[CICD] Github workflow for publishing Docker images (#764)

# What does this PR do?

- Add Github workflow for publishing docker images. 
- Manual Inputs
- We can use a (1) TestPyPi version / (2) build via released PyPi
version

**Notes**
- Keep this workflow manually triggered as we don't want to publish
nightly docker images

**Additional Changes**
- Resolve issue with running llama stack build in non-terminal device
```
  File "/home/runner/.local/lib/python3.12/site-packages/llama_stack/distribution/utils/exec.py", line 25, in run_with_pty
    old_settings = termios.tcgetattr(sys.stdin)
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
termios.error: (25, 'Inappropriate ioctl for device')
```
- Modified build_container.sh to work in non-terminal environment


## Test Plan

- Triggered workflow:
3562217878
<img width="1076" alt="image"
src="https://github.com/user-attachments/assets/f1b5cef6-05ab-49c7-b405-53abc9264734"
/>


- Tested published docker image
<img width="702" alt="image"
src="https://github.com/user-attachments/assets/e7135189-65c8-45d8-86f9-9f3be70e380b"
/>


- /tools API endpoints are served so that docker is correctly using the
TestPyPi package
<img width="296" alt="image"
src="https://github.com/user-attachments/assets/bbcaa7fe-c0a4-4d22-b600-90e3c254bbfd"
/>

- Published tagged images:
https://hub.docker.com/repositories/llamastack
<img width="947" alt="image"
src="https://github.com/user-attachments/assets/2a0a0494-4d45-4643-bc29-72154ecc54a5"
/>


## Sources

Please link relevant resources if necessary.


## Before submitting

- [ ] This PR fixes a typo or improves the docs (you can dismiss the
other checks if that's the case).
- [ ] Ran pre-commit to handle lint / formatting issues.
- [ ] Read the [contributor
guideline](https://github.com/meta-llama/llama-stack/blob/main/CONTRIBUTING.md),
      Pull Request section?
- [ ] Updated relevant documentation.
- [ ] Wrote necessary unit or integration tests.

.github/workflows/publish-to-docker.yml

@@ -0,0 +1,99 @@
+name: Docker Build and Publish
+
+on:
+  workflow_dispatch:
+    inputs:
+      version:
+        description: 'TestPyPI or PyPI version to build (e.g., 0.0.63.dev20250114)'
+        required: true
+        type: string
+
+jobs:
+  build-and-push:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Log in to the Container registry
+        uses: docker/login-action@v3
+        with:
+          username: ${{ secrets.DOCKERHUB_USERNAME }}
+          password: ${{ secrets.DOCKERHUB_TOKEN }}
+
+      - name: Set version
+        id: version
+        run: |
+          if [ "${{ github.event_name }}" = "push" ]; then
+            echo "VERSION=0.0.63.dev20250114" >> $GITHUB_OUTPUT
+          else
+            echo "VERSION=${{ inputs.version }}" >> $GITHUB_OUTPUT
+          fi
+
+      - name: Check package version availability
+        run: |
+            # Function to check if version exists in a repository
+            check_version() {
+                local repo=$1
+                local status_code=$(curl -s -o /dev/null -w "%{http_code}" "https://$repo.org/project/llama-stack/${{ steps.version.outputs.version }}")
+                return $([ "$status_code" -eq 200 ])
+            }
+
+            # Check TestPyPI first, then PyPI
+            if check_version "test.pypi"; then
+                echo "Version ${{ steps.version.outputs.version }} found in TestPyPI"
+                echo "PYPI_SOURCE=testpypi" >> $GITHUB_ENV
+            elif check_version "pypi"; then
+                echo "Version ${{ steps.version.outputs.version }} found in PyPI"
+                echo "PYPI_SOURCE=pypi" >> $GITHUB_ENV
+            else
+                echo "Error: Version ${{ steps.version.outputs.version }} not found in either TestPyPI or PyPI"
+                exit 1
+            fi
+
+      - name: Install llama-stack
+        run: |
+            if [ "${{ github.event_name }}" = "push" ]; then
+                pip install -e .
+            else
+                if [ "$PYPI_SOURCE" = "testpypi" ]; then
+                    pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple llama-stack==${{ steps.version.outputs.version }}
+                else
+                    pip install llama-stack==${{ steps.version.outputs.version }}
+                fi
+            fi
+
+      - name: Build docker image
+        run: |
+          TEMPLATES=("ollama" "bedrock" "remote-vllm" "fireworks" "together" "tgi" "meta-reference-gpu")
+          for template in "${TEMPLATES[@]}"; do
+            if [ "$PYPI_SOURCE" = "testpypi" ]; then
+                TEST_PYPI_VERSION=${{ steps.version.outputs.version }} llama stack build --template $template --image-type docker
+            else
+                PYPI_VERSION=${{ steps.version.outputs.version }} llama stack build --template $template --image-type docker
+            fi
+          done
+
+      - name: List docker images
+        run: |
+          docker images
+
+      - name: Push to dockerhub
+        run: |
+          TEMPLATES=("ollama" "bedrock" "remote-vllm" "fireworks" "together" "tgi" "meta-reference-gpu")
+          for template in "${TEMPLATES[@]}"; do
+            if [ "$PYPI_SOURCE" = "testpypi" ]; then
+                docker tag distribution-$template:test-${{ steps.version.outputs.version }} llamastack/distribution-$template:test-${{ steps.version.outputs.version }}
+                docker push llamastack/distribution-$template:test-${{ steps.version.outputs.version }}
+            else
+                docker tag distribution-$template:${{ steps.version.outputs.version }} llamastack/distribution-$template:${{ steps.version.outputs.version }}
+                docker push llamastack/distribution-$template:${{ steps.version.outputs.version }}
+            fi
+          done

llama_stack/cli/stack/build.py

@@ -107,7 +107,8 @@ class StackBuild(Subcommand):
                             f"Please specify a image-type (docker | conda | venv) for {args.template}"
                         )
                     self._run_stack_build_command_from_build_config(
-                        build_config, template_name=args.template
+                        build_config,
+                        template_name=args.template,
                     )
                     return
 
