feat: Implement pluggable multi-provider sandbox architecture (#12820)

## Summary

Implement a flexible sandbox provider system supporting both
self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for
secure code execution in agent workflows.

**Key Changes:**
-  Aliyun Code Interpreter provider using official
`agentrun-sdk>=0.0.16`
-  Self-managed provider with gVisor (runsc) security
-  Arguments parameter support for dynamic code execution
-  Database-only configuration (removed fallback logic)
-  Configuration scripts for quick setup

Issue #12479

## Features

### 🔌 Provider Abstraction Layer

**1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`)
- Wraps existing executor_manager HTTP API
- gVisor (runsc) for secure container isolation
- Configurable pool size, timeout, retry logic
- Languages: Python, Node.js, JavaScript
- ⚠️ **Requires**: gVisor installation, Docker, base images

**2. Aliyun Code Interpreter**
(`agent/sandbox/providers/aliyun_codeinterpreter.py`)
- SaaS integration using official agentrun-sdk
- Serverless microVM execution with auto-authentication
- Hard timeout: 30 seconds max
- Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`,
`AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION`
- Automatically wraps code to call `main()` function

**3. E2B Provider** (`agent/sandbox/providers/e2b.py`)
- Placeholder for future integration

### ⚙️ Configuration System

- `conf/system_settings.json`: Default provider =
`aliyun_codeinterpreter`
- `agent/sandbox/client.py`: Enforces database-only configuration
- Admin UI: `/admin/sandbox-settings`
- Configuration validation via `validate_config()` method
- Health checks for all providers

### 🎯 Key Capabilities

**Arguments Parameter Support:**
All providers support passing arguments to `main()` function:
```python
# User code
def main(name: str, count: int) -> dict:
    return {"message": f"Hello {name}!" * count}

# Executed with: arguments={"name": "World", "count": 3}
# Result: {"message": "Hello World!Hello World!Hello World!"}
```

**Self-Describing Providers:**
Each provider implements `get_config_schema()` returning form
configuration for Admin UI

**Error Handling:**
Structured `ExecutionResult` with stdout, stderr, exit_code,
execution_time

## Configuration Scripts

Two scripts for quick Aliyun sandbox setup:

**Shell Script (requires jq):**
```bash
source scripts/configure_aliyun_sandbox.sh
```

**Python Script (interactive):**
```bash
python3 scripts/configure_aliyun_sandbox.py
```

## Testing

```bash
# Unit tests
uv run pytest agent/sandbox/tests/test_providers.py -v

# Aliyun provider tests
uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v

# Integration tests (requires credentials)
uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v

# Quick SDK validation
python3 agent/sandbox/tests/verify_sdk.py
```

**Test Coverage:**
- 30 unit tests for provider abstraction
- Provider-specific tests for Aliyun
- Integration tests with real API
- Security tests for executor_manager

## Documentation

- `docs/develop/sandbox_spec.md` - Complete architecture specification
- `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy
sandbox
- `agent/sandbox/tests/QUICKSTART.md` - Quick start guide
- `agent/sandbox/tests/README.md` - Testing documentation

## Breaking Changes

⚠️ **Migration Required:**

1. **Directory Move**: `sandbox/` → `agent/sandbox/`
   - Update imports: `from sandbox.` → `from agent.sandbox.`

2. **Mandatory Configuration**: 
   - SystemSettings must have `sandbox.provider_type` configured
   - Removed fallback default values
- Configuration must exist in database (from
`conf/system_settings.json`)

3. **Aliyun Credentials**:
   - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`)
   - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID)

4. **Self-Managed Provider**:
   - gVisor (runsc) must be installed for security
   - Install: `go install gvisor.dev/gvisor/runsc@latest`

## Database Schema Changes

```python
# SystemSettings.value: CharField → TextField
api/db/db_models.py: Changed for unlimited config length

# SystemSettingsService.get_by_name(): Fixed query precision
api/db/services/system_settings_service.py: startswith → exact match
```

## Files Changed

### Backend (Python)
- `agent/sandbox/providers/base.py` - SandboxProvider ABC interface
- `agent/sandbox/providers/manager.py` - ProviderManager
- `agent/sandbox/providers/self_managed.py` - Self-managed provider
- `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider
- `agent/sandbox/providers/e2b.py` - E2B provider (placeholder)
- `agent/sandbox/client.py` - Unified client (enforces DB-only config)
- `agent/tools/code_exec.py` - Updated to use provider system
- `admin/server/services.py` - SandboxMgr with registry & validation
- `admin/server/routes.py` - 5 sandbox API endpoints
- `conf/system_settings.json` - Default: aliyun_codeinterpreter
- `api/db/db_models.py` - TextField for SystemSettings.value
- `api/db/services/system_settings_service.py` - Exact match query

### Frontend (TypeScript/React)
- `web/src/pages/admin/sandbox-settings.tsx` - Settings UI
- `web/src/services/admin-service.ts` - Sandbox service functions
- `web/src/services/admin.service.d.ts` - Type definitions
- `web/src/utils/api.ts` - Sandbox API endpoints

### Documentation
- `docs/develop/sandbox_spec.md` - Architecture spec
- `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide
- `agent/sandbox/tests/QUICKSTART.md` - Quick start
- `agent/sandbox/tests/README.md` - Testing guide

### Configuration Scripts
- `scripts/configure_aliyun_sandbox.sh` - Shell script (jq)
- `scripts/configure_aliyun_sandbox.py` - Python script

### Tests
- `agent/sandbox/tests/test_providers.py` - 30 unit tests
- `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests
- `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` -
Integration tests
- `agent/sandbox/tests/verify_sdk.py` - SDK validation

## Architecture

```
Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B]
                                      ↓
                                  SystemSettings
```

## Usage

### 1. Configure Provider

**Via Admin UI:**
1. Navigate to `/admin/sandbox-settings`
2. Select provider (Aliyun Code Interpreter / Self-Managed)
3. Fill in configuration
4. Click "Test Connection" to verify
5. Click "Save" to apply

**Via Configuration Scripts:**
```bash
# Aliyun provider
export AGENTRUN_ACCESS_KEY_ID="xxx"
export AGENTRUN_ACCESS_KEY_SECRET="yyy"
export AGENTRUN_ACCOUNT_ID="zzz"
export AGENTRUN_REGION="cn-shanghai"
source scripts/configure_aliyun_sandbox.sh
```

### 2. Restart Service

```bash
cd docker
docker compose restart ragflow-server
```

### 3. Execute Code in Agent

```python
from agent.sandbox.client import execute_code

result = execute_code(
    code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}',
    language="python",
    timeout=30,
    arguments={"name": "World"}
)

print(result.stdout)  # {"message": "Hello World!"}
```

## Troubleshooting

### "Container pool is busy" (Self-Managed)
- **Cause**: Pool exhausted (default: 1 container in `.env`)
- **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+

### "Sandbox provider type not configured"
- **Cause**: Database missing configuration
- **Fix**: Run config script or set via Admin UI

### "gVisor not found"
- **Cause**: runsc not installed
- **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp
~/go/bin/runsc /usr/local/bin/`

### Aliyun authentication errors
- **Cause**: Wrong environment variable names
- **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`)

## Checklist

- [x] All tests passing (30 unit tests + integration tests)
- [x] Documentation updated (spec, migration guide, quickstart)
- [x] Type definitions added (TypeScript)
- [x] Admin UI implemented
- [x] Configuration validation
- [x] Health checks implemented
- [x] Error handling with structured results
- [x] Breaking changes documented
- [x] Configuration scripts created
- [x] gVisor requirements documented

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
This commit is contained in:
Zhichang Yu
2026-01-28 13:28:21 +08:00
committed by GitHub
parent b57c82b122
commit fd11aca8e5
72 changed files with 6914 additions and 404 deletions

