How to Kill All Processes That Have Open Connection in a SQL Server Database[关闭数据库链接 最佳方法] -摘自网络

SQL Server database administrators may frequently need in especially development and test environments  instead of the production environments to kill all the open connections to a  specific database in order to process SQL Server maintenance task over the SQL Server database.

In such situations when you need to kill or close all the active or open  connections to the SQL Server database, you may manage this task by using the  Microsoft SQL Server Management Studio or by running t-sql commands or codes. Actually, this task can be thought as a batch task to kill sql process running on a SQL Server.

If you open the SQL Server Management Studio and connect to a SQL Server  instance you will see the Activity Monitor object in the Object Explorer screen  of the related database instance. You can double click the Activity Monitor  object or right click to view the context menu and then select a desired item to  display the activities to be monitored on the Activity Monitor screen.

As seen on below you can monitor and view process id‘s and process details on  the list of prcesses running on the database instance. If you want you can  filter processes based on specific values like user, database or status.

Note that default view when displayed the screen is first opened is filtered  only for non-system processes which means system processes which own the first  50 reserved processid‘s are not listed in the view by default. You can view  system processes by removing the filter on "Show System Processes" criteria in  the filter settings screen.

SQL Server 2005 SQL Server Management Studio Activity Monitor screen

You can kill a process by a right click on the process in the grid and  selecting the Kill Process menu item. You will be asked for a confirmation to  kill the related process and then will kill the open connection to the database  over this process. This action is just like running to kill sql process t-sql command for a single process.

A second method which I do not recommend but can be used in some situations  may be using the Detach Database screen to drop connections and detaching the  database and then re-attaching the database. You can open the Detach Database  screen from the context menu displayed by a right click on the related daabase  for example for the below screen shot the name of the database is Works. On the  menu, highlight menu item Tasks then select the Detach... menu item. This  selection will open the detach database dialog screen. Note that if in the  message column it is declared that active connections exists as for our case the  number of active connections is 2, you will not be able to detach the database  unless the Drop Connections checkbox is also selected.

The above configuration as the Drop Connections check box is cleared and  active connections exist, the detach task will fail:

Detach database failed for Server ‘{DatabaseInstanceName}‘. (Microsoft.SqlServer.Smo)
For help, click: http://go.microsoft.com/fwlink?ProdName=Microsoft+SQL+Server& ProdVer=9.00.2047.00& EvtSrc=Microsoft.SqlServer.Management.Smo.ExceptionTemplates. FailedOperationExceptionText& EvtID=Detach+database+Server& LinkId=20476
An exception occurred while executing a Transact-SQL statement or batch.  (Microsoft.SqlServer.ConnectionInfo)
Cannot detach the database ‘{DatabaseName}‘ because it is currently in use.  (Microsoft SQL Server, Error: 3703)

But if the Drop Connections is selected you can successfully detach the  database. Then you will have to re-attach the database by selecting the Attach  command from the context menu item displayed on the Databases node of the SQL  Server instance.

For SQL Server 2000 the default behaviour was different than the SQL Server  2005. Because in SQL Server 2000, when you run the detach command from the menu  item, you are prompted if you want to drop all active connections. Then you can  confirm closing of all open connections, but the nice thing is that you can  cancel detach process after the open connections are dropped or closed. But for  SQL Server 2005, this behaviour is not valid.

How to Kill All Processes using T-SQL Code

By using t-sql commands or sql codes, similarly closing connections can be implemented by a few methods. One of the methods is using a cursor which loops for all the active connections of the related database and kill these processes.  This method was also mention on SQL Server article named How to Alter a SQL Server Database as Single User Mode and as Multi User Mode

The below code block can be used to kill all processes which are connected to the sql database named @DatabaseName except the process that the code block is running in the scope of. You can also set the SQL Server database name by the DB_NAME() property.

The base of the below t-sql script is sql Kill SPId command. Within the sql cursor which loops through each process in sysProcesses view, each time a tsql Kill process command is called passing the SPId as an argument. And after Kill SPId sql statement is executed for each process, all database connections are dropped.

SQL Server database administrators can use below t-sql script in order to drop connections SQL Server 2008, or SQL Server 2005 and also drop connections to sql server 2000 database.

DECLARE @DatabaseName nvarchar(50) DECLARE @SPId int DECLARE @SQL nvarchar(100)
--SET @DatabaseName = N‘AdventureWorks2008‘ SET @DatabaseName = DB_NAME() DECLARE my_cursor CURSOR FAST_FORWARD FOR SELECT SPId FROM MASTER..SysProcesses WHERE DBId = DB_ID(@DatabaseName) AND SPId <> @@SPId
OPEN my_cursor
FETCH NEXT FROM my_cursor INTO @SPId
WHILE @@FETCH_STATUS = 0 BEGIN  SET @SQL = ‘KILL ‘ + CAST(@SPId as nvarchar(10))  print @SQL  EXEC sp_executeSQL @SQL  --KILL @SPId -- Causing Incorrect syntax near ‘@spid‘.
 FETCH NEXT FROM my_cursor INTO @SPId END
CLOSE my_cursor DEALLOCATE my_cursor

UPDATE : Please note that I altered above t-sql script and removed KILL @SPId sql statement. It was causing the below error. Msg 102, Level 15, State 1, Line 19 Incorrect syntax near ‘@SPId‘. This is because the sql processid can not be used using a variable with sql KILL command. The solution is using dynamic t-sql statement as shown in the above sql cursor code.

A second way to drop all active connections of a database can be implemented by generating dynamic sql commands that runs a list of "Kill @spId" commands.

