fix(docker): eliminate venv, use system python with uv#209
Merged
fyzanshaik-atlan merged 1 commit intomainfrom Feb 17, 2026
Merged
fix(docker): eliminate venv, use system python with uv#209fyzanshaik-atlan merged 1 commit intomainfrom
fyzanshaik-atlan merged 1 commit intomainfrom
Conversation
Remove the multi-stage builder and venv in favor of installing directly to system python using uv, mirroring how the pyatlan base image itself is built. The pyatlan base image already provides pyatlan, pydantic, httpx, cryptography, authlib, h11, anyio, and other common packages at the system level. Only MCP-specific packages (fastmcp, uvicorn, and their transitive deps) need to be installed on top. This eliminates: - Builder stage (ghcr.io/astral-sh/uv:python3.11-bookworm-slim) - Python venv and the broken symlink workaround (#206) - Bundled pip/setuptools/wheel and their vendored dependencies which carried 2 HIGH vulnerabilities: - jaraco.context 5.3.0 (CVE-2026-23949) - wheel 0.45.1 (CVE-2026-24049) Trivy scan: 0 HIGH/CRITICAL on the resulting image.
abhinavmathur-atlan
approved these changes
Feb 15, 2026
firecast
approved these changes
Feb 17, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Replaces the multi-stage venv-based Dockerfile with a single-stage system install using
uv, mirroring how the pyatlan base image itself is built.Problem
The current Dockerfile:
ghcr.io/astral-sh/uv:python3.11-bookworm-slim) to create a venv and install packages/usr/local/bin/python) doesn't exist in the runtime image (/usr/bin/python3)pip,setuptools, andwheelinside the venv — these carry 2 HIGH vulnerabilities from vendored dependencies:jaraco.context 5.3.0— CVE-2026-23949 (path traversal via malicious tar archives)wheel 0.45.1— CVE-2026-24049 (privilege escalation via malicious wheel files)Solution
The pyatlan base image (
registry.atlan.com/public/pyatlan:main-latest) already provides at the system level:Only MCP-specific packages are missing:
fastmcp,uvicorn, and their transitive deps (mcp,starlette,websockets, etc.).The new Dockerfile:
uv pip install --system--no-depsfor the project itself to avoid re-installing packages already in the base imageBefore → After
Scan results
Test plan
docker buildsucceeds (single stage, ~6s)fastmcp 2.14.5,uvicorn 0.40.0,pyatlan 8.5.2,pydantic 2.12.5,cryptography 46.0.5,starlette 0.52.10.0.0.0:8000