Assessing the 00Z 27 Aug GFS run, it looks like New Orleans will get hit by 0800 29 Aug 05 w/70KTs sustained. By 2000 29 Aug 05, storm will be very tightly wrapped about 90 miles W of Hattiesburg w/40 Kts sustained (she'll be about 90 miles NW of Tupelo by 0800 Tue w/25Kts sustained). In this scenario, I'd expect this storm to be dropping tornados all over the place.
NGP seems to concurs with this, except landfall appears about 50 miles W of New Orleans (Morgan City, Lafeyette), and with about 55 KTs sustained and not as tightly wrapped (tornadoes shouldn't be as numerous).
NGP, on the other hand, shows Katrina making her way inland near Biloxi. The NGP model doesn't show the storm as tightly wrapped as the GFS model however. As a result, the windfield is pronouncedly dispersed over a greater area compared to the GFS model. Despite peak sustained winds projected to be merely 70 KTS, they'll extend all the way to Pensacola (and perhaps as far as Panama City), albeit significantly attenuated there. While NGP projects the storm to be centered over Jackson by sundown Monday, 40 KT winds will still be pummeling the Gulf coast to at least Mobile (and possibly as far E as Pensacola) even so winds over Jackson will be a mere 20 KTs sustained.
Nevertheless, and all that notwithstanding, at least 1" of rain per hour for the duration and area mentioned can be expected at a minimum Local areas of heavier rainfall is to be expected as squall lines move through the area. I wouldn't expect rainfall to be more than 1 1/2" per hour in any one particular area for any significant amount of time. While it may rain the equivalent of 3"/hr for about 10 minutes, it'll return back to the normal expected rate of 1" per hour. And that's going to go on for at least 12 hours (proably 18).
Looks like you guys are going to get a little wet down there (if 1 1/2' of water falling out of the sky in 1/2 a day can be called "getting a little wet").