Yandex Cloud
Search
Contact UsGet started
  • Blog
  • Pricing
  • Documentation
  • All Services
  • System Status
    • Featured
    • Infrastructure & Network
    • Data Platform
    • Containers
    • Developer tools
    • Serverless
    • Security
    • Monitoring & Resources
    • ML & AI
    • Business tools
  • All Solutions
    • By industry
    • By use case
    • Economics and Pricing
    • Security
    • Technical Support
    • Customer Stories
    • Gateway to Russia
    • Cloud for Startups
    • Education and Science
  • Blog
  • Pricing
  • Documentation
Yandex project
© 2025 Yandex.Cloud LLC
Yandex Compute Cloud
    • All guides
      • Updating an instance group
      • Updating an instance group based on a YAML specification
      • Moving an instance group to a different availability zone
      • Moving an instance group with a network load balancer to a different availability zone
      • Moving an instance group with an L7 load balancer to a different availability zone
      • Configuring application health checks on a VM instance
        • Rolling updates
        • Zero-downtime updates
      • Pausing an instance group
      • Resuming an instance group
      • Sequentially restarting instances in a group
      • Sequentially recreating instances in a group
      • Stopping an instance group
      • Starting an instance group
      • Configuring instance group access permissions
      • Removing an instance group from a placement group
      • Deleting an instance group
    • Viewing operations with resources
  • Yandex Container Solution
  • Access management
  • Terraform reference
  • Monitoring metrics
  • Audit Trails events
  • Release notes
  1. Step-by-step guides
  2. Managing an instance group
  3. Updating a group
  4. Rolling updates

Rolling updates

Written by
Yandex Cloud
Updated at February 12, 2025

To perform a rolling update of your service, discontinuing no more than N virtual machine instances at a time:

  1. Define the instance template you need.
  2. Set the deployment policy with a non-zero value for max_unavailable, which stands for the maximum number of instances to discontinue at a time.
  3. Start group updating.

For example, to update an instance group sequentially, discontinuing one VM instance at a time, configure deploy-policy as follows:

...
deploy-policy:
    max_unavailable: 1
    max_expansion: 0
    ...
...

The group will be updated as follows:

Rolling update

Was the article helpful?

Previous
Configuring application health checks on a VM instance
Next
Zero-downtime updates
Yandex project
© 2025 Yandex.Cloud LLC