### Node.js Handler with Initializer Function

Source: https://www.volcengine.com/docs/6662/116908

This example demonstrates adding an initializer function to perform global setup, such as database connections, before the handler is invoked. This helps reduce cold start times.

```javascript
// vefaas-nodejs14-default
exports.handler = async function handler (event, context) {
return {
statusCode: 200,
headers: {'Content-Type': 'application/json'},
body: JSON.stringify({ 'vefaas-nodejs14-default': 'hello world'}),
};
};

// vefaas will try to do initializer when start up the instance, to avoid cold start from user perspective. 
// initializer will only be executed for one time when there is no exception. 
// otherwise initializer will be executed again before every invoke, till successfully inited.
exports.initializer = async function (context) {
console.log('initializer begin');
// replace the following line, with the real init logic
await new Promise(resolve => setTimeout(resolve, 3 * 1000));
console.log('initializer end');
};
```

--------------------------------

### TLS Configuration Example

Source: https://www.volcengine.com/docs/6662/1327609

Example of TLS configuration for logging, including project ID and topic ID.

```json
{
  "EnableLog": true,
  "TlsProjectId": "a40202c5-fe22-4ea7-******",
  "TlsTopicId": "2ff955cb-1f03-4b9b-******"
}
```

--------------------------------

### CreateTimer Request and Response Examples

Source: https://www.volcengine.com/docs/6662/1391867?lang=zh

Examples showing the HTTP POST request structure for creating a timer and the corresponding JSON response from the server.

```json
POST http://open.volcengineapi.com/?Action=CreateTimer&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "FunctionId": "qtqzj**",
    "Name": "doc-timer",
    "Description": "Timer触发器",
    "Enabled": true,
    "Crontab": "*/5 * * * *",
    "EnableConcurrency": false
}
```

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "202411291616311B603AD90**",
        "Action": "CreateTimer",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "FunctionId": "qtqzj**",
        "Id": "ing3qa2**",
        "Name": "doc-timer",
        "Description": "Timer触发器",
        "Enabled": true,
        "Crontab": "*/5 * * * *",
        "Payload": "",
        "CreationTime": "2024-11-29 16:16:31.540584873 +0800 CST m=+65768.359252459",
        "LastUpdateTime": "2024-11-29 16:16:31.540584953 +0800 CST m=+65768.359252522",
        "EnableConcurrency": false,
        "Retries": 0
    }
}
```

--------------------------------

### ListReleaseRecords Response Example

Source: https://www.volcengine.com/docs/6662/1327610?lang=zh

Example of a successful 200 OK response containing the list of release records and metadata.

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "ListReleaseRecords",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "Items": [
            {
                "Id": "cytso41iugitj01x",
                "Status": "finished",
                "FinishTime": "2024-06-14 15:31:07.143863491 +0800 CST m=+88714.902440914",
                "FunctionId": "nuvpugx5",
                "Description": "Support xxx feature",
                "CreationTime": "2024-06-14 15:31:07.143863441 +0800 CST m=+88714.902440864",
                "LastUpdateTime": "2024-06-14 15:31:07.143863491 +0800 CST m=+88714.902440914",
                "SourceRevisionNumber": 9,
                "TargetRevisionNumber": 10
            }
        ],
        "Total": 1
    }
}
```

--------------------------------

### GetReleaseStatus Request and Response Examples

Source: https://www.volcengine.com/docs/6662/1327338

Examples of the HTTP POST request and the corresponding JSON response for checking function release status.

```json
POST http://open.volcengineapi.com/?Action=GetReleaseStatus&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "FunctionId": "nw******"
}
```

