Merge pull request #211 from stacklok/restore_backup_in_image

feat: expose an entrypoint to restore from backup

Author: yrobla

Dockerfile

@@ -30,7 +30,7 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     && rm -rf /var/lib/apt/lists/*
 
 # Create a non-root user and switch to it
-RUN adduser --system --no-create-home codegate --uid 1000
+RUN useradd -m -u 1000 -r codegate 
 USER codegate
 WORKDIR /app
 
@@ -41,9 +41,6 @@ COPY --from=builder /app /app
 # Set the PYTHONPATH environment variable
 ENV PYTHONPATH=/app/src
 
-# Allow to expose weaviate_data volume
-VOLUME ["/app/weaviate_data"]
-
 # Set the container's default entrypoint
 EXPOSE 8989
-ENTRYPOINT ["python", "-m", "src.codegate.cli", "serve", "--port", "8989", "--host", "0.0.0.0"]
+ENTRYPOINT ["/app/scripts/entrypoint.sh", "/app/weaviate_backup", "backup"]

scripts/entrypoint.sh

@@ -0,0 +1,14 @@
+#!/bin/bash
+
+# Check if the backup directory exists and handle the restore
+if [ -d "$1" ] && [ -d "$1/$2" ]; then
+    echo "Restoring backup from $1/$2..."
+    # Your restore logic here, e.g., running a Python script or restoring a database
+    python -m src.codegate.cli restore_backup "$1" "$2"
+else
+    echo "No backup found at $1/$2. Skipping restore."
+fi
+
+# Step 2: Start the main application (serve)
+echo "Starting the application..."
+exec python -m src.codegate.cli serve --port 8989 --host 0.0.0.0

scripts/import_packages.py

@@ -113,6 +113,7 @@ async def run_import(self):
         await self.add_data()
         self.take_backup()
 
+
 if __name__ == "__main__":
     importer = PackageImporter()
     asyncio.run(importer.run_import())

src/codegate/cli.py

@@ -5,6 +5,7 @@
 from typing import Dict, Optional
 
 import click
+from src.codegate.storage.utils import restore_storage_backup
 import structlog
 
 from codegate.codegate_logging import LogFormat, LogLevel, setup_logging
@@ -191,6 +192,29 @@ def serve(
         sys.exit(1)
 
 
+@cli.command()
+@click.option(
+    "--backup-path",
+    type=click.Path(exists=True, file_okay=False, path_type=Path),
+    required=True,
+    help="Directory path where the backup file is located.",
+)
+@click.option(
+    "--backup-name",
+    type=str,
+    required=True,
+    help="Name of the backup file to restore.",
+)
+def restore_backup(backup_path: Path, backup_name: str) -> None:
+    """Restore the database from the specified backup."""
+    try:
+        restore_storage_backup(backup_path, backup_name)
+        click.echo(f"Successfully restored the backup '{backup_name}' from {backup_path}.")
+    except Exception as e:
+        click.echo(f"Error restoring backup: {e}", err=True)
+        sys.exit(1)
+
+
 def main() -> None:
     """Main entry point for the CLI."""
     cli()

src/codegate/storage/utils.py

@@ -0,0 +1,20 @@
+import weaviate
+from weaviate.embedded import EmbeddedOptions
+
+
+def restore_storage_backup(backup_path: str, backup_name: str):
+    client = weaviate.WeaviateClient(
+        embedded_options=EmbeddedOptions(
+            persistence_data_path="./weaviate_data", grpc_port=50052,
+            additional_env_vars={"ENABLE_MODULES": "backup-filesystem",
+                                  "BACKUP_FILESYSTEM_PATH": backup_path}
+        )
+    )
+    client.connect()
+    try:
+        client.backup.restore(backup_id=backup_name,
+                              backend="filesystem", wait_for_completion=True)
+    except Exception as e:
+        print(f"Failed to restore backup: {e}")
+    finally:
+        client.close()