You can set the ServerName property of a connection manager in a package configuration, which can come from an XML file, registry setting, environment variable, parent package variable, or SQL...

The problems I had would probably not be an issue if I worked primarily with Teradata, or at least had access to a DBA that knew what they were talking...

Any use of the outer table in the WHERE clause will change your outer join to an inner join. You can use the subquery method shown above, or move...

In SQLCMD mode you can use

:r file.sql

to include input from an external file. Read about it in Books Online, the external file is run after GO and not...

I've had some experience connecting to a client's Teradata system to run queries and moving data back and forth to my SQL Server systems. I didn't like the client...

Replace your primary key with a clustered unique index and use WITH (IGNORE_DUP_KEY).

You'll see information messages about "Duplicate key was ignored", but you won't get errors.

You could use SET TRANSACTION ISOLATION LEVEL REPEATABLE READ or SET TRANSACTION ISOLATION LEVEL SERIALIZABLE before your BEGIN TRAN statement to lock all SELECTED rows until the transaction is finished....

What works best is to synchronize the SID values for SQL logins across all your servers, so the users don't get orphaned in the first place when moving databases between...

I don't have anythng to add to the variable vs. temp table discussion, but I have a correction for the original script.

The line "if not exists (select * from sysobjects...

You can use CRTL-Enter to add a line break in a quoted string in an expression, but I believe it will show up in the email as double-spaced (i.e. extra...

In the Job Step Properties dialog, change the Type value to "Operating system (CmdExec)" and put the command you want to run in the Command text box.

Apex SQL Edit has very good source control integration. Tools from other vendors like Quest and RedGate probably do, too.

But why not just use the source code integration in...

I think you have conflicting goals. If you track daily schema changes how are you going to keep old archive data in anything resembling its original form?

Maybe you should...

The best method I've found is to store packages in files (and in source control) for development and only store them in msdb when they are deployed (if at all)....

I was backing up to a network fileshare that was about 1.5TB, so size was not an issue.  It was an IDE-based piece of junk, so throughput was an issue. ...