I tried capturing a replayable trace on a SQL Server 2012 production environment and then replay it against a SQL 2016 test environment.
After doing so , I got the following results :
Can anyone help me identify what each of the below highlighted term represents :
- Total events = 89
- Replayed events = 39
- Provider errors = 13
- Internal errors = 26
- Total errors = 39
- Provider hit ratio = 66.67 %
- Internal hit ratio = 33.33 %
- Overall hit ratio = 0.00 %
- Replay time = 0 hr 2 min 12 sec 38 ms
Will be of great help as I am not able to understand how the upgrade to 2016 might perform based on the profiler results.
Best Answer
I am afraid this replay will not help you at all. It will not give you any indication about performance in your new test server as all the replay events failed.
I recommend you read this when you take your trace at source, especially following sections:
Event Classes Required for Replay
Data Columns Required for Replay
I will use these terms so I do not repeat the same thing:
SQL Server 2012 production environment = Source
SQL 2016 test environment = Destination
textdata
column for details on eachprovider warnings or error
Errors and Warnings
column for detailsReplayed events
andProviders errors
Replayed events
andInternal errors
Replayed events
all of those timed out either andProvide errors
orInternal error
. In your case it 13+26=39.You can load your test result into a table and then you should be easily able find details about errors. As an example for
Provider Errors
you can use following query which I copied from this blog post.