Words of true desperation:
Several days ago, my system SSD broke, so I send it for RMA. I was in last step of making my bachelor thesis, so I had to be up and running. I used my second, older 60GB SSD for temporal system. Today I got new SSD from RMA. Since I have one of those P55 motherboards without proper native support of SATA3, which is implemented using shitty Marvel controller, I have never really used it. Still, I just wanted to see how it would go, and plugged it into SATA3 port, installed Windows 7 and all drivers and applications. So at this time, I had 2 SSDs, each having Windows 7 copy installed, and I also added boot record (using bcdboot) of my new system to the older windows bootloader (temporary solution before setting new SSD as first boot option). Everything worked quite well. BUT...
As a last thing I need to work, I began installing Visual Studio 2013 (from DreamSpark). As I returned from kitchen, I saw Windows boot error screen, with code 0xc00....0f. Well, f**k. I quickly understood that my fresh Windows installation is possibly gone, with few hours of work. So, I thought I can use my older Windows to see what hapened. But it was corrupted too! And not its MBR. It began to load normally, but freezed. Same with safe mode.
I used my copy of Windows DVD to repair "startup problems" of the older system (new one could not be detected at all). As repair process finished, I booted into system, running chkdsk on the new SSD. So many bad sectors! At first I thought SSD is just gone. But after many restars, formats and everything I reconnected it from SATA3 to SATA2 port. Suddenly I was able to do low level format, and actually managed to install, once again, new copy of Widnows onto the drive. Without any issue at all, at least so far.
If you made it so far, I am wondering about 2 things:
- How the hell could crash of SSD, connected to different port and host chip, also somehow affected my older system? Yes, it had boot record for it, but it freezed during bootup stage, in the middle of Windows start animation, so obviously its MBR was not corrupted.
- Could Marvell SATA3 controller really caused the issue in the first place? Most concers about it is about performance, not about stability nor destroying MBRs... Could plain installation of Visual Studio corrupt MBR? Even if there was any system crash not having anything to do with both SSDs (power surge, BSOD etc..), nothing normally writes in the MBR sectors.
Please pardon some typos and possible grammar erros, I am too depressed and tired to even think straight after more than 12 hour marathon of dealing with s**t.