如何用SQL查询快速定位MASTER包中未完成的SSIS子包

• 在Visual Studio中，我们有一个包含约120个SSIS包的项目
• 这些包通过MASTER包启动：每个待运行的包对应一个Execute Package Task，其中仅70个包处于激活状态，任务相互独立可并行运行，由SQL Server代理调度
• MASTER包每日夜间自动运行
• 出错时可在SSMS中查看错误详情（多为网络类错误）
• 当前需要对照文档核对所有应运行的激活包，手动重新执行未运行的包

当MASTER包执行出错时，如何编写SQL查询快速找出其中未完成的包？

作业脚本 #

USE [msdb]
GO

/****** Object:  Job [SSIS_EXTRANET_DATAMART]    Script Date: 04/04/2023 09:17:34 ******/
BEGIN TRANSACTION
DECLARE @ReturnCode INT
SELECT @ReturnCode = 0
/****** Object:  JobCategory [[Uncategorized (Local)]]    Script Date: 04/04/2023 09:17:34 ******/
IF NOT EXISTS (SELECT name FROM msdb.dbo.syscategories WHERE name=N'[Uncategorized (Local)]' AND category_class=1)
BEGIN
EXEC @ReturnCode = msdb.dbo.sp_add_category @class=N'JOB', @type=N'LOCAL', @name=N'[Uncategorized (Local)]'
IF (@@ERROR <> 0 OR @ReturnCode <> 0) GOTO QuitWithRollback

END

DECLARE @jobId BINARY(16)
EXEC @ReturnCode =  msdb.dbo.sp_add_job @job_name=N'SSIS_EXTRANET_DATAMART', 
        @enabled=1, 
        @notify_level_eventlog=0, 
        @notify_level_email=2, 
        @notify_level_netsend=0, 
        @notify_level_page=0, 
        @delete_level=0, 
        @description=N'Pas de description disponible.', 
        @category_name=N'[Uncategorized (Local)]', 
        @owner_login_name=N'XXX\xxxxx', 
        @notify_email_operator_name=N'EmailBI', @job_id = @jobId OUTPUT
IF (@@ERROR <> 0 OR @ReturnCode <> 0) GOTO QuitWithRollback
/****** Object:  Step [DATAMART_V2]    Script Date: 04/04/2023 09:17:34 ******/
EXEC @ReturnCode = msdb.dbo.sp_add_jobstep @job_id=@jobId, @step_name=N'DATAMART_V2', 
        @step_id=1, 
        @cmdexec_success_code=0, 
        @on_success_action=3, 
        @on_success_step_id=0, 
        @on_fail_action=2, 
        @on_fail_step_id=0, 
        @retry_attempts=0, 
        @retry_interval=0, 
        @os_run_priority=0, @subsystem=N'SSIS', 
        @command=N'/ISSERVER &quot;\&quot;\SSISDB\EXTRANET\SSIS_EXTRANET\DATAMART_V2.dtsx\&quot;&quot; /SERVER &quot;\&quot;XXXXXXSSIS\SSIS\&quot;&quot; /ENVREFERENCE 29 /Par &quot;\&quot;$ServerOption::LOGGING_LEVEL(Int16)\&quot;&quot;;1 /Par &quot;\&quot;$ServerOption::SYNCHRONIZED(Boolean)\&quot;&quot;;True /CALLERINFO SQLAGENT /REPORTING E', 
        @database_name=N'master', 
        @flags=0, 
        @proxy_name=N'xxxxxxxx'
IF (@@ERROR <> 0 OR @ReturnCode <> 0) GOTO QuitWithRollback
/****** Object:  Step [Send Sucess Extranet]    Script Date: 04/04/2023 09:17:34 ******/
EXEC @ReturnCode = msdb.dbo.sp_add_jobstep @job_id=@jobId, @step_name=N'Send Sucess Extranet', 
        @step_id=2, 
        @cmdexec_success_code=0, 
        @on_success_action=1, 
        @on_success_step_id=0, 
        @on_fail_action=2, 
        @on_fail_step_id=0, 
        @retry_attempts=0, 
        @retry_interval=0, 
        @os_run_priority=0, @subsystem=N'TSQL', 
        @command=N'EXEC msdb.dbo.sp_send_dbmail  
@profile_name = ''P_BI'',  
@recipients = ''bi@xxxxx.fr'' ,
@body = ''Refresh Ok '',  
@subject = ''Refresh Ok''', 
        @database_name=N'master', 
        @flags=0
IF (@@ERROR <> 0 OR @ReturnCode <> 0) GOTO QuitWithRollback
EXEC @ReturnCode = msdb.dbo.sp_update_job @job_id = @jobId, @start_step_id = 1
IF (@@ERROR <> 0 OR @ReturnCode <> 0) GOTO QuitWithRollback
EXEC @ReturnCode = msdb.dbo.sp_add_jobserver @job_id = @jobId, @server_name = N'(local)'
IF (@@ERROR <> 0 OR @ReturnCode <> 0) GOTO QuitWithRollback
COMMIT TRANSACTION
GOTO EndSave
QuitWithRollback:
    IF (@@TRANCOUNT > 0) ROLLBACK TRANSACTION
EndSave:
GO