View File

@ -14,6 +14,7 @@
# limitations under the License.
#
import json
import os
import logging
import re
@ -372,7 +373,23 @@ class SettingsMgr:
elif len(settings) > 1:
raise AdminException(f"Can't update more than 1 setting: {name}")
else:
raise AdminException(f"No setting: {name}")
# Create new setting if it doesn't exist
# Determine data_type based on name and value
if name.startswith("sandbox."):
data_type = "json"
elif name.endswith(".enabled"):
data_type = "boolean"
else:
data_type = "string"
new_setting = {
"name": name,
"value": str(value),
"source": "admin",
"data_type": data_type,
}
SystemSettingsService.save(**new_setting)
class ConfigMgr:
@ -407,3 +424,300 @@ class EnvironmentsMgr:
result.append(env_kv)
return result
class SandboxMgr:
"""Manager for sandbox provider configuration and operations."""
# Provider registry with metadata
PROVIDER_REGISTRY = {
"self_managed": {
"name": "Self-Managed",
"description": "On-premise deployment using Daytona/Docker",
"tags": ["self-hosted", "low-latency", "secure"],
},
"aliyun_codeinterpreter": {
"name": "Aliyun Code Interpreter",
"description": "Aliyun Function Compute Code Interpreter - Code execution in serverless microVMs",
"tags": ["saas", "cloud", "scalable", "aliyun"],
},
"e2b": {
"name": "E2B",
"description": "E2B Cloud - Code Execution Sandboxes",
"tags": ["saas", "fast", "global"],
},
}
@staticmethod
def list_providers():
"""List all available sandbox providers."""
result = []
for provider_id, metadata in SandboxMgr.PROVIDER_REGISTRY.items():
result.append({
"id": provider_id,
**metadata
})
return result
@staticmethod
def get_provider_config_schema(provider_id: str):
"""Get configuration schema for a specific provider."""
from agent.sandbox.providers import (
SelfManagedProvider,
AliyunCodeInterpreterProvider,
E2BProvider,
)
schemas = {
"self_managed": SelfManagedProvider.get_config_schema(),
"aliyun_codeinterpreter": AliyunCodeInterpreterProvider.get_config_schema(),
"e2b": E2BProvider.get_config_schema(),
}
if provider_id not in schemas:
raise AdminException(f"Unknown provider: {provider_id}")
return schemas.get(provider_id, {})
@staticmethod
def get_config():
"""Get current sandbox configuration."""
try:
# Get active provider type
provider_type_settings = SystemSettingsService.get_by_name("sandbox.provider_type")
if not provider_type_settings:
# Return default config if not set
provider_type = "self_managed"
else:
provider_type = provider_type_settings[0].value
# Get provider-specific config
provider_config_settings = SystemSettingsService.get_by_name(f"sandbox.{provider_type}")
if not provider_config_settings:
provider_config = {}
else:
try:
provider_config = json.loads(provider_config_settings[0].value)
except json.JSONDecodeError:
provider_config = {}
return {
"provider_type": provider_type,
"config": provider_config,
}
except Exception as e:
raise AdminException(f"Failed to get sandbox config: {str(e)}")
@staticmethod
def set_config(provider_type: str, config: dict, set_active: bool = True):
"""
Set sandbox provider configuration.
Args:
provider_type: Provider identifier (e.g., "self_managed", "e2b")
config: Provider configuration dictionary
set_active: If True, also update the active provider. If False,
only update the configuration without switching providers.
Default: True
Returns:
Dictionary with updated provider_type and config
"""
from agent.sandbox.providers import (
SelfManagedProvider,
AliyunCodeInterpreterProvider,
E2BProvider,
)
try:
# Validate provider type
if provider_type not in SandboxMgr.PROVIDER_REGISTRY:
raise AdminException(f"Unknown provider type: {provider_type}")
# Get provider schema for validation
schema = SandboxMgr.get_provider_config_schema(provider_type)
# Validate config against schema
for field_name, field_schema in schema.items():
if field_schema.get("required", False) and field_name not in config:
raise AdminException(f"Required field '{field_name}' is missing")
# Type validation
if field_name in config:
field_type = field_schema.get("type")
if field_type == "integer":
if not isinstance(config[field_name], int):
raise AdminException(f"Field '{field_name}' must be an integer")
elif field_type == "string":
if not isinstance(config[field_name], str):
raise AdminException(f"Field '{field_name}' must be a string")
elif field_type == "bool":
if not isinstance(config[field_name], bool):
raise AdminException(f"Field '{field_name}' must be a boolean")
# Range validation for integers
if field_type == "integer" and field_name in config:
min_val = field_schema.get("min")
max_val = field_schema.get("max")
if min_val is not None and config[field_name] < min_val:
