||[Feb. 4th, 2004|09:47 pm]
Besides the usual avoidance that seems common at work, I kept pretty busy today. Especially while fighting with a test server I am using to test a migration strategy for Friday night.
It appears despite the precedence setting in the BIOS of IBM Netfinity servers, Windows 2000 will take the PCI slot order into consideration when choosing what device to boot from. Since the server has 2 RAID controllers (one in slot 4 (data) and one in slot 5 (boot)), I kept ending up with two conditions: either bootup properly with no data volume (because I deleted the array) or a bootup error which could only be cured by deleting the array on the controller in slot 4. So...I had to flip flop the physical chassis and drives of each controller, blow away the existing configs, copy the old configs from the drives to the new controller and hope it all worked. I was amazed it did. Yes, I might have been able to physically move the controllers to each others slots but I seem to recall once when I did that I was greeted by a warm blue screen of death. What I am curious about is if this will work on two different models of controllers in the same line - say a ServeRAID 4l and a ServeRAID 4lx. My assumption is yes but I may have to test that on Friday when the real migration occurs.
My video capture project is slowly moving along. I cannot replicate my earlier audio sync problems so all the upgrading I did MAY have fixed the problem. I have one more test to run (tonight maybe) and I will feel comfortable it is pretty much resolved. Then I get to burn hours and hours of time I should spend studying.
Time for some tech humor: http://chroniclesofgeorge.nanc.com/