Sql-server – SSMS Disk Usage report does not show auto grow events

auto-growthsql serversql server 2014ssms

I have for some time suspected that the Disk Usage report in SSMS have stopped showing autogrow events. To test I created an empty database with a very small auto grow setting. I then inserted a lot of data into a table in the DB so that I knew it would have to grow. The file was bigger when looking at file properties of the DB, but the Disk Usage report did not show any autogrow events.

I then tried running Aaron Bertrand's script from this post, and that clearly shows every autogrow event.

So, why will not the Disk Usage report show the same growth? Any suggestions?
Btw, it is SQL Server 2014 SP2.

Best Answer

The logic for the disk usage report is baked into SSMS and while we can't know what the RDL looks like (and if any filtering is done) I grabbed this query sent by SSMS 2016 using Profiler:

if (select convert(int,value_in_use) from sys.configurations where name = 'default trace enabled' ) = 1
begin
declare @curr_tracefilename varchar(500) ;
declare @base_tracefilename varchar(500) ;
declare @indx int ;

select @curr_tracefilename = path from sys.traces where is_default = 1 ;
set @curr_tracefilename = reverse(@curr_tracefilename);
select @indx  = patindex('%\%', @curr_tracefilename) ;
set @curr_tracefilename = reverse(@curr_tracefilename) ;
set @base_tracefilename = left( @curr_tracefilename,len(@curr_tracefilename) - @indx) + '\log.trc' ;


    select  (dense_rank() over (order by StartTime desc))%2 as l1
,       convert(int, EventClass) as EventClass
,       DatabaseName
,       Filename
,       (Duration/1000) as Duration
,       StartTime
,       EndTime
,       (IntegerData*8.0/1024) as ChangeInSize
from ::fn_trace_gettable( @base_tracefilename, default )
left outer join sys.databases as d on (d.name = DB_NAME())
where EventClass >=  92      and EventClass <=  95        and ServerName = @@servername   and DatabaseName = db_name()  and (d.create_date < EndTime)
order by StartTime desc ;

This isn't substantially different to Aaron's script and I don't see how it could return different results. It looks to me that even if the traces roll over, it will still be iterating all of them (and they wouldn't roll over 5x in the space of a few minutes or even hours of testing).

Related Question