DeadLock

  • We are facing deadlock issue ,now a days more frequently . I have enclosed the deadlock graph.

    We have  indexes  on the  table  and the query seems to be  the same update from different servers (not sure though) . Can anyone help me resolve this?

    Thanks In Advance

  • You need to get to the code before and after the listings that are in the input buffer of the deadlock graph you have posted. One of the queries is incomplete in the buffer, so there's no way to make any suggestions. If you can post both sets of processing, it will make it much easier to make suggestions as to what's wrong. It could be the classic of referencing tables in different orders within a batch. This is generally difficult to troubleshoot using only the deadlock graph. The deadlock graph is a pointer to where the problem is, not the actual problem.

    As to whether or not your indexes are in place, you'll want to look to the execution plans to understand how the indexes you have are being used by these queries. Fundamentally a deadlock is a performance problem. Make the queries fast enough and deadlocks go away or are radically reduced.

    "The credit belongs to the man who is actually in the arena, whose face is marred by dust and sweat and blood"
    - Theodore Roosevelt

    Author of:
    SQL Server Execution Plans
    SQL Server Query Performance Tuning

  • Grant Fritchey - Tuesday, February 7, 2017 6:11 AM

    You need to get to the code before and after the listings that are in the input buffer of the deadlock graph you have posted. One of the queries is incomplete in the buffer, so there's no way to make any suggestions. If you can post both sets of processing, it will make it much easier to make suggestions as to what's wrong. It could be the classic of referencing tables in different orders within a batch. This is generally difficult to troubleshoot using only the deadlock graph. The deadlock graph is a pointer to where the problem is, not the actual problem.

    As to whether or not your indexes are in place, you'll want to look to the execution plans to understand how the indexes you have are being used by these queries. Fundamentally a deadlock is a performance problem. Make the queries fast enough and deadlocks go away or are radically reduced.

    Thanks Grant , Took the queries from buffer and created indexes based on those which helped to eliminate DL

Viewing 3 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic. Login to reply