New
The Gartner Hype Cycle Emerging Technologies Report is out!

Download now

Home
/
Guides
/
Sunsetting Coder v1 and v2 Migration FAQ

Sunsetting Coder v1 and v2 Migration FAQ

Sunsetting Coder v1 and v2 Migration FAQ

Coder v2 is Coder's 2nd generation remote development platform launched in June 2022. This document lists frequently asked questions for customers planning to migrate from Coder v1 to v2.

Coder v2 Dashboard

When will Coder no longer support v1?

Coder v1 will have 3 sunset or end-of-life dates.

DateSupport Sunset ParametersExample
06/30/2023End of feature requests and enhancementsimage tag decommissioning coming in 1.38.0
12/31/2023End of feature-related bug fixesorganization sort order in 1.37.0
03/31/2024End of security vulnerability fixesmalicious DevURL redirect link fix in 1.37.1
How will we continue to get v1 support?

Continue to either coordinate with your Coder account executive or leverage the Slack channel.

Why did Coder build v2?

Coder v2 addresses compute and integration limitations of Coder v1. In particular, 100% control over a workspace's Kubernetes pod spec, flexibility to make workspace compute a Kubernetes pod, a VM, or a Docker container, and an open-source platform for the community to get the developer-centric benefits of remote development without a license fee. See the blog post Lessons learned from v1.

What Coder v2 features are open-source and paid?

Functionality for a developer to be productive are in the v2 OSS, while scalability, governance and control features for DevOps teams are in the v2 Enterprise paid version. See this page for feature comparisons.

Will I pay the same for Coder v2?

Contact Sales to learn how your v1 license fee converts to v2.

Is v2 a different code base?

Yes. Because we wanted to make v2 open-source and use Terraform as the workspace provisioning engine, it was easier to re-build Coder's remote development platform into a new code base. v2 also has a different Postgres database schema.

How are the v2 concepts different than v1 at a high-level?

In v1, workspaces are Kubernetes pods with an inner container based on container images with an optional configure script in the image that runs additional configurations as the non-root user after the workspace is built. Coder v1 has an optional workspace template yaml spec to define compute and additional bash scripting steps.

In Coder v2, workspaces are defined as Terraform templates with Terraform resources to specify the infrastructure provider and compute type. e.g., Kubernetes pod, Docker container, or VM. Docker or alternatively VM images are specified in the template. The template includes an agent resource and startup_script configuration that can run the configure script in the image or additional steps like in v1 workspace templates. Coder v1 workspace applications are configured as coder_app resources in the Terraform template.

v1 organizations are groups in v2.

Integration points remain the same like OIDC for single-sign-on and specifying image registries in v2 templates.

Out-of-the-box Git authentication in v1 is an OAuth app and SSH. In v2, OAuth is used as well but Coder intercepts git actions, forcing the user to authenticate to their git provider. Coder stores the user's OAuth token in the Coder database and using it for subsequent git actions. In v2, Coder issues an SSH key to each user if that is preferred.

Are there migration scripts from v1 to v2?

No. The database schema and architectural concepts are so different in v2, it is not reasonable to build migration scripts that meet all customer deployment scenarios. See the migration strategy and recommendations on moving workspace contents.

Is there a community to support v2?

Yes. In v1, Coder maintains a Slack channel. In v2, customers can file GitHub Issues or use our Discord or Slack channels to ask questions to the community.


For more information on Coder v2 concepts, please review the Moving to Coder v2 guide.

See an opportunity to improve our docs? Make an edit.