MASTER包结构 #

由于你的MASTER包部署在SSIS目录（SSISDB）中，可通过查询SSISDB系统视图快速定位未完成的子包。

核心查询：找出未完成的子包 #

USE SSISDB;
GO

-- 定位最近一次失败的MASTER包执行实例
DECLARE @MasterExecutionId BIGINT;
SELECT TOP 1 @MasterExecutionId = execution_id
FROM catalog.executions
WHERE package_name = 'DATAMART_V2.dtsx' -- MASTER包名称
    AND folder_name = 'EXTRANET' -- 目录名
    AND project_name = 'SSIS_EXTRANET' -- 项目名
    AND status = 3 -- 状态3代表执行失败
ORDER BY start_time DESC;

-- 查询该MASTER实例下未完成的子包
SELECT
    ex.package_name AS 子包名称,
    ex.start_time AS 开始时间,
    ex.end_time AS 结束时间,
    CASE ex.status
        WHEN 1 THEN '正在运行'
        WHEN 2 THEN '已成功'
        WHEN 3 THEN '已失败'
        WHEN 4 THEN '已取消'
        WHEN 5 THEN '已暂停'
        WHEN 6 THEN '正在挂起'
        ELSE '未知状态'
    END AS 执行状态
FROM catalog.executions ex
WHERE ex.parent_execution_id = @MasterExecutionId
    AND ex.status NOT IN (2) -- 排除已成功完成的包
ORDER BY ex.start_time;

查询说明 #

1. 定位MASTER包失败实例：通过catalog.executions视图筛选出指定MASTER包的最近一次失败记录，获取其execution_id作为父执行ID。
2. 筛选未完成子包：根据父执行ID关联所有子包的执行记录，排除状态为“已成功”（值为2）的包，剩下的即为未完成的子包（包括失败、运行中、取消等状态）。

扩展：查看子包错误详情 #

如果需要查看未完成子包的具体错误信息，可关联catalog.operation_messages视图：

USE SSISDB;
GO

DECLARE @MasterExecutionId BIGINT;
SELECT TOP 1 @MasterExecutionId = execution_id
FROM catalog.executions
WHERE package_name = 'DATAMART_V2.dtsx'
    AND folder_name = 'EXTRANET'
    AND project_name = 'SSIS_EXTRANET'
    AND status = 3
ORDER BY start_time DESC;

-- 查询子包的错误日志
SELECT
    ex.package_name AS 子包名称,
    om.message_time AS 错误时间,
    om.message AS 错误信息
FROM catalog.executions ex
JOIN catalog.operation_messages om
    ON ex.execution_id = om.operation_id
WHERE ex.parent_execution_id = @MasterExecutionId
    AND om.message_type = 120 -- 120代表错误类型消息
ORDER BY om.message_time;

注意事项 #

• 请确保查询中的package_name、folder_name、project_name与实际环境一致（可从作业脚本的SSIS命令参数中获取）。
• catalog.executions视图中status字段的含义：
  • 1: 正在运行
  • 2: 已成功
  • 3: 已失败
  • 4: 已取消
  • 5: 已暂停
  • 6: 正在挂起

内容的提问来源于stack exchange，提问作者Salah K.