CentOS Win7 Dual boot on NVRaid (fakeraid)
Hello everyone,
For my project this weekend, I decided I would dual boot CentOS 5.6 and Windows 7 on my desktop. I have 2 500GB SATA drives, running in (fakeraid) RAID0 from the onboard Nvidia chip on the motherboard. I installed Windows 7 on a ~500GB partition first, everything went great. Next, I installed CentOS. I had 3 partitions, one for root (~300GB), one for swap (4GB), and a third for /usr (~200GB). The install went well, but when I restarted, I was still booting straight into Windows. At no time did I see a grub menu.
I booted from a live cd and reinstalled grub manually. But after I did that, the system would just boot to a grub prompt. I tried so many different things over the course of 6 hours, that I can't remember each of them. But as I tried manually configuring/installing grub, I would receive various errors from "File not found" to "stage1 not read correctly." At one point, I remember fdisk telling me "the number of cylinders for this disk is set to... larger than 1024." I eventually wondered if it was a problem with how I installed CentOS (since I'm usually an Ubuntu guy). So I put in my 10.04LTS CD and installed over the same partitions with Ubuntu. But at the end of the install, I kept receiving errors that grub couldn't be installed to /dev/sda. I kept pointing it to /dev/mapper/nvidia_XXXXXX, but it would just go back to trying /dev/sda.
Both CentOS and Ubuntu recognized the Raid configuration. Both of them were aware of the fact that the devices were mapped to /dev/mapper/nvidia_blahblahX. It's just as if grub wasn't yet able to access the partitions at that point of the boot process.
I ended up restoring the Windows 7 boot manager, so at least I'm back to square 1 now. But does anyone have any idea why I'm having so much trouble with this? Is what I'm wanting to do even possible?
Thank you in advance for your help.
|