|
|
-
Roji Thomas Commented 9 Months ago through Blogs | 1 Point
Thanks.
Couple of points need clarification.
1. Using the sys.dm__tran__locks to inspect the lock requirements may not yield the desired result since the DMV only gives a point in time snapshot of the locks. To see and understand what locks are ...
|
-
Roji Thomas Commented 9 Months ago through Blogs | 1 Point
Below is the (incomplete) list of locks required to run a REORGANIZE on a Non Clustered index. Apart from the locks listed, there may be other METADATA lock requirement based on the security context of the executing user and features like Auditing and t...
|
-
Roji Thomas Commented 9 Months ago through Blogs | 1 Point
Thanks Madhi. Trying to be bit more active these days.
I'll also offer couple of solutions.
DECLARE @i INT = 987654321
DECLARE @c VARCHAR(250) = CONVERT(VARCHAR, @i)
SELECT @c = REPLACE(@c,n,REPLICATE('*',n))
FROM (SELECT TOP 10 ROW_NUMBER() ...
|
|
|
-
Roji Thomas Commented 10 Months ago through Blogs | 1 Point
I thought I should offer an explanation on how we can prove they are unrelated.
1. Run the trace and try to run a proc that does not exist. The RPC Starting and RPC completed Events will show the proc call, but it will fail with "cannot find procedur...
|
-
Roji Thomas Commented 10 Months ago through Blogs | 1 Point
Nakul,
Allow me to disagree. AFAIK, The correlation between the execultion plan (or instance specific execution context) and the profiler events are not correct. The RPC (Remote Procedure Call) events are traced from the higher stacks, Not the execut...
|
|
|
-
Roji Thomas Commented 10 Months ago through Blogs | 1 Point
Thanks. I was trying to validate what you are suggesting in the blog. So are you saying that this is just an issue with the client and it has nothing to with Profiler?...
|
-
Roji Thomas Commented 10 Months ago through Blogs | 1 Point
On what environments you have tested this? Was Foreced Parameterization in effect? From which client you invoked the queries?
In my tests in both 28R2 and 2012 I get the exact text in the "SQL:StmtStarting” and “SQL:StmtCompleted" events....
|
-
Roji Thomas Commented 10 Months ago through Blogs | 1 Point
Thanks for the clarifications Nakul and Good to see you here CELKO.
The point I was trying to drive home is that, in the context of datatype, the properties and behavior of TIMESTAMP and ROWVERSION are exactly same since they are just synonyms of the...
|