DECLARE @DatabaseName nvarchar(50) SET @DatabaseName = N‘Works‘ --SET @DatabaseName = DB_NAME()
DECLARE @SQL varchar(max) SET @SQL = ‘‘
SELECT @SQL = @SQL + ‘Kill ‘ + Convert(varchar, SPId) + ‘;‘ FROM MASTER..SysProcesses WHERE DBId = DB_ID(@DatabaseName) AND SPId <> @@SPId
-- SELECT @SQL EXEC(@SQL)

A very similar to the sql code above, an other code block can be used by using the COALESCE as shown below

DECLARE @DatabaseName nvarchar(50) SET @DatabaseName = N‘Works‘
DECLARE @SQL varchar(max)
SELECT @SQL = COALESCE(@SQL,‘‘) + ‘Kill ‘ + Convert(varchar, SPId) + ‘;‘ FROM MASTER..SysProcesses WHERE DBId = DB_ID(@DatabaseName) AND SPId <> @@SPId
--SELECT @SQL EXEC(@SQL)

The above sql queries can be modified further for specific needs. For example you may create a sql stored procedure that drops all existing active connections. You may pass SQL database name or database id as parameter or use the current database information to kill processes except its own process, etc.

时间: 2024-10-29 16:16:59

How to Kill All Processes That Have Open Connection in a SQL Server Database[关闭数据库链接 最佳方法] -摘自网络的相关文章

(转)How To Kill runaway processes After Terminating Concurrent Request

Every concurrent Request uses some resources for running. If we find that the concurrent request is taking long time and decided to terminate the concurrent request , the resources may not be released soon. These processes are called runaway processe

SQL Server会话KILL不掉,一直处于KILLED /ROLLBACK状态情形浅析[转]

本文将为您描述SQL Server会话KILL不掉,一直处于KILLED /ROLLBACK状态情形浅析,教程操作方法: 今天遇到一个很奇怪的情况,发现一个会话异常,这个会话只是在执行一个简单的存储过程,里面使用了链接服务器(Linked Server)查询另外一台服务器数据(存储过程里面没有任何显性事务.UPDATE.DELETE操作,只有几个简单的SELECT查询,其中有两个查询使用了链接服务器Linked Server,由于生产环境,不好贴出SQL语句),在DPA监控工具里面,发现该会话引

SQL 2012群集添加节点失败“Please wait while Microsoft SQL Server 2012 Service Pack 1 Setup processes

问题描述(Issue Symptoms) SQL Server 2012 STD cluster安装在Windows Server 2012时,添加节点时,在以下界面超过4小时无法通过: "Please wait while Microsoft SQL Server 2012 Service Pack 1 Setup processes the current operation." 第二天存在以下界面,但是无法选择已经安装的群集 原因分析(Cause) 1. 查看了安装日志发现 De

sql server block如何查询并kill

本帖提供两种做法,可避免在 SQL Server 事务锁定时产生的不正常或长时间阻塞,让用户和程序也无限期等待,甚至引起 connection pooling 连接数超过容量. 所谓的「阻塞」,是指当一个数据库会话中的事务,正在锁定其他会话事务想要读取或修改的资源,造成这些会话发出的请求进入等待的状态.SQL Server 默认会让被阻塞的请求无限期地一直等待,直到原来的事务释放相关的锁,或直到它超时 (根据 SET LOCK_TIMEOUT,本文后续会提到).服务器关闭.进程被杀死.一般的系统

SQL Server会话KILL不掉,一直处于KILLED /ROLLBACK状态情形浅析

原文:SQL Server会话KILL不掉,一直处于KILLED /ROLLBACK状态情形浅析 今天遇到一个很奇怪的情况,发现一个会话异常,这个会话只是在执行一个简单的存储过程,里面使用了链接服务器(Linked Server)查询另外一台服务器数据(存储过程里面没有任何显性事务.UPDATE.DELETE操作,只有几个简单的SELECT查询,其中有两个查询使用了链接服务器Linked Server,由于生产环境,不好贴出SQL语句),在DPA监控工具里面,发现该会话引起了非常长的OLEDB等

sql server 查询和Kill死锁进程

查询死锁进程语句 select        request_session_id spid,       OBJECT_NAME(resource_associated_entity_id) tableName    from        sys.dm_tran_locks   where        resource_type='OBJECT' 杀死死锁进程语句 kill spid

sql server 查看表的死锁和Kill 死锁进程

查询出来 select        request_session_id spid,       OBJECT_NAME(resource_associated_entity_id) tableName    from        sys.dm_tran_locks   where        resource_type='OBJECT 杀死死锁进程 kill spid 另: exec master.dbo.sp_who_lock --查看当前死锁进程 exec master.dbo.p_

处理sql server的死锁 [kill spid]

出现死锁的解决办法 在master中创建查看死锁的存储过程 [sql] view plaincopy use master go create procedure sp_who_lock as begin declare @spid int,@bl int, @intTransactionCountOnEntry  int, @intRowcount    int, @intCountProperties   int, @intCounter    int create table #tmp_l

慎用kill -9,kill -15的作用

详见:http://blog.yemou.net/article/query/info/tytfjhfascvhzxcyt334 Perl语言专家Randal Schwartz在一篇短文里这样写: no no no.不要使用kill -9. 它没有给进程留下善后的机会: 1) 关闭socket链接 2) 清理临时文件 3) 将自己将要被销毁的消息通知给子进程 4) 重置自己的终止状态 等等. 通常,应该发送15,等一两秒钟,如果没效果,发送2,如果还不行,发送1.如果还不行,那你应该把那个程序删