raise AdminException(f"Field '{field_name}' must be >= {min_val}")
if max_val is not None and config[field_name] > max_val:
raise AdminException(f"Field '{field_name}' must be <= {max_val}")
# Provider-specific custom validation
provider_classes = {
"self_managed": SelfManagedProvider,
"aliyun_codeinterpreter": AliyunCodeInterpreterProvider,
"e2b": E2BProvider,
}
provider = provider_classes[provider_type]()
is_valid, error_msg = provider.validate_config(config)
if not is_valid:
raise AdminException(f"Provider validation failed: {error_msg}")
# Update provider_type only if set_active is True
if set_active:
SettingsMgr.update_by_name("sandbox.provider_type", provider_type)
# Always update the provider config
config_json = json.dumps(config)
SettingsMgr.update_by_name(f"sandbox.{provider_type}", config_json)
return {"provider_type": provider_type, "config": config}
except AdminException:
raise
except Exception as e:
raise AdminException(f"Failed to set sandbox config: {str(e)}")
@staticmethod
def test_connection(provider_type: str, config: dict):
"""
Test connection to sandbox provider by executing a simple Python script.
This creates a temporary sandbox instance and runs a test code to verify:
- Connection credentials are valid
- Sandbox can be created
- Code execution works correctly
Args:
provider_type: Provider identifier
config: Provider configuration dictionary
Returns:
dict with test results including stdout, stderr, exit_code, execution_time
"""
try:
from agent.sandbox.providers import (
SelfManagedProvider,
AliyunCodeInterpreterProvider,
E2BProvider,
)
# Instantiate provider based on type
provider_classes = {
"self_managed": SelfManagedProvider,
"aliyun_codeinterpreter": AliyunCodeInterpreterProvider,
"e2b": E2BProvider,
}
if provider_type not in provider_classes:
raise AdminException(f"Unknown provider type: {provider_type}")
provider = provider_classes[provider_type]()
# Initialize with config
if not provider.initialize(config):
raise AdminException(f"Failed to initialize provider '{provider_type}'")
# Create a temporary sandbox instance for testing
instance = provider.create_instance(template="python")
if not instance or instance.status != "READY":
raise AdminException(f"Failed to create sandbox instance. Status: {instance.status if instance else 'None'}")
# Simple test code that exercises basic Python functionality
test_code = """
# Test basic Python functionality
import sys
import json
import math
print("Python version:", sys.version)
print("Platform:", sys.platform)
# Test basic calculations
result = 2 + 2
print(f"2 + 2 = {result}")
# Test JSON operations
data = {"test": "data", "value": 123}
print(f"JSON dump: {json.dumps(data)}")
# Test math operations
print(f"Math.sqrt(16) = {math.sqrt(16)}")
# Test error handling
try:
x = 1 / 1
print("Division test: OK")
except Exception as e:
print(f"Error: {e}")
# Return success indicator
print("TEST_PASSED")
"""
# Execute test code with timeout
execution_result = provider.execute_code(
instance_id=instance.instance_id,
code=test_code,
language="python",
timeout=10 # 10 seconds timeout
)
# Clean up the test instance (if provider supports it)
try:
if hasattr(provider, 'terminate_instance'):
provider.terminate_instance(instance.instance_id)
logging.info(f"Cleaned up test instance {instance.instance_id}")
else:
logging.warning(f"Provider {provider_type} does not support terminate_instance, test instance may leak")
except Exception as cleanup_error:
logging.warning(f"Failed to cleanup test instance {instance.instance_id}: {cleanup_error}")
# Build detailed result message
success = execution_result.exit_code == 0 and "TEST_PASSED" in execution_result.stdout
message_parts = [
f"Test {success and 'PASSED' or 'FAILED'}",
f"Exit code: {execution_result.exit_code}",
f"Execution time: {execution_result.execution_time:.2f}s"
]
if execution_result.stdout.strip():
stdout_preview = execution_result.stdout.strip()[:200]
message_parts.append(f"Output: {stdout_preview}...")
if execution_result.stderr.strip():
stderr_preview = execution_result.stderr.strip()[:200]
message_parts.append(f"Errors: {stderr_preview}...")
message = " | ".join(message_parts)
return {
"success": success,
"message": message,
"details": {
"exit_code": execution_result.exit_code,
"execution_time": execution_result.execution_time,
"stdout": execution_result.stdout,
"stderr": execution_result.stderr,
}
}
except AdminException:
raise
except Exception as e:
import traceback
error_details = traceback.format_exc()
raise AdminException(f"Connection test failed: {str(e)}\\n\\nStack trace:\\n{error_details}")