fix: run cognee in Docker [COG-1961] (#775) (#779)

<!-- .github/pull_request_template.md -->

## Description
Resolve issue with .venv being broken when using docker compose with
Cognee

## DCO Affirmation
I affirm that all code in every commit of this pull request conforms to
the terms of the Topoteretes Developer Certificate of Origin.

---------

<!-- .github/pull_request_template.md -->

## Description
<!-- Provide a clear description of the changes in this PR -->

## DCO Affirmation
I affirm that all code in every commit of this pull request conforms to
the terms of the Topoteretes Developer Certificate of Origin.

Co-authored-by: Igor Ilic <30923996+dexters1@users.noreply.github.com>

Author: borisarzentar

.github/workflows/dockerhub-mcp.yml

@@ -36,9 +36,10 @@ jobs:
         id: build
         uses: docker/build-push-action@v5
         with:
-          context: cognee-mcp
+          context: .
           platforms: linux/amd64,linux/arm64
           push: true
+          file: cognee-mcp/Dockerfile
           tags: ${{ steps.meta.outputs.tags }}
           labels: ${{ steps.meta.outputs.labels }}
           cache-from: type=registry,ref=cognee/cognee-mcp:buildcache

Dockerfile

@@ -25,7 +25,7 @@ ENV DEBUG=${DEBUG}
 ENV PIP_NO_CACHE_DIR=true
 ENV PATH="${PATH}:/root/.poetry/bin"
 
-RUN apt-get update && apt-get install
+RUN apt-get update
 
 RUN apt-get install -y \
   gcc \

alembic/versions/482cd6517ce4_add_default_user.py

@@ -0,0 +1,28 @@
+"""Add default user
+
+Revision ID: 482cd6517ce4
+Revises: 8057ae7329c2
+Create Date: 2024-10-16 22:17:18.634638
+
+"""
+
+from typing import Sequence, Union
+
+from sqlalchemy.util import await_only
+
+from cognee.modules.users.methods import create_default_user, delete_user
+
+
+# revision identifiers, used by Alembic.
+revision: str = "482cd6517ce4"
+down_revision: Union[str, None] = "8057ae7329c2"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = "8057ae7329c2"
+
+
+def upgrade() -> None:
+    await_only(create_default_user())
+
+
+def downgrade() -> None:
+    await_only(delete_user("default_user@example.com"))

alembic/versions/8057ae7329c2_initial_migration.py

@@ -0,0 +1,27 @@
+"""Initial migration
+
+Revision ID: 8057ae7329c2
+Revises:
+Create Date: 2024-10-02 12:55:20.989372
+
+"""
+
+from typing import Sequence, Union
+from sqlalchemy.util import await_only
+from cognee.infrastructure.databases.relational import get_relational_engine
+
+# revision identifiers, used by Alembic.
+revision: str = "8057ae7329c2"
+down_revision: Union[str, None] = None
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    db_engine = get_relational_engine()
+    await_only(db_engine.create_database())
+
+
+def downgrade() -> None:
+    db_engine = get_relational_engine()
+    await_only(db_engine.delete_database())

cognee-mcp/Dockerfile

@@ -22,7 +22,7 @@ RUN apt-get update && apt-get install -y \
     libpq-dev
 
 # Copy pyproject.toml and lockfile first for better caching
-COPY pyproject.toml uv.lock ./
+COPY ./cognee-mcp/pyproject.toml ./cognee-mcp/uv.lock ./cognee-mcp/entrypoint.sh ./
 
 # Install the project's dependencies using the lockfile and settings
 RUN --mount=type=cache,target=/root/.cache/uv \
@@ -31,19 +31,25 @@ RUN --mount=type=cache,target=/root/.cache/uv \
 # Copy .env file first if it exists (for environment variables)
 COPY .env* /app/
 
+# Copy Alembic configuration
+COPY alembic.ini /app/alembic.ini
+COPY alembic/ /app/alembic
+
 # Then, add the rest of the project source code and install it
 # Installing separately from its dependencies allows optimal layer caching
-COPY . /app
+COPY ./cognee-mcp /app
 RUN --mount=type=cache,target=/root/.cache/uv \
     uv sync --frozen --no-dev --no-editable
 
 FROM python:3.12-slim-bookworm
 
 WORKDIR /app
- 
+
 COPY --from=uv /root/.local /root/.local
 COPY --from=uv /app /app
 
+RUN chmod +x /app/entrypoint.sh
+
 # Place executables in the environment at the front of the path
 ENV PATH="/app/.venv/bin:$PATH"
 