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "GetReleaseStatus",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "FunctionId": "nwwzoaq1",
        "Status": "done",
        "StatusMessage": "Function release finished successfully",
        "StableRevisionNumber": 1,
        "NewRevisionNumber": 1,
        "OldRevisionNumber": 1,
        "StartTime": "2024-06-27 07:05:12",
        "TargetTrafficWeight": 100,
        "CurrentTrafficWeight": 50,
        "ErrorCode": "function_start_failed",
        "FailedInstanceLogs": "https://vefaas-sit.tos-s3-cn-boe.volces.com/release_failed_instance_logs/******",
        "ReleaseRecordId": "xozo******"
    }
}
```

--------------------------------

### ListFunctionInstances Request Example

Source: https://www.volcengine.com/docs/6662/1327336?lang=zh

This example demonstrates how to make a POST request to the ListFunctionInstances API to query function instances. Ensure the Content-Type is set to application/json.

```http
POST http://open.volcengineapi.com/?Action=ListFunctionInstances&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "FunctionId": "nw******"
}

```

--------------------------------

### VPC Configuration Example

Source: https://www.volcengine.com/docs/6662/1327609

Example of VPC configuration for a function, including VPC ID, subnet IDs, security group IDs, and internet access settings.

```json
{
  "EnableVpc": true,
  "VpcId": "vpc-rrag******",
  "SubnetIds": [
    "subnet-milb******"
  ],
  "SecurityGroupIds": [
    "sg-rrag******"
  ],
  "EnableSharedInternetAccess": true
}
```

--------------------------------

### PrecacheSandboxImages Request and Response Examples

Source: https://www.volcengine.com/docs/6662/1824712?lang=zh

Examples showing the HTTP POST request structure and the corresponding JSON response for the PrecacheSandboxImages API.

```JSON
POST http://open.volcengineapi.com/?Action=PrecacheSandboxImages&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
"ImageUrls": [
"vefaas-******.cr.volces.com/vefaas-cr/******:latest"
]
}
```

```JSON
HTTP/1.1 200 OK
Content-Type: application/json
{
"ResponseMetadata": {
"RequestId": "20230604110420****100232280022D31",
"Action": "PrecacheSandboxImages",
"Version": "2024-06-06",
"Service": "vefaas",
"Region": "cn-beijing"
},
"Result": {
"TicketId": "948410ee80954******"
}
}
```

--------------------------------

### Create Function HTTP Request Example

Source: https://www.volcengine.com/docs/6662/1262132

This example shows the HTTP POST request structure for creating a function, including the endpoint and headers.

```http
POST http://open.volcengineapi.com/?Action=CreateFunction&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
```

--------------------------------

### ListFunctionInstances Response Example

Source: https://www.volcengine.com/docs/6662/1327336?lang=zh

This is an example of a successful response from the ListFunctionInstances API, showing the structure of returned function instance data, including metadata and a list of items.

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "ListFunctionInstances",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "Items": [
            {
                "Id": "nw******",
                "CreationTime": "2024-06-25 11:17:11.492301904 +0800 CST",
                "InstanceName": "nw******-vd9qztck6t-74b9cf8975-bkd99",
                "InstanceStatus": "Starting",
                "RevisionNumber": 1
            }
        ],
        "Total": 1
    }
}

```

--------------------------------

### Startup script for host and port configuration

Source: https://www.volcengine.com/docs/6662/1336108?lang=zh

This section of the startup script demonstrates how to configure the listening address (HOST) and port (PORT) from environment variables or command-line arguments. The HOST must be '0.0.0.0' for deployment.

```bash
# 配置监听地址和端口。
# 监听地址必须为 0.0.0.0，否则会导致部署失败。
HOST="0.0.0.0"
PORT=${_FAAS_RUNTIME_PORT}
TIMEOUT=${_FAAS_FUNC_TIMEOUT}

# 解析参数
while [[ $# -gt 0 ]]; do
case $1 in
--port) 
PORT="$2"
shift 2
;; 
--host) 
HOST="$2"
shift 2
;; 
*) 
shift
;; 
esac
done

```

--------------------------------

### Initialize Volcengine Provider for Sandbox

Source: https://www.volcengine.com/docs/6662/1851199?lang=zh

Demonstrates how to initialize the Volcengine provider using environment variables for credentials and region.

```python
from __future__ import print_function

import os
import sys

# Add the parent directory to Python path so we can import agent_sandbox
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from agent_sandbox.providers import VolcengineProvider


def main():
"""
Main function demonstrating Volcengine provider usage.
"""
# Configuration - replace with your actual credentials
access_key = os.getenv("VOLC_ACCESSKEY")
secret_key = os.getenv("VOLC_SECRETKEY")
region = os.getenv("VOLCENGINE_REGION", "cn-beijing")
```

--------------------------------

### Get Kafka Trigger Request Example

Source: https://www.volcengine.com/docs/6662/1451302?lang=zh

Use this POST request to retrieve the details of a specific Kafka trigger. Ensure you provide the correct `Id` and `FunctionId`.

```http
POST http://open.volcengineapi.com/?Action=GetKafkaTrigger&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "Id": "nw******",
    "FunctionId": "wrw******"
}

```

```json
{
    "Id": "nw******",
    "FunctionId": "wrw******"
}
```

--------------------------------

### Get Kafka Trigger Response Example

Source: https://www.volcengine.com/docs/6662/1451302?lang=zh

This is a successful HTTP 200 OK response for the GetKafkaTrigger API call, detailing the Kafka trigger's configuration and status.

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "GetKafkaTrigger",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "FunctionId": "wrw******",
        "Name": "reviews-consumer",
        "Description": "示例 kafka 触发器",
        "Enabled": true,
        "MqInstanceId": "kafka-cnngl4if******",
        "TopicName": "reviews-topic",
        "StartingPosition": "Latest",
        "MaximumRetryAttempts": 100,
        "Id": "18lt****",
        "ConsumerGroup": "reviews-topic_18lt****",
        "Status": "pending",
        "CreationTime": "2025-02-03 06:47:41.423691239 +0000 UTC",
        "LastUpdateTime": "2025-02-03 06:47:41.423691239 +0000 UTC"
    }
}

```

--------------------------------

### 使用 Python SDK 创建沙箱实例

Source: https://www.volcengine.com/docs/6662/1852301

通过 VolcengineProvider 初始化并调用 create_sandbox 方法创建沙箱实例，需配置有效的访问密钥和区域信息。

```python
from __future__ import print_function

import os
import sys

# Add the parent directory to Python path so we can import agent_sandbox
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from agent_sandbox.providers import VolcengineProvider


def main():
"""
Main function demonstrating Volcengine provider usage.
"""
# Configuration - replace with your actual credentials
access_key = os.getenv("VOLC_ACCESSKEY")
secret_key = os.getenv("VOLC_SECRETKEY")
region = os.getenv("VOLCENGINE_REGION", "cn-beijing")

# Initialize the Volcengine provider
provider = VolcengineProvider(
access_key=access_key,
secret_key=secret_key,
region=region
)

print("=== Volcengine Sandbox Provider Example ===\n")

function_id = "yato****"

print("1. Creating a sandbox...")
sandbox_id = provider.create_sandbox(function_id=function_id)
print(f"Create response: {sandbox_id}")


if __name__ == "__main__":
main()
```

--------------------------------

### Get Kafka Trigger Response Example

Source: https://www.volcengine.com/docs/6662/1451302

This is a successful response (HTTP 200 OK) for the GetKafkaTrigger API call. It includes metadata about the request and the detailed configuration of the Kafka trigger.

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "GetKafkaTrigger",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "FunctionId": "wrw******",
        "Name": "reviews-consumer",
        "Description": "示例 kafka 触发器",
        "Enabled": true,
        "MqInstanceId": "kafka-cnngl4if******",
        "TopicName": "reviews-topic",
        "StartingPosition": "Latest",
        "MaximumRetryAttempts": 100,
        "Id": "18lt****",
        "ConsumerGroup": "reviews-topic_18lt****",
        "Status": "pending",
        "CreationTime": "2025-02-03 06:47:41.423691239 +0000 UTC",
        "LastUpdateTime": "2025-02-03 06:47:41.423691239 +0000 UTC"
    }
}
```

--------------------------------

### FastAPI application example

Source: https://www.volcengine.com/docs/6662/1336108?lang=zh

A basic FastAPI application demonstrating GET and POST request handling for the root path. This code is intended for use within the Native Python runtime.

```python
from fastapi import FastAPI
import time

# 创建一个 FastAPI 应用实例
app = FastAPI()

# 定义根路径 '/' 的 GET 路由
@app.get('/')
def index_handler():
return 'Hello FastAPI From FaaS nativepython3 (and bye python2)!'

# 定义根路径 '/' 的 POST 路由
@app.post('/')
async def root_post_handler(payload: dict):
return {"message": "POST request received", "payload": payload}

```

--------------------------------

### Make run.sh executable

Source: https://www.volcengine.com/docs/6662/1336108?lang=zh

Ensure your startup script is executable before deployment. This command should be run locally.

```shell
# 注意是在本地执行该命令
# 假设启动脚本为 run.sh
chmod a+x run.sh

```

--------------------------------

### Get Kafka Trigger Request Example

Source: https://www.volcengine.com/docs/6662/1451302

Use this POST request to retrieve the details of a specific Kafka trigger by providing its ID and the associated Function ID. Ensure the Content-Type is set to application/json.

```http
POST http://open.volcengineapi.com/?Action=GetKafkaTrigger&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "Id": "nw******",
    "FunctionId": "wrw******"
}
```

--------------------------------

### Deploy Project to Volcengine Function Service

Source: https://www.volcengine.com/docs/6662/2206937?lang=zh

Deploy projects to Volcengine Function Service using `vefaas cli`. Consult `vefaas -h` and the quickstart guide for assistance. Ensure non-interactive mode is used for deployment to avoid terminal waiting.

```bash
vefaas -h
```

```bash
vefaas deploy -h
```

--------------------------------

### Create a Microservice Application Function with Python SDK

Source: https://www.volcengine.com/docs/6662/1386373

Demonstrates how to initialize the client and call the create_function method to deploy a microservice application using a function template.

```python
from __future__ import print_function​
import volcenginesdkvefaas​
import volcenginesdkcore​
from pprint import pprint​
from volcenginesdkcore.rest import ApiException​
​
region = "cn-beijing"​
access_key = "yourAccessKeyId"​
secret_key = "yourAccessKeySecret"​
​
​
def create_function_micro_service_code():​
configuration = volcenginesdkcore.Configuration()​
configuration.ak = access_key​
configuration.sk = secret_key​
configuration.region = region​
configuration.client_side_validation = True​
volcenginesdkcore.Configuration.set_default(configuration)​
​
client = volcenginesdkvefaas.VEFAASApi(volcenginesdkcore.ApiClient(configuration))​
​
try:​
resp = client.create_function(​
volcenginesdkvefaas.CreateFunctionRequest(​
name="micro-service-doctest",​
runtime="native/v1",​
vpc_config=volcenginesdkvefaas.VpcConfigForCreateFunctionInput(​
enable_vpc=True,​
vpc_id="vpc-id",​
subnet_ids=["subnet-id"],​
security_group_ids=["sg-id"]​
),​
cpu_strategy="always",​
command="./run.sh"​
)​
)​
pprint(resp)​
except ApiException as e:​
print("Exception when calling API: %s\n" % e)​
​
​
if __name__ == "__main__":​
create_function_micro_service_code()
```

--------------------------------

### Create Sandbox Instance using Volcengine Python SDK

Source: https://www.volcengine.com/docs/6662/1851199

This Python script demonstrates how to create a sandbox instance using the Volcengine provider from the All-in-One SDK. It requires setting environment variables for access key, secret key, and region. Replace 'yatoczqh' with your actual function ID.

```python
from __future__ import print_function

import os
import sys

# Add the parent directory to Python path so we can import agent_sandbox
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))

from agent_sandbox.providers import VolcengineProvider


def main():
"""
Main function demonstrating Volcengine provider usage.
"""
# Configuration - replace with your actual credentials
access_key = os.getenv("VOLC_ACCESSKEY")
secret_key = os.getenv("VOLC_SECRETKEY")
region = os.getenv("VOLCENGINE_REGION", "cn-beijing")

# Initialize the Volcengine provider
provider = VolcengineProvider(
access_key=access_key,
secret_key=secret_key,
region=region
)

print("=== Volcengine Sandbox Provider Example ===\n")

function_id = "yatoczqh"

print("1. Creating a sandbox...")
sandbox_id = provider.create_sandbox(function_id=function_id)
print(f"Create response: {sandbox_id}")


if __name__ == "__main__":
main()
```

--------------------------------

### GetRocketMQTrigger Request Example

Source: https://www.volcengine.com/docs/6662/1327337?lang=zh

Example of a POST request to retrieve RocketMQ trigger details.

```http
POST http://open.volcengineapi.com/?Action=GetRocketMQTrigger&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "Id": "kh******",
    "FunctionId": "aj******"
}
```

--------------------------------

### GetRocketMQTrigger Response Example

Source: https://www.volcengine.com/docs/6662/1327337?lang=zh

Example of a successful JSON response containing the RocketMQ trigger configuration.

```json
HTTP/1.1 200 OK
Content-Type: application/json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "GetRocketMQTrigger",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "FunctionId": "aj******",
        "Name": "rmq1",
        "Description": "轻量级官网",
        "Enabled": true,
        "MqInstanceId": "rocketmq-cnoe******",
        "TopicName": "test",
        "StartingPosition": "Latest",
        "MaximumRetryAttempts": 100,
        "Id": "kh******",
        "ConsumerGroup": "GID_test",
        "Status": "ready",
        "Orderly": true,
        "QPSLimit": 0,
        "CreationTime": "2024-02-13 02:36:23.033 +0000 UTC",
        "LastUpdateTime": "2024-02-13 02:36:23.033 +0000 UTC",
        "Endpoint": "http://rocketmq-cnoeaf70cd8b1234.rocketmq.ivolces.com:9876"
    }
}
```

--------------------------------

### 编写 Golang 编译脚本 build.sh

Source: https://www.volcengine.com/docs/6662/1206695?lang=zh

用于编译 Golang 代码的脚本，需确保二进制文件在 Linux 环境下运行。

```bash
#! /bin/bash
set -ex
cd `dirname $0`

go mod tidy

# Note: the binary should be compiled using linux env so as to run on FaaS Platform XD.
GOOS=linux GOARCH=amd64 go build -v -ldflags '-w -extldflags "-static"' -o main
```

--------------------------------

### Example Request Data for SandboxFusion

Source: https://www.volcengine.com/docs/6662/1585066

This is an example of the data payload structure for requesting code execution from SandboxFusion.

```json
"data": {
"code": "print(\"Hello, SandboxFusion!\")",
"language": "python"
}
```

--------------------------------

### Create Function Request Example

Source: https://www.volcengine.com/docs/6662/1262132

This JSON payload demonstrates how to configure a function, including source, environment variables, VPC, TLS, NAS storage, and TOS mounts.

```json
{
"Name": "official-site",
"Description": "Lightweight Official Website",
"Runtime": "python3.8/v1",
"ExclusiveMode": false,
"RequestTimeout": 30,
"MaxConcurrency": 100,
"MemoryMB": 1024,
"SourceType": "tos",
"Source": "pqyqo9qa/3kwem******.zip",
"Envs": [
{
"Key": "business_type",
"Value": "free"
}
],
"VpcConfig": {
"VpcId": "vpc-2d73mgm****",
"EnableVpc": true,
"SubnetIds": [
"subnet-mj1l******"
],
"SecurityGroupIds": [
"sg-3re8ffwjclybk5zsk****"
],
"EnableSharedInternetAccess": true
},
"TlsConfig": {
"EnableLog": true,
"TlsTopicId": "0cd4e4c4-5436-4b05-b39e-****",
"TlsProjectId": "48f49d74-cd50-4591-a3d0-****"
},
"SourceAccessConfig": {
"Password": "password ",
"Username": "operator"
},
"Tags": [
{
"Key": "business",
"Value": "ecommerce"
}
],
"NasStorage": {
"EnableNas": true,
"NasConfigs": [
{
"RemotePath": "/",
"FileSystemId": "enas-cns******",
"MountPointId": "mount-63******",
"LocalMountPath": "/mnt/nas"
}
]
},
"TosMountConfig": {
"EnableTos": false,
"Credentials": {
"AccessKeyId": "******",
"SecretAccessKey": "******"
},
"MountPoints": [
{
"Endpoint": "http://tos-cn-beijing.ivolces.com",
"ReadOnly": true,
"BucketName": "ai-model",
"BucketPath": "/",
"LocalMountPath": "/mnt/tos"
}
]
},
"InitializerSec": 30,
"Command": "./run.sh",
"Port": 8000,
"CpuStrategy": "always",
"ProjectName": "default",
"CpuMilli": 10,
"Cell": "2",
"Role": "trn:iam::140******:role/VeFaaSDefaultRole"
}
```

--------------------------------

### ListReleaseRecords Request Example

Source: https://www.volcengine.com/docs/6662/1327610?lang=zh

Example of a POST request to the ListReleaseRecords endpoint, including pagination and filtering parameters.

```json
POST http://open.volcengineapi.com/?Action=ListReleaseRecords&Version=2024-06-06 HTTP/1.1
Content-Type: application/json
{
    "FunctionId": "nuvpugx5",
    "PageNumber": 1,
    "PageSize": 10,
    "Filters": [
        {
            "Name": "Id",
            "Value": [
                "7d2******"
            ]
        }
    ],
    "OrderBy": {
        "Key": "CreationTime",
        "Ascend": false
    }
}
```

--------------------------------

### Create Sandbox Instance

Source: https://www.volcengine.com/docs/6662/1824706?lang=zh

Creates a sandbox instance with specified configurations.

```APIDOC
## POST /api/createSandbox

### Description
Creates a sandbox instance. This endpoint allows you to provision a new sandbox environment with various configuration options.

### Method
POST

### Endpoint
http://open.volcengineapi.com/?Action=CreateSandbox&Version=2024-06-06

### Parameters
#### Query Parameters
- **Action** (String) - Required - The name of the API action. Must be `CreateSandbox`.
- **Version** (String) - Required - The API version. Must be `2024-06-06`.

#### Request Body
- **FunctionId** (String) - Required - The ID of the sandbox application to which the sandbox instance belongs.
- **Timeout** (Integer) - Optional - The survival time of the sandbox instance in minutes. Defaults to 60. Range: 3-1440.
- **Metadata** (JSON Map) - Optional - Metadata for the sandbox instance, used for tagging and filtering. Format: `{"key":"value"}`.
- **InstanceTosMountConfig** (Object) - Optional - Configuration for mounting Object Storage (TOS) for the sandbox instance.
  - **Enable** (Boolean) - Required - Whether instance-level TOS mounting is enabled. `true` or `false`.
  - **TosMountPoints** (Array of Object) - Required if `Enable` is true. Specific TOS mount point information.
    - **BucketPath** (String) - Required - The remote directory in TOS to mount.
    - **LocalMountPath** (String) - Required - The local directory in the sandbox instance to mount the TOS bucket to.
- **Envs** (Array of Object) - Optional - Environment variables for the sandbox instance.
  - **Key** (String) - Required - The key of the environment variable.
  - **Value** (String) - Required - The value of the environment variable.
- **InstanceImageInfo** (Object) - Optional - Information about the sandbox instance image.
  - **Id** (String) - Optional - The ID of the pre-warmed image to use.
  - **Port** (Integer) - Required - The listening port of the sandbox instance image. Avoid ports 9000, 9001, 9990.
  - **Image** (String) - Optional - The address of the pre-warmed image to use.
  - **Command** (String) - Required - The command to start the sandbox instance program. Use absolute paths for scripts.
- **CpuMilli** (Integer) - Optional - The CPU specification for the sandbox instance in milli CPU. Defaults to 1000. Range: 250-16000.
- **MemoryMB** (Integer) - Optional - The memory specification for the sandbox instance in MiB. Defaults to 2048. Range: 512-131072.
- **MaxConcurrency** (Integer) - Optional - The maximum number of concurrent requests per instance. Defaults to 100. Range: 10-1000.
- **RequestTimeout** (Integer) - Optional - The request timeout in seconds. Defaults to 30. Range: 1-900.

### Request Example
```json
{
  "FunctionId": "wrws****",
  "Timeout": 60,
  "Metadata": {
    "app": "test"
  },
  "InstanceTosMountConfig": {
    "Enable": true,
    "TosMountPoints": [
      {
        "BucketPath": "/sandbox",
        "LocalMountPath": "/mnt/tos"
      }
    ]
  },
  "Envs": [
    {
      "Key": "business_type",
      "Value": "free"
    }
  ],
  "InstanceImageInfo": {
    "Id": "xGWv****",
    "Port": 8080,
    "Image": "cr-vefaas-cn-beijing.cr.volces.com/wdf/test/****:latest",
    "Command": "bash /root/sandbox/scripts/run.sh"
  },
  "CpuMilli": 500,
  "MemoryMB": 1024,
  "MaxConcurrency": 20,
  "RequestTimeout": 30
}
```

### Response
#### Success Response (200)
- **SandboxId** (String) - The ID of the created sandbox instance, which is also its name.

#### Response Example
```json
{
  "ResponseMetadata": {
    "RequestId": "20230604110420****100232280022D31",
    "Action": "CreateSandbox",
    "Version": "2024-06-06",
    "Service": "vefaas",
    "Region": "cn-beijing"
  },
  "Result": {
    "SandboxId": "vefaas-e8m6xarn-p2zamuv2fr-******"
  }
}
```

### Error Handling
Refer to Common Error Codes for troubleshooting API call failures.
```

--------------------------------

### 执行结果评测

Source: https://www.volcengine.com/docs/6662/2191799

使用 sb-cli 工具对推理结果进行评测。

```Bash
sb-cli verified test --predictions_path preds/preds.json --workers=5 --output_dir eval
```

--------------------------------

### GetFunctionInstanceLogs API Response Example

Source: https://www.volcengine.com/docs/6662/1327340?lang=zh

This is an example of a successful response from the GetFunctionInstanceLogs API. It includes ResponseMetadata and the Result containing the Logs.

```json
{
    "ResponseMetadata": {
        "RequestId": "20230604110420****100232280022D31",
        "Action": "GetFunctionInstanceLogs",
        "Version": "2024-06-06",
        "Service": "vefaas",
        "Region": "cn-beijing"
    },
    "Result": {
        "Logs": "2024-06-25T11:18:25.136415053+08:00 Model loaded in 11.1s (load weights from disk: 3.7s, create model: 0.6s, apply weights to model: 5.5s, load textual inversion embeddings: 0.2s, calculate empty prompt: 1.0s).\n"
    }
}
```

--------------------------------

### 执行 SWE-Bench 推理流程

Source: https://www.volcengine.com/docs/6662/2191799

使用 mini-extra 工具启动单次或批量推理任务。

```Bash
mini-extra swebench-single -c swebench_vefaas.yaml
```

```Bash
mini-extra swebench -c swebench_vefaas.yaml --subset=verified --split=test --shuffle --output=preds --workers=5
```