Top priority for me today will be to research and fix that problem. Likely a configuration problem with the new box.
If I'm not mistaken, the post still "takes" on that particular error and the error is happening on the way to taking you to your message.
Nearly all stored procedures use "with (nolock)" to limit our exposure to this problem, but that's not true of most of the T-SQL, which post_info is still full of. The new version, which still isn't ready for production, is about half the size and almost exclusively uses stored procs, and only about half as many as the current version's T-SQL statements.
Interestingly, I've found that "with (nolock)" slows down some processes, but have yet to find any rhyme or reason to it.
Anyway, half an hour before the open and the new db is acting like a Greyhound bus being used to do a VW microbus's job. Not even close to breaking a sweat, and realtime search indexing is enabled.
Edit: Live is looking comparatively good for the web-server, too. Likely because it can get its data from the db server so much more quickly and not have requests queueing up like was probably happening before. Still, it's having to work hard. Just not quite as hard. I hope to have the new webserver into production this week or next, which has around 50% more horsepower than the current one.