 3.X only support Python 3.12.

Note

More Python versions may be supported in future. Python 3.12, which is its current primary version, will remain in support until the end of 2028, as per the Python kernel support policy.

Key stages based on version availability in Yandex Cloud:

Version status Description Applies in (from the kernel release date)
Supported versions
Availability in Managed Service for Apache Airflow™ You can use this version to create and manage new clusters. 0 to 1 months
Unavailable for new clusters You cannot use this version to create new clusters, but the existing ones will continue to operate. You may be prompted to upgrade in the management console. 6 months
Unsupported versions
Forced update Existing clusters are focibly updated to the supported version. 12 months

Tip

We recommend updating clusters manually to the supported versions without waiting for an automatic update. This enables you to test compatibility of DAG files in advance.

Version support scheduleVersion support schedule

Versions 2.XVersions 2.X

Available Python versions: 3.10, 3.12.

Version Current status Forced update
2.8 Unavailable for creating new clusters February 2026
2.10 Available February 2026
2.11 (LTS) Scheduled (February 2026) Q4 2026

All versions 2.8 and 2.10 will be updated to the official open-source community-supported LTS version 2.11 no earlier than in February 2026.

Versions 3.XVersions 3.X

Available Python versions: 3.12 (more versions may be added).

Version Available in Managed Service for Apache Airflow™ Unavailable for new clusters Forced update
3.0 September 2025 March 2026 September 2026
3.1 November 2025 June 2026 Q4 2026
3.X TBD TBD + 6 months TBD + 12 months

Update policyUpdate policy

Version updates within the same branch (e.g., from 3.0 to 3.1) take place as follows:

  1. As soon as the version switches to Forced update, cluster update is initiated.
  2. The cluster is updated to the nearest supported version within the major branch.

Update notifications

The notification system provides the following notifications:

  • 30 days prior to a version’s transition to the Forced update status.
  • 7 days prior to the start of a version’s forced update.

Specifics of switching from versions 2.X to 3.XSpecifics of switching from versions 2.X to 3.X

Due to architectural differences, automatic migration from versions 2.X to 3.X is not supported.

You need to upgrade to version 3.X manually. Proceed as follows:

  1. Save DAG files in an external storage (S3 or git repository).
  2. Create a new Managed Service for Apache Airflow™ version 3.X cluster.
  3. Connect the repository with the DAG files to the cluster you created.

See alsoSee also

Versioning in Managed Service for Apache Airflow™

Was the article helpful?

Previous
Available Apache Airflow™ versions
Next
Quotas and limits
© 2026 Direct Cursus Technology L.L.C.