chore: sync downstream changes to midstream main

Signed-off-by: Nathan Weinberg <nweinber@redhat.com>

Author: nathan-weinberg

.github/workflows/redhat-distro-container-build.yml

@@ -51,7 +51,7 @@ jobs:
         uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # v6.18.0
         with:
           context: .
-          file: redhat-distribution/Containerfile
+          file: distribution/Containerfile
           platforms: ${{ matrix.platform }}
           push: ${{ github.event_name == 'push' }}
           tags: ${{ env.IMAGE_NAME }}:${{ github.sha }},${{ env.IMAGE_NAME }}:latest

.pre-commit-config.yaml

@@ -0,0 +1,49 @@
+exclude: 'build/'
+
+default_language_version:
+    python: python3.12
+
+repos:
+-   repo: https://github.com/pre-commit/pre-commit-hooks
+    rev: v5.0.0  # Latest stable version
+    hooks:
+    -   id: check-merge-conflict
+        args: ['--assume-in-merge']
+    -   id: trailing-whitespace
+        exclude: '\.py$'  # Exclude Python files as Ruff already handles them
+    -   id: check-added-large-files
+        args: ['--maxkb=1000']
+    -   id: end-of-file-fixer
+        exclude: '^(.*\.svg)$'
+    -   id: no-commit-to-branch
+    -   id: check-yaml
+        args: ["--unsafe"]
+    -   id: detect-private-key
+    -   id: requirements-txt-fixer
+    -   id: mixed-line-ending
+        args: [--fix=lf] # Forces to replace line ending by LF (line feed)
+    -   id: check-executables-have-shebangs
+    -   id: check-json
+    -   id: check-shebang-scripts-are-executable
+    -   id: check-symlinks
+    -   id: check-toml
+
+-   repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.9.4
+    hooks:
+    -   id: ruff
+        args: [ --fix ]
+    -   id: ruff-format
+
+-   repo: local
+    hooks:
+      - id: pkg-gen
+        name: Distribution Build
+        entry: ./distribution/build.py
+        language: python
+        pass_filenames: false
+        require_serial: true
+        always_run: true
+        files: ^distribution/.*$
+        additional_dependencies:
+          - llama-stack==0.2.18

README.md

@@ -15,7 +15,7 @@ The Containerfile is auto-generated from a template. To generate it:
 1. Make sure you have the `llama` CLI tool installed
 2. Run the build script from root of this git repo:
    ```bash
-   ./redhat-distribution/build.py
+   ./distribution/build.py
    ```
 
 This will:
@@ -25,7 +25,7 @@ This will:
 
 ## Editing the Containerfile
 
-The Containerfile is auto-generated from a template. To edit it, you can modify the template in `redhat-distribution/Containerfile.in` and run the build script again.
+The Containerfile is auto-generated from a template. To edit it, you can modify the template in `distribution/Containerfile.in` and run the build script again.
 NEVER edit the generated `Containerfile` manually.
 
 ## Building the Container Image
@@ -35,7 +35,7 @@ Once the Containerfile is generated, you can build the image using either Podman
 ### Using Podman build image for x86_64
 
 ```bash
-podman build --platform linux/amd64 -f redhat-distribution/Containerfile -t rh .
+podman build --platform linux/amd64 -f distribution/Containerfile -t rh .
 ```
 
 ## Notes

redhat-distribution/Containerfile distribution/Containerfileredhat-distribution/Containerfile renamed to distribution/Containerfile

@@ -1,5 +1,5 @@
 # WARNING: This file is auto-generated. Do not modify it manually.
-# Generated by: redhat-distribution/build.py
+# Generated by: distribution/build.py
 
 FROM registry.access.redhat.com/ubi9/python-312@sha256:95ec8d3ee9f875da011639213fd254256c29bc58861ac0b11f290a291fa04435
 WORKDIR /opt/app-root
@@ -17,7 +17,7 @@ RUN pip install \
     llama_stack_provider_lmeval==0.1.7 \
     llama_stack_provider_trustyai_fms==0.1.2 \
     matplotlib \
-    mcp \
+    mcp>=1.8.1 \
     nltk \
     numpy \
     openai \
@@ -40,9 +40,9 @@ RUN pip install \
     uvicorn
 RUN pip install --index-url https://download.pytorch.org/whl/cpu torch torchvision
 RUN pip install --no-deps sentence-transformers
-RUN pip install --no-cache llama-stack==0.2.14
+RUN pip install --no-cache llama-stack==0.2.18
 RUN mkdir -p ${HOME}/.llama/providers.d ${HOME}/.cache
-COPY redhat-distribution/run.yaml ${APP_ROOT}/run.yaml
-COPY redhat-distribution/providers.d/ ${HOME}/.llama/providers.d/
+COPY distribution/run.yaml ${APP_ROOT}/run.yaml
+COPY distribution/providers.d/ ${HOME}/.llama/providers.d/
 
-ENTRYPOINT ["python", "-m", "llama_stack.distribution.server.server", "--config", "/opt/app-root/run.yaml"]
+ENTRYPOINT ["python", "-m", "llama_stack.core.server.server", "/opt/app-root/run.yaml"]

redhat-distribution/Containerfile.in distribution/Containerfile.inredhat-distribution/Containerfile.in renamed to distribution/Containerfile.in