@@ -261,7 +262,9 @@ class StackBuild(Subcommand):
         )
 
     def _run_stack_build_command_from_build_config(
-        self, build_config: BuildConfig, template_name: Optional[str] = None
+        self,
+        build_config: BuildConfig,
+        template_name: Optional[str] = None,
     ) -> None:
         import json
         import os

llama_stack/distribution/build.py

@@ -6,6 +6,7 @@
 
 import importlib.resources
 import logging
+import sys
 from enum import Enum
 
 from pathlib import Path
@@ -20,7 +21,7 @@ from llama_stack.distribution.distribution import get_provider_registry
 
 from llama_stack.distribution.utils.config_dirs import BUILDS_BASE_DIR
 
-from llama_stack.distribution.utils.exec import run_with_pty
+from llama_stack.distribution.utils.exec import run_command, run_with_pty
 from llama_stack.providers.datatypes import Api
 
 log = logging.getLogger(__name__)
@@ -102,7 +103,10 @@ def print_pip_install_help(providers: Dict[str, List[Provider]]):
     print()
 
 
-def build_image(build_config: BuildConfig, build_file_path: Path):
+def build_image(
+    build_config: BuildConfig,
+    build_file_path: Path,
+):
     docker_image = build_config.distribution_spec.docker_image or "python:3.10-slim"
 
     normal_deps, special_deps = get_provider_dependencies(
@@ -144,7 +148,12 @@ def build_image(build_config: BuildConfig, build_file_path: Path):
     if special_deps:
         args.append("#".join(special_deps))
 
+    is_terminal = sys.stdin.isatty()
+    if is_terminal:
         return_code = run_with_pty(args)
+    else:
+        return_code = run_command(args)
+
     if return_code != 0:
         log.error(
             f"Failed to build target {build_config.name} with return code {return_code}",

llama_stack/distribution/build_container.sh

@@ -82,13 +82,17 @@ fi
 # Add pip dependencies first since llama-stack is what will change most often
 # so we can reuse layers.
 if [ -n "$pip_dependencies" ]; then
-  add_to_docker "RUN pip install --no-cache $pip_dependencies"
+  add_to_docker << EOF
+RUN pip install --no-cache $pip_dependencies
+EOF
 fi
 
 if [ -n "$special_pip_deps" ]; then
   IFS='#' read -ra parts <<<"$special_pip_deps"
   for part in "${parts[@]}"; do
-    add_to_docker "RUN pip install --no-cache $part"
+    add_to_docker <<EOF
+RUN pip install --no-cache $part
+EOF
   done
 fi
 
@@ -104,14 +108,19 @@ if [ -n "$LLAMA_STACK_DIR" ]; then
   # Install in editable format. We will mount the source code into the container
   # so that changes will be reflected in the container without having to do a
   # rebuild. This is just for development convenience.
-  add_to_docker "RUN pip install --no-cache -e $stack_mount"
+  add_to_docker << EOF
+RUN pip install --no-cache -e $stack_mount
+EOF
 else
   if [ -n "$TEST_PYPI_VERSION" ]; then
     # these packages are damaged in test-pypi, so install them first
-    add_to_docker "RUN pip install fastapi libcst"
+    add_to_docker << EOF
+RUN pip install fastapi libcst
+EOF
     add_to_docker << EOF
 RUN pip install --no-cache --extra-index-url https://test.pypi.org/simple/ \
   llama-models==$TEST_PYPI_VERSION llama-stack-client==$TEST_PYPI_VERSION llama-stack==$TEST_PYPI_VERSION
+
 EOF
   else
     if [ -n "$PYPI_VERSION" ]; then
@@ -119,7 +128,9 @@ EOF
     else
       SPEC_VERSION="llama-stack"
     fi
-    add_to_docker "RUN pip install --no-cache $SPEC_VERSION"
+    add_to_docker << EOF
+RUN pip install --no-cache $SPEC_VERSION
+EOF
   fi
 fi
 

llama_stack/distribution/utils/exec.py

@@ -98,9 +98,11 @@ def run_with_pty(command):
 
 
 def run_command(command):
-    process = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+    try:
-    output, error = process.communicate()
+        result = subprocess.run(command, capture_output=True, text=True, check=True)
-    if process.returncode != 0:
+        print("Script Output\n", result.stdout)
-        log.error(f"Error: {error.decode('utf-8')}")
+        return result.returncode
-        sys.exit(1)
+    except subprocess.CalledProcessError as e:
-    return output.decode("utf-8")
+        print("Error running script:", e)
+        print("Error output:", e.stderr)
+        return e.returncode