Forum Replies Created

Viewing 15 posts - 2,161 through 2,175 (of 8,416 total)

  • RE: Is this a Single Resource Deadlock?

    Alex Webber (2/24/2011)

    I am one of the developers 😉 (accidental DBA scenario). And yes it's a consequence of the data access tech, more specifically the ORM tool (LINQ-to-SQL). SERIALIZABLE isn't...

    February 24, 2011 at 5:02 am

    #1290872

  • RE: Locking Question: sys.dm_tran_locks

    glennhill (2/24/2011)

    I do have a snapshot of dm_os_waiting_tasks at the time of the problem. The wait_type is LOCK for all 21 rows (LOCK_M_S, LOCK_M_U or LOCK_M_X). The wait durations...

    February 24, 2011 at 4:50 am

    #1290869

  • RE: Paramaterized query

    onixsoft (2/24/2011)

    Thanks Paul, I was able to recreate table structure and did some more research and found what following line turn autoparameterization off :

    CREATE NONCLUSTERED INDEX [IDX_SSNHistory_IndID] ON [dbo].[SSNHistory] ([IndID])...

    February 24, 2011 at 4:37 am

    #1290862

  • RE: Is this a Single Resource Deadlock?

    Alex Webber (2/24/2011)

    However what I don't get is why range locks are coming into play at all when it's an IDENTITY based table and the two queries in question are...

    February 24, 2011 at 4:29 am

    #1290858

  • RE: Locking Question: sys.dm_tran_locks

    glennhill (2/24/2011)

    Can you offer any other advice on how we should debug this problem next time it occurs?

    I would look not just at sys.dm_tran_locks, but also at sys.dm_os_waiting_tasks. I...

    February 24, 2011 at 4:07 am

    #1290842

  • RE: Locking Question: sys.dm_tran_locks

    glennhill (2/24/2011)

    I do have the full output from dm_tran_locks. But my suspicion is that the 'hobt_id's have changed in sys.partitions due to our index rebuild.

    Yes, I read and understood that....

    February 24, 2011 at 4:01 am

    #1290840

  • RE: Is this a Single Resource Deadlock?

    Alex Webber (2/24/2011)

    Thanks for the replies SQLKiwi. The table in question is clustered/pk'ed on an identity column and thus there is no "SELECT to see if exists and then INSERT"...

    February 24, 2011 at 3:46 am

    #1290833

  • RE: Locking Question: sys.dm_tran_locks

    glennhill (2/24/2011)

    Not easy to map back this back to object names. The 'resource_associated_entity_id's no longer exist (as hobt_id in sys.partitions).

    Ok, no problem, but the full output from the locks DMV...

    February 24, 2011 at 3:40 am

    #1290828

  • RE: Is this a Single Resource Deadlock?

    Alex Webber (2/24/2011)

    The INSERT and SELECT are from 1 table with 1 index (the clustered one), so how does the UPDLOCK come into play here?

    Typically, the broken pattern is to...

    February 24, 2011 at 3:21 am

    #1290819

  • RE: Grant permission to create table in SP

    atikdesai (2/23/2011)

    Hi everyone,

    There is one user who has permission to execute one SP.

    In that SP, one table is created in database and finally it is dropped at the end of...

    February 24, 2011 at 3:13 am

    #1290818

  • RE: Is this a Single Resource Deadlock?

    That looks like an ordinary conversion deadlock to me, which can be fixed by accessing objects in a consistent order or by using the UPDLOCK hint appropriately. You will...

    February 24, 2011 at 3:08 am

    #1290813

  • RE: Paramaterized query

    onixsoft (2/24/2011)

    No, by run in the Management Studio I do not get auto-parameterized plan also.

    The following script gives me an auto-parameterized query in both cases:

    CREATE TABLE dbo.SSNhistory

    ...

    February 24, 2011 at 3:03 am

    #1290809

  • RE: Paramaterized query

    onixsoft (2/24/2011)

    So no way to fix it with less blood?

    If you run the IndId= GUID query from Management Studio, do you get an auto-parameterized plan? I would expect that...

    February 24, 2011 at 2:34 am

    #1290796

  • RE: Locking Question: sys.dm_tran_locks

    glennhill (2/24/2011)

    The head of the blocking chain is not waiting on any resource - i.e. waitresource in sysprocesses is blank. The process is sleeping, awaiting command.

    You said the 21 rows...

    February 24, 2011 at 2:30 am

    #1290795

  • RE: Weird exception raised by server

    dant12 (2/23/2011)

    qxcntxt.cpp:956 "No exceptions should be raised by this code"

    As far as I can tell, this bug was fixed in SP2 for SQL Server 2005. Which version are you...

    February 24, 2011 at 2:26 am

    #1290793

Viewing 15 posts - 2,161 through 2,175 (of 8,416 total)