I’ve been itching all day to write about this, but I got caught in way too many other small issues, but now, that I’m home, I hope I get the chance to write about this little mishap that I got today.
I was documenting an emergency role change for a secondary involved in log shipping and guess what, I managed to f… screw log shippingJ. Yeap, on the prod servers.
Here’s a little repro of what happened.
I’ve created log shipping on the same instance, AdventureWorks being the primary DB, while test is the secondary. See below for details.
As you can see, I decided to have my database as read-only, so users can actually query tables. Sure enough, I can access the DB, as seen below:
Heck, I can even open a query window to query some data, right? Wrong. If you open a new query window and forget about it, this is what happens:
And this is the actual print screen from production; apologies for all the rectangles and stuff, but it’s a bit of a confidential stuff (except for the error log):
I put this here because I find it fascinating: the actual error message is hidden in one of those “successful” stepsJ.
Anyway, a quick look using sp_who2 has shown the culprit:
Apologies for the extremely truncated picture, but well, you get the idea.
I just ran KILL 55 and I was good to go.
I hope this helps.
Best regards,
io
SQL-PLAY 