Serverless Offline使用EventBridge Scheduler报错，咨询是否支持该服务

核心结论 #

EventBridge Scheduler 目前未被Serverless Offline官方支持，这是你遇到ValidationException: The execution role you provide must allow AWS EventBridge Scheduler to assume the role错误的根本原因。

错误原因 #

Serverless Offline仅能模拟Lambda、API Gateway等基础服务的核心逻辑，并未实现EventBridge Scheduler的完整模拟：

• 本地环境不会创建你在resources.ts中定义的IAM角色，也无法验证该角色的信任策略是否允许scheduler.amazonaws.com执行sts:AssumeRole
• 当你的Lambda代码调用CreateScheduleCommand时，本地模拟环境无法处理真实的IAM权限校验逻辑，直接抛出权限错误

本地开发可行方案 #

1. Mock Scheduler API 调用（推荐） #

通过环境变量判断是否处于Offline模式，跳过真实的Scheduler调用，直接触发目标Lambda的业务逻辑：

// 修改你的调度器代码
const isOffline = process.env.IS_OFFLINE === 'true';

if (isOffline) {
  // 本地直接调用sendGreeting的处理函数
  const sendGreetingHandler = require('./path-to-sendGreeting').handler;
  await sendGreetingHandler({ /* 传入你的Input参数 */ }, {});
} else {
  // 线上环境正常调用Scheduler
  const schedulerClient = new SchedulerClient();
  const params = {
    // 你的原有参数
  };
  await schedulerClient.send(new CreateScheduleCommand(params));
}

2. 使用第三方离线模拟插件 #

可以尝试第三方社区开发的Serverless插件（如serverless-offline-scheduler），这类插件会补充EventBridge Scheduler的本地模拟能力，但需要注意版本兼容性，部分插件可能只支持基础的定时触发逻辑。

3. 临时绕过权限校验（仅本地测试用） #

在本地环境下，修改CreateScheduleCommand的参数，去掉RoleArn字段或使用占位值，同时配合Mock工具（如jest）拦截Scheduler的API调用，避免触发权限校验。

附你的配置与代码：

resources.ts 配置

export const resources = {
  Resources: {
    ....
    SchedulerExecutionRole: {
      Type: 'AWS::IAM::Role',
      Properties: {
        RoleName: '${self:service}-scheduler-execution-role-${self:provider.stage}',
        AssumeRolePolicyDocument: {
          Version: '2012-10-17',
          Statement: [
            {
              Effect: 'Allow',
              Principal: {
                Service: 'scheduler.amazonaws.com',
              },
              Action: 'sts:AssumeRole',
            },
          ],
        },
        Policies: [
          {
            PolicyName: '${self:service}-scheduler-lambda-policy-${self:provider.stage}',
            PolicyDocument: {
              Version: '2012-10-17',
              Statement: [
                {
                  Effect: 'Allow',
                  Action: [
                    'lambda:InvokeFunction',
                  ],
                  Resource: [
                    {
                      'Fn::GetAtt': ['sendGreeting', 'Arn'],
                    }
                  ],
                },
              ],
            },
          },
          {
            PolicyName: '${self:service}-scheduler-management-policy-${self:provider.stage}',
            PolicyDocument: {
              Version: '2012-10-17',
              Statement: [
                {
                  Effect: 'Allow',
                  Action: [
                    'scheduler:CreateSchedule',
                    'scheduler:DeleteSchedule',
                    'scheduler:GetSchedule',
                    'scheduler:UpdateSchedule',
                  ],
                  Resource: [
                    'arn:aws:scheduler:${self:provider.region}:${self:custom.AWS_ACCOUNT_ID}:schedule/${self:service}-*',
                  ],
                },
              ],
            },
          },
        ],
      },
    },
    SendGreetingLambdaPermission: {
        Type: 'AWS::Lambda::Permission',
        Properties: {
        FunctionName: {
            'Fn::GetAtt': ['sendGreeting', 'Arn'],
        },
        StatementId: 'EventBridgeSchedulerInvokePermission',
        Action: 'lambda:InvokeFunction',
        Principal: 'scheduler.amazonaws.com',
        SourceArn: 'arn:aws:scheduler:${self:provider.region}:${self:custom.AWS_ACCOUNT_ID}:schedule/${self:service}-*',
      },
    }
  }
};

调度器Lambda代码

const schedulerClient = new SchedulerClient();
const params = {
  Name: scheduleName,
  ScheduleExpression: `at(${schedule.scheduleTime})`,
  FlexibleTimeWindow: {
    Mode: 'OFF' as const,
  },
  Target: {
    Arn: `arn:aws:lambda:${process.env.AWS_REGION}:${process.env.AWS_ACCOUNT_ID}:function:${process.env.SERVICE_NAME}-${process.env.STAGE}-sendGreeting`,
    RoleArn: `arn:aws:iam::${process.env.AWS_ACCOUNT_ID}:role/${process.env.SERVICE_NAME}-scheduler-execution-role-${process.env.STAGE}`,
    Input: JSON.stringify({
      // params here
    }),
  },
};

console.log('Creating scheduler schedule:', params);

await schedulerClient.send(new CreateScheduleCommand(params));

内容的提问来源于stack exchange，提问作者EJCris