I have found the following SQL very handy to do pretty much the same thing -- check on the progress of a back up. You can substitute whatever other command you want (DBCC, RESTORE, etc.). We have a couple of servers where the collating sequence is binary so you must specify the case properly.
r.start_time AS 'CmdStartTime'
, r.status AS 'CmdStatus'
, r.command AS 'ShortCommand'
, DB_NAME(r.database_id) AS 'DataBase'
, r.wait_type AS 'CmdWaitType'
, r.blocking_session_id AS 'CmdBlockingSessionID'
, r.percent_complete AS 'CmdPctComplete'
-- , r.estimated_completion_time
, CASE WHEN ((r.estimated_completion_time/1000)/3600) < 10 THEN '0' +
END + ':' +
CASE WHEN ((r.estimated_completion_time/1000)%3600/60) < 10 THEN '0' +
END + ':' +
CASE WHEN ((r.estimated_completion_time/1000)%60) < 10 THEN '0' +
END AS 'TimeRemainingForCmd'
, CAST(((r.total_elapsed_time/1000.0)/60.0) AS dec(5,1)) AS 'CmdElapsedMinutes'
, sqltxt.text AS 'CmdFullSQLCommandText'
from msdb.sys.dm_exec_requests r
--join msdb.sys.dm_os_tasks t -- if you want i/o statistics
--on r.task_address = t.task_address
OUTER APPLY msdb.sys.dm_exec_sql_text(r.sql_handle) sqltxt
where command like 'BACK%'; -- Remember, some servers are case sensitive
I remembered that there was a command for figuring out what percent a KILL command had completed and went looking for it (I'm old... looking for something in my library is easier than remembering lately :-D). Did a quick study and wrote some code. When I finally return to this thread, there you are with code that has remarkably similar and sometimes identically named columns as what I came up with. We even selected mostly the same functionality. Great minds think alike. 🙂
I'll post mine anyway because I calculated things a bit differently and I also believe you'll like what I've done with the QUERY column (thank you Orlando!). The duration and "time to go" work for more than 24 hours like I believe yours do.
SELECT DBName = DB_NAME(r.database_id)
,Command = r.command
,SessionStartDT = CONVERT(CHAR(16),r.start_time,120)
,CurrentDT = CONVERT(CHAR(16),GETDATE(),120)
,EstCmdEndDT = CONVERT(CHAR(16),DATEADD(ss,estimated_completion_time/1000, GETDATE()),120)
,CmdTimeToGo = CONVERT(VARCHAR(10),DATEDIFF(hh,0,DATEADD(ss,estimated_completion_time/1000, 0)))
+ RIGHT(CONVERT(CHAR(8),DATEADD(ss,estimated_completion_time/1000, 0),108),6)
,CmdPctComplete = r.percent_complete
,SessionDuration = CONVERT(VARCHAR(10),DATEDIFF(hh,0,GETDATE()-r.start_time))
,UserName = USER_NAME(r.[user_id])
,SPID = r.session_id
,BlockingSPID = r.blocking_session_id
,Query = (SELECT '--' + CHAR(10) + a.[text] + CHAR(10) AS [processing-instruction(QueryCode)] FOR XML PATH(''),TYPE)
FROM sys.dm_exec_requests r
CROSS APPLY sys.dm_exec_sql_text(r.[sql_handle]) a
WHERE r.command IN ('BACKUP DATABASE','RESTORE DATABASE','DBCC CHECKDB','KILL','UPDATE STATISTICS')
ORDER BY DBName
[EDIT] p.s. Do you also dislike Extended Events as much as I do? :w00t:
[EDIT2] The CmdTimeToGo and CmdPctComplete don't appear to include the Transaction Log for Backups (which is usually very short for me because of the PIT backups I do) nor the Redo/Undo phases of a Restore. I can live with that but just wanted folks to know.
is pronounced "ree-bar
" and is a "Modenism
" for R
First step towards the paradigm shift of writing Set Based code:
________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.
"Change is inevitable... change for the better is not".
"If "pre-optimization" is the root of all evil, then what does the resulting no optimization lead to?"
How to post code problems
How to Post Performance Problems
Create a Tally Function (fnTally)