That worked. Thanks. Wish I could have found that in BOL.

Thanks,

Chris

It worked. Once I copied all tables over to the primary files group I dropped the old tables and then that file, from the DB In the primary file group...

That's my suspicion, that the file group was corrupted somehow and DBCC doesn't catch the problem. My fix for this was to script out each table on that file group...

Bouncing the server made no difference. Still the same error. I am testing with

DBCC ShrinkFile(FG2Data, EmptyFile)

rather than attempting the data load every time.

Thanks,

Chris

I have a VB script now, and I could ignore the error, but I'm trying to use the script to decide whether to run the objects in the sequence container,...

Prod:

DBINFO @0x6902EF8C

dbi_dbid = 211 dbi_status = 65544 ...

I ran Ola's original script, with the data purity arg. Fortunately this is a UAT server, so it didn't affect Prod, but because of that the usage is much lower...

I left it to run all weekend and it turns out it only ran 4:41.00. Is CheckDB cumulative, in that whatever it did before is saved and it continues where...

The maintenance job runs:DBCC CHECKDB ([DMART_MSRIPOS]) WITH NO_INFOMSGS, ALL_ERRORMSGS, DATA_PURITY

(Ola Hallengren's scripts)

Interactively, I run "DBCC CheckDB(DMART_MSRIPOS)"

If I watch the wait resources, I can see that they are changing, suggesting that...

The waits I have are PageIOLatch_SH and CXPacket, on multiple instances of the command "DBCC Table Check". DM_Exec_Requests show the extact query is being: "DECLARE @BlobEater VARBINARY(8000) SELECT @BlobEater...

The main goal is to remove the login, but I don't want to leave orphans around, so I am trying to remove the users associated with the login from each...

I second GilaMonster. Very large DB files seem to take forever to shrink. I have had them take 2 days, myself. This is a great argument for a clustered index...

It seems to me that you'll need to create a script object in a package to do the job. That will give you the most flexibility. Select the entire record...

I don't think it will do any good to install a SQL 2005 driver on a 2008 server. I can see where that would bite me in the butt.

Thanks,

Chris

If I got handed that mess I'd allocate each instance 4G of RAM, and leaving 4G for the OS, that would allow the other node to handle it should there...