@@ -3,9 +3,9 @@ WORKDIR /opt/app-root
 
 RUN pip install sqlalchemy # somehow sqlalchemy[asyncio] is not sufficient
 {dependencies}
-RUN pip install --no-cache llama-stack==0.2.14
+RUN pip install --no-cache llama-stack==0.2.18
 RUN mkdir -p ${{HOME}}/.llama/providers.d ${{HOME}}/.cache
-COPY redhat-distribution/run.yaml ${{APP_ROOT}}/run.yaml
-COPY redhat-distribution/providers.d/ ${{HOME}}/.llama/providers.d/
+COPY distribution/run.yaml ${{APP_ROOT}}/run.yaml
+COPY distribution/providers.d/ ${{HOME}}/.llama/providers.d/
 
-ENTRYPOINT ["python", "-m", "llama_stack.distribution.server.server", "--config", "/opt/app-root/run.yaml"]
+ENTRYPOINT ["python", "-m", "llama_stack.core.server.server", "/opt/app-root/run.yaml"]

redhat-distribution/build.py distribution/build.pyredhat-distribution/build.py renamed to distribution/build.py

@@ -5,15 +5,15 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-# Usage: ./redhat-distribution/build.py
+# Usage: ./distribution/build.py
 
 import shutil
 import subprocess
 import sys
 from pathlib import Path
 
 BASE_REQUIREMENTS = [
-    "llama-stack==0.2.14",
+    "llama-stack==0.2.18",
 ]
 
 
@@ -59,9 +59,11 @@ def check_llama_stack_version():
 
 def get_dependencies():
     """Execute the llama stack build command and capture dependencies."""
-    cmd = "llama stack build --config redhat-distribution/build.yaml --print-deps-only"
+    cmd = "llama stack build --config distribution/build.yaml --print-deps-only"
     try:
-        result = subprocess.run(cmd, shell=True, capture_output=True, text=True, check=True)
+        result = subprocess.run(
+            cmd, shell=True, capture_output=True, text=True, check=True
+        )
         # Categorize and sort different types of pip install commands
         standard_deps = []
         torch_deps = []
@@ -74,7 +76,9 @@ def get_dependencies():
                 parts = line.replace("uv ", "RUN ", 1).split(" ", 3)
                 if len(parts) >= 4:  # We have packages to sort
                     cmd_parts = parts[:3]  # "RUN pip install"
-                    packages = sorted(set(parts[3].split()))  # Sort the package names and remove duplicates
+                    packages = sorted(
+                        set(parts[3].split())
+                    )  # Sort the package names and remove duplicates
 
                     # Determine command type and format accordingly
                     if "--index-url" in line:
@@ -104,13 +108,14 @@ def get_dependencies():
     except subprocess.CalledProcessError as e:
         print(f"Error executing command: {e}")
         print(f"Command output: {e.output}")
+        print(f"Command stderr: {e.stderr}")
         sys.exit(1)
 
 
 def generate_containerfile(dependencies):
     """Generate Containerfile from template with dependencies."""
-    template_path = Path("redhat-distribution/Containerfile.in")
-    output_path = Path("redhat-distribution/Containerfile")
+    template_path = Path("distribution/Containerfile.in")
+    output_path = Path("distribution/Containerfile")
 
     if not template_path.exists():
         print(f"Error: Template file {template_path} not found")
@@ -121,10 +126,12 @@ def generate_containerfile(dependencies):
         template_content = f.read()
 
     # Add warning message at the top
-    warning = "# WARNING: This file is auto-generated. Do not modify it manually.\n# Generated by: redhat-distribution/build.py\n\n"
+    warning = "# WARNING: This file is auto-generated. Do not modify it manually.\n# Generated by: distribution/build.py\n\n"
 
     # Process template using string formatting
-    containerfile_content = warning + template_content.format(dependencies=dependencies.rstrip())
+    containerfile_content = warning + template_content.format(
+        dependencies=dependencies.rstrip()
+    )
 
     # Write output
     with open(output_path, "w") as f:

distribution/build.yaml

@@ -0,0 +1,36 @@
+version: 2
+distribution_spec:
+  description: Red Hat distribution of Llama Stack
+  providers:
+    inference:
+    - provider_type: remote::vllm
+    - provider_type: inline::sentence-transformers
+    vector_io:
+    - provider_type: inline::milvus
+    safety:
+    - provider_type: remote::trustyai_fms
+    agents:
+    - provider_type: inline::meta-reference
+    eval:
+    - provider_type: remote::trustyai_lmeval
+    datasetio:
+    - provider_type: remote::huggingface
+    - provider_type: inline::localfs
+    scoring:
+    - provider_type: inline::basic
+    - provider_type: inline::llm-as-judge
+    - provider_type: inline::braintrust
+    telemetry:
+    - provider_type: inline::meta-reference
+    tool_runtime:
+    - provider_type: remote::brave-search
+    - provider_type: remote::tavily-search
+    - provider_type: inline::rag-runtime
+    - provider_type: remote::model-context-protocol
+  container_image: registry.redhat.io/ubi9/python-311:9.6-1749631027
+additional_pip_packages:
+- aiosqlite
+- sqlalchemy[asyncio]
+image_type: container
+image_name: llama-stack-rh
+external_providers_dir: distribution/providers.d