什么是死锁?
==============
A deadlock occurs when two system server process IDs (SPIDs) are waiting for a resource and neither process can advance because the other process is preventing it from getting the resource. For example, Andrew holds a lock on table A and requests a lock on table B; Lindsay holds a lock on table B and requests a lock on table A.
什么东西可以引发死锁?
==============
- Locks
- Worker threads
- Memory
- Parallel query execution-related resources
- Multiple Active Result Sets (MARS) resources.
如何排查死锁?
==============
1. 开启死锁日志输出(deadlock trace)
DBCC TRACEON(1204,1222)
Trace flag 1204 reports deadlock information formatted by each node involved in the deadlock.
Trace flag 1222 formats deadlock information, first by processes and then by resources.
开启了上面的选项之后, SQL会输出死锁的细节信息到SQL Error Log中(默认位置Program Files\Microsoft SQL Server\MSSQL.n\MSSQL\LOG\ERRORLOG and ERRORLOG.n)
2. 开启SQL Profiler.
- Start SQL profiler
- On the Trace Properties dialog box, on the General tab, check Save to file and specify a path to save the trace
- Click the Events tab, only add Locks\Lock:deadlock and Locks\Lock:deadlock chain
- Click the Data columns tab, add DatabaseID, IndexID, ObjectID
可以通过下面的语句把DatabaseID和ObjectID换成DatabaseName和ObjectName
SELECT database_name(DatabaseID)
SELECT object_name(ObjectID)
3. 使用下面的查询语句来检查那个进程被锁住了.
SELECT * FROM sys.sysprocesses WHERE blocked <> 0
从Blocked列中得到SPID
DBCC inputbuffer (SPID)
sp_who2
sp_lock2
参考资料:
SQL: How to debug SQL deadlocks
http://www.dalun.com/blogs/10.13.2006.htm
SQL Server technical bulletin - How to resolve a deadlock
http://support.microsoft.com/?kbid=832524
Detecting and Ending Deadlocks