How Sql server Deadlock Mechanism works

Deadlock Detection

Note: To enable deadlock , Please refer : How To Enable Deadlock Trace in Sql Server

 Deadlock detection is performed by a lock monitor thread that periodically initiates a search through all of the tasks in an instance of the Database Engine. The following points describe the search process:

· The default interval is 5 seconds.
· If the lock monitor thread finds deadlocks, the deadlock detection interval will drop from 5 seconds to as low as 100 milliseconds depending on the frequency of deadlocks.
· If the lock monitor thread stops finding deadlocks, the Database Engine increases the intervals between searches to 5 seconds.
· If a deadlock has just been detected, it is assumed that the next threads that must wait for a lock are entering the deadlock cycle. The first couple of lock waits after a deadlock has been detected will immediately trigger a deadlock search rather than wait for the next deadlock detection interval. For example, if the current interval is 5 seconds, and a deadlock was just detected, the next lock wait will kick off the deadlock detector immediately. If this lock wait is part of a deadlock, it will be detected right away rather than during next deadlock search.
The Database Engine typically performs periodic deadlock detection only. Because the number of deadlocks encountered in the system is usually small, periodic deadlock detection helps to reduce the overhead of deadlock detection in the system.
When the lock monitor initiates deadlock search for a particular thread, it identifies the resource on which the thread is waiting. The lock monitor then finds the owner(s) for that particular resource and recursively continues the deadlock search for those threads until it finds a cycle. A cycle identified in this manner forms a deadlock.
After a deadlock is detected, the Database Engine ends a deadlock by choosing one of the threads as a deadlock victim. The Database Engine terminates the current batch being executed for the thread, rolls back the transaction of the deadlock victim, and returns a 1205 error to the application. Rolling back the transaction for the deadlock victim releases all locks held by the transaction. This allows the transactions of the other threads to become unblocked and continue. The 1205 deadlock victim error records information about the threads and resources involved in a deadlock in the error log.
By default, the Database Engine chooses as the deadlock victim the session running the transaction that is least expensive to roll back. Alternatively, a user can specify the priority of sessions in a deadlock situation using the SET DEADLOCK_PRIORITY statement. DEADLOCK_PRIORITY can be set to LOW, NORMAL, or HIGH, or alternatively can be set to any integer value in the range (-10 to 10). The deadlock priority defaults to NORMAL. If two sessions have different deadlock priorities, the session with the lower priority is chosen as the deadlock victim. If both sessions have the same deadlock priority, the session with the transaction that is least expensive to roll back is chosen. If sessions involved in the deadlock cycle have the same deadlock priority and the same cost, a victim is chosen randomly.
When working with CLR, the deadlock monitor automatically detects deadlock for synchronization resources (monitors, reader/writer lock and thread join) accessed inside managed procedures. However, the deadlock is resolved by throwing an exception in the procedure that was selected to be the deadlock victim. It is important to understand that the exception does not automatically release resources currently owned by the victim; the resources must be explicitly released. Consistent with exception behavior, the exception used to identify a deadlock victim can be caught and dismissed.
ms-help://MS.SQLCC.v10/MS.SQLSVR.v10.en/s10de_1devconc/local/collapse.gifDeadlock Information Tools
To view deadlock information, the Database Engine provides monitoring tools in the form of two trace flags, and the deadlock graph event in SQL Server Profiler.
Trace Flag 1204 and Trace Flag 1222
When deadlocks occur, trace flag 1204 and trace flag 1222 return information that is captured in the SQL Server 2005 error log. 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. It is possible to enable both trace flags to obtain two representations of the same deadlock event.
 Note: To enable deadlock , Please refer : How To Enable Deadlock Trace in Sql Server

No comments:

Post a Comment