May 23, 2008 at 9:32 am
If we have a Dead lock situation,we can check in the error log and find out which spids are involved in the process.After that the next step is to kill one of the process by using KILL (spid)
In this situation i have a small doubt can we able to directly issue a kill command.Is this approach correct one or any other solution is there.
And i want to know can we have to consider anything before issuing a
kill command.
Is this is the only solution or is there any approach.
And i have one more doubt , if we kill a process then what about the data that the transaction is processing
Plz guide me.
May 23, 2008 at 9:46 am
There are a couple of things I would do before killing a process.
1. Run sp_lock [spid] for each spid to see what is object(s) are causing the deadlock and the types of locks. You will see a wait for the blocked spid.
2. Run DBCC InputBuffer([spid]) for each spid involved to find the statements that are causing the deadlock.
Once you find this you can then determine if indexing may help (possibly reduce the type of lock from table to page/row), refactor the code in the offending statements to access tables in the same order, and determine there are long-running transactions that can be broken down in to smaller transactions.
Jack Corbett
Consultant - Straight Path Solutions
Check out these links on how to get faster and more accurate answers:
Forum Etiquette: How to post data/code on a forum to get the best help
Need an Answer? Actually, No ... You Need a Question
May 23, 2008 at 3:05 pm
In case you have recurring deadlock issues then go thru the link mentioned below and you really dont knwo how to troubleshoot them:
http://blogs.msdn.com/bartd/archive/2006/09/09/Deadlock-Troubleshooting_2C00_-Part-1.aspx
Manu Jaidka
May 24, 2008 at 2:22 pm
If you have a true deadlock, then you won't need to kill processes. SQL has a deadlock detector that will notice a deadlock condition, pick one of the participants and kill it automatically.
Do you have true deadlocks (if so, the users will be getting occasional errors about processes been picked as the deadlock victim and rolled back) or do you just have long-term blocking?
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
May 24, 2008 at 2:35 pm
if u killed a process its statment will be rolled back, so u dont have to wory that much.
but if the code have "begin transaction and commit transaction" if u killed the session after the commit then the roll back wont effect this batch.
..>>..
MobashA
May 27, 2008 at 6:43 am
You dont need to kill the process because SQL Server automatically kills the process that is creating that deadlock by using its internal deadlock manager.
Basit Ali Farooq
MCITP Database Administrator
Microsoft Certified Professional Developer (Web Applications)
Microsoft Certified Database Administrator
Microsoft Certified Systems Engineer
Microsoft Certified Systems Administrator
CIW Security Analyst
Cisco Certified Network Associate
May 28, 2008 at 12:18 am
To avoid deadlock, you must have through knowledge of your application.
1. Identify the deadlocking processes and the object sparticipating in deadlock.
2. Use program monitoring table, say Program_session having columns program_id and running_date and modify your SPs/or queries such that, it will first check entry in Program_session and verify if running_date is NULL.
3. If running_date is NULL, then SP can continue with setting running_date to getdate(). And after executing set running_date to NULL.
4. Else use wait() command to follow steps 3.
Viewing 7 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply