Xrhrft

Just implemented Ola's Parallel execution for backups, integrity checks and index maintenance. Everything appears to be running great but I am seeing sql_message_id 50000 in the sysjobhistory table for the new executions. I am also seeing [SQLSTATE 01000] on each line of the job output files. Is this normal behavior? I can't seem to find any in-depth documentation online.

The Message 50000 in msdb.dbo.Sysjobhistory is intended.

Take part of Ola Hallengren's index optimize procedure:[dbo].[IndexOptimize], part of the integrity checks: [dbo].[DatabaseIntegrityCheck] or part of the backup procedure: [dbo].[DatabaseBackup], the same RAISERROR() Statements will be shown:

  SET @StartMessage = 'Date and time: ' + CONVERT(nvarchar,@StartTime,120)

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Server: ' + CAST(SERVERPROPERTY('ServerName') AS nvarchar(max))

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Version: ' + CAST(SERVERPROPERTY('ProductVersion') AS nvarchar(max))

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Edition: ' + CAST(SERVERPROPERTY('Edition') AS nvarchar(max))

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Platform: ' + @HostPlatform

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Procedure: ' + QUOTENAME(DB_NAME(DB_ID())) + '.' + QUOTENAME(@SchemaName) + '.' + QUOTENAME(@ObjectName)

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Parameters: ' + @Parameters

  SET @StartMessage = REPLACE(@StartMessage,'%','%%')

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Version: ' + @VersionTimestamp

  RAISERROR(@StartMessage,10,1) WITH NOWAIT



  SET @StartMessage = 'Source: https://ola.hallengren.com' + CHAR(13) + CHAR(10) + ' '



  RAISERROR(@StartMessage,10,1) WITH NOWAIT

The RAISERRROR()'s are going to be executing when the procedure runs.

Looking into the history, we see the 'error's' shown:

select message From msdb..sysjobhistory 

where sql_message_id =50000

Result

Executed as user: DOMAINtest. ...0000)  Server: test [SQLSTATE 01000] (Message 50000)  Version: 12.0.5546.0 [SQLSTATE 01000] (Message 50000) ...

With the same Server and Version info as defined in the previous T-SQL examples.

Why does the 'error' start at 50000?

It starts at 50000, because starting from 50000 and upwards is where user generated errors can be defined.

Why are there raiserror()'s in the procedures?

My opinion is that the RAISERRROR() is added to add additional information to the history DMV, as to know what is being executed, and have additional information without it being a mess. It gives you control over what will be shown.

I will leave more in depth reasoning to Ola himself.

Subsequentially the history is added to the SSMS Job Activity monitor that we (Most of us) know and use:

So to answer the question, it is coded into the proc with reasoning behind it.

RAISERROR() 50000 default example

You can only generate custom errors with raiserrror when using an error number starting from 50000.

Testing with a job, with no specific error number:

Result:

Starts at 50000.