You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
SAM cannot build Docker images - bundled docker-py API version too old #8466
SAM Version: 1.144.0 Installation: Official AWS installer (/usr/local/aws-sam-cli/) OS: Linux (Ubuntu/Debian) Docker Version: Modern (API 1.44+)
Issue
SAM fails to build Lambda container images with error:
Error: Building image for generateRecruiticsFlags requires Docker. is
Docker running?
Debug output shows:
docker.errors.APIError: 400 Client Error: client version 1.35 is too
old.
Minimum supported API version is 1.44, please upgrade your client to a
newer version
Root Cause
SAM 1.144.0 bundles docker-py that only supports Docker API 1.35 (from
~2018). Modern Docker daemons require API 1.44+.
Expected Behavior
SAM should bundle a modern docker-py version compatible with current
Docker releases.
Workaround
Manually build/push Docker image to ECR, bypass SAM's build step. Or downgrade Docker to 2018 version (unsafe to do this in practice)
Request
Please update the bundled docker-py library to support Docker API
1.44+.
Comments on closed issues are hard for our team to see.
If you need more assistance, please either tag a team member or open a new issue that references this one.
If you wish to keep having a conversation with other community members under this issue feel free to do so.
Activity
valerena commentedon Dec 1, 2025
Hi. Support for newer Docker versions was added in SAM CLI 1.147.1.
We recommend to update to the latest version SAM CLI 1.149.0. You can find the installer directly there on the release page on GitHub, or in the public AWS documentation to install the latest version: https://docs.aws.amazon.com/serverless-application-model/latest/developerguide/install-sam-cli.html
github-actions commentedon Dec 1, 2025
Comments on closed issues are hard for our team to see.
If you need more assistance, please either tag a team member or open a new issue that references this one.
If you wish to keep having a conversation with other community members under this issue feel free to do so.