@@ -54,4 +60,4 @@ ENV PYTHONPATH=/app
 
 # Use the application name from pyproject.toml for normal operation
 # For testing, we'll override this with a direct command
-ENTRYPOINT ["cognee"]
+ENTRYPOINT ["/app/entrypoint.sh"]

cognee-mcp/README.md

@@ -84,3 +84,22 @@ To apply new changes while developing cognee you need to do:
 1. `poetry lock` in cognee folder
 2. `uv sync --dev --all-extras --reinstall`
 3. `mcp dev src/server.py`
+
+### Development
+In order to use local cognee build, run in root of the cognee repo:
+```bash
+poetry build -o ./cognee-mcp/sources
+```
+After the build process is done, change the cognee library dependency inside the `cognee-mcp/pyproject.toml` from
+```toml
+cognee[postgres,codegraph,gemini,huggingface]==0.1.38
+```
+to
+```toml
+cognee[postgres,codegraph,gemini,huggingface]
+```
+After that add the following snippet to the same file (`cognee-mcp/pyproject.toml`).
+```toml
+[tool.uv.sources]
+cognee = { path = "sources/cognee-0.1.38-py3-none-any.whl" }
+```

cognee-mcp/entrypoint.sh

@@ -0,0 +1,41 @@
+#!/bin/bash
+
+set -e  # Exit on error
+echo "Debug mode: $DEBUG"
+echo "Environment: $ENVIRONMENT"
+
+# Run Alembic migrations with proper error handling.
+# Note on UserAlreadyExists error handling:
+# During database migrations, we attempt to create a default user. If this user
+# already exists (e.g., from a previous deployment or migration), it's not a
+# critical error and shouldn't prevent the application from starting. This is
+# different from other migration errors which could indicate database schema
+# inconsistencies and should cause the startup to fail. This check allows for
+# smooth redeployments and container restarts while maintaining data integrity.
+echo "Running database migrations..."
+MIGRATION_OUTPUT=$(uv run alembic upgrade head 2>&1) || {
+    if [[ $MIGRATION_OUTPUT == *"UserAlreadyExists"* ]] || [[ $MIGRATION_OUTPUT == *"User default_user@example.com already exists"* ]]; then
+        echo "Warning: Default user already exists, continuing startup..."
+    else
+        echo "Migration failed with unexpected error:"
+        echo "$MIGRATION_OUTPUT"
+        exit 1
+    fi
+}
+
+echo "Starting Cognee MCP Server"
+
+# Add startup delay to ensure DB is ready
+sleep 2
+
+# Modified Gunicorn startup with error handling
+if [ "$ENVIRONMENT" = "dev" ] || [ "$ENVIRONMENT" = "local" ]; then
+    if [ "$DEBUG" = "true" ]; then
+        echo "Waiting for the debugger to attach..."
+        exec python -m debugpy --wait-for-client --listen 0.0.0.0:5678 -m cognee
+    else
+        exec cognee
+    fi
+else
+    exec cognee
+fi

cognee-mcp/src/server.py

@@ -36,11 +36,11 @@ async def list_tools() -> list[types.Tool]:
                     },
                     "graph_model_file": {
                         "type": "string",
-                        "description": "The path to the graph model file",
+                        "description": "The path to the graph model file (Optional)",
                     },
                     "graph_model_name": {
                         "type": "string",
-                        "description": "The name of the graph model",
+                        "description": "The name of the graph model (Optional)",
                     },
                 },
                 "required": ["text"],

docker-compose.yml

@@ -7,8 +7,8 @@ services:
       context: .
       dockerfile: Dockerfile
     volumes:
-      - .:/app
-      - /app/cognee-frontend/ # Ignore frontend code
+      - ./cognee:/app/cognee
+      - .env:/app/.env
     environment:
       - DEBUG=false # Change to true if debugging
       - HOST=0.0.0.0
@@ -26,10 +26,8 @@ services:
           cpus: "4.0"
           memory: 8GB
 
-# NOTE: Frontend is a work in progress and is not intended to be used by users yet.
-#       If you want to use Cognee with a UI environment you can run the cognee-gui.py script or
-#       integrate the Cognee MCP Server to Cursor / Claude Desktop / Visual Studio Code ( through Cline/Roo )
-
+  # NOTE: Frontend is a work in progress and supports minimum amount of features required to be functional.
+  # If you want to use Cognee with a UI environment you can integrate the Cognee MCP Server into Cursor / Claude Desktop / Visual Studio Code (through Cline/Roo)
   frontend:
     container_name: frontend
     profiles: