Docker部署本地DynamoDB报错：无法打开数据库文件

运行DynamoDB本地版容器并使用Docker卷持久化数据，创建第一个表时重复出现以下报错，最终脚本执行失败：

dynamodb-local           | Aug 31, 2022 3:32:49 PM com.almworks.sqlite4java.Internal log
dynamodb-local           | WARNING: [sqlite] SQLiteQueue[shared-local-instance.db]: stopped abnormally, reincarnating in 3000ms
dynamodb-local           | Aug 31, 2022 3:32:52 PM com.almworks.sqlite4java.Internal log
dynamodb-local           | WARNING: [sqlite] cannot open DB[5]: com.almworks.sqlite4java.SQLiteException: [14] unable to open database file
dynamodb-local           | Aug 31, 2022 3:32:52 PM com.almworks.sqlite4java.Internal log
dynamodb-local           | SEVERE: [sqlite] SQLiteQueue[shared-local-instance.db]: error running job queue
dynamodb-local           | com.almworks.sqlite4java.SQLiteException: [14] unable to open database file
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteConnection.open0(SQLiteConnection.java:1480)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteConnection.open(SQLiteConnection.java:282)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteConnection.open(SQLiteConnection.java:293)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteQueue.openConnection(SQLiteQueue.java:464)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteQueue.queueFunction(SQLiteQueue.java:641)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteQueue.runQueue(SQLiteQueue.java:623)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteQueue.access$000(SQLiteQueue.java:77)
dynamodb-local           |      at com.almworks.sqlite4java.SQLiteQueue$1.run(SQLiteQueue.java:205)
dynamodb-local           |      at java.lang.Thread.run(Thread.java:750)

使用的docker-compose.yml内容如下：

version: "3.9"  # optional since v1.27.0
services:
  server:
    build:
      context: ./server/
      dockerfile: Dockerfile.consumer
    ports:
      - "5558:5558"
    depends_on:
      - "dynamodb-local"
    environment:
      AWS_ACCESS_KEY_ID: 'DUMMYIDEXAMPLE'
      AWS_SECRET_ACCESS_KEY: 'DUMMYEXAMPLEKEY'

  dynamodb-local:
    image: "amazon/dynamodb-local:latest"
    ports:
      - "8000:8000"
    volumes:
      - "data:/home/dynamodblocal/data"
    command: "-jar DynamoDBLocal.jar -sharedDb -dbPath ./data"
    container_name: dynamodb-local
    working_dir: /home/dynamodblocal

volumes:
  logvolume01: {}
  data: {}

创建表的Python代码：

import boto3
from pprint import pprint
from botocore.exceptions import ClientError


TABLE_NAME = "check_tbl"
dynamodb = boto3.client(
            'dynamodb',
            region_name="us-east-1",
            aws_access_key_id="dummy_access_key",
            aws_secret_access_key="dummy_secret_key",
            endpoint_url="http://localhost:8000",
        )

try:
    response = dynamodb.describe_table(TableName=TABLE_NAME)

except ClientError as ce:
    if ce.response['Error']['Code'] == 'ResourceNotFoundException':
        result = dynamodb.create_table(
            TableName=TABLE_NAME,
                KeySchema=[
                    {
                        'AttributeName': 'id',
                        'KeyType': 'HASH'  # Partition key
                    }
                ],
                AttributeDefinitions=[
                    {
                        'AttributeName': 'timestamp',
                        'AttributeType': 'D'
                    }
                ],
                ProvisionedThroughput={
                    'ReadCapacityUnits': 10,
                    'WriteCapacityUnits': 10
                }
            )

    else:
        print("Unknown exception occurred while querying for the " + TABLE_NAME + " table. Printing full error:")
        pprint(ce.response)

注：Docker卷路径/var/lib/docker/volumes/data属主为root，但用户认为此非问题根源。

解决方案 #

1. 解决容器权限问题 #

DynamoDB本地版容器默认以非root用户运行（通常是nobody），而Docker匿名卷的默认属主为root，导致进程无法写入数据库文件。可通过以下两种方式解决：

方式一：临时切换为root用户运行容器

修改docker-compose.yml中dynamodb-local服务的配置，添加user: root：

dynamodb-local:
    image: "amazon/dynamodb-local:latest"
    ports:
      - "8000:8000"
    volumes:
      - "data:/home/dynamodblocal/data"
    command: "-jar DynamoDBLocal.jar -sharedDb -dbPath ./data"
    container_name: dynamodb-local
    working_dir: /home/dynamodblocal
    user: root  # 添加此行

方式二：使用本地目录绑定并设置权限

在主机创建本地目录，调整权限后绑定到容器：

# 创建本地目录
mkdir -p ./dynamodb-data
# 设置权限，允许容器进程写入
chmod 777 ./dynamodb-data

修改docker-compose.yml中的卷配置：

dynamodb-local:
    # 其他配置不变
    volumes:
      - ./dynamodb-data:/home/dynamodblocal/data  # 替换原匿名卷配置

2. 修正创建表的代码错误 #

当前代码中KeySchema指定了id作为分区键，但AttributeDefinitions中未声明该属性的类型，这会导致创建表请求直接失败。需补充id的属性定义：

result = dynamodb.create_table(
    TableName=TABLE_NAME,
    KeySchema=[
        {
            'AttributeName': 'id',
            'KeyType': 'HASH'  # Partition key
        }
    ],
    AttributeDefinitions=[
        {
            'AttributeName': 'id',
            'AttributeType': 'S'  # 根据实际业务需求调整类型，如S=字符串、N=数字等
        },
        {
            'AttributeName': 'timestamp',
            'AttributeType': 'D'
        }
    ],
    ProvisionedThroughput={
        'ReadCapacityUnits': 10,
        'WriteCapacityUnits': 10
    }
)

内容的提问来源于stack exchange，提问作者Brian