On 4 Mar 2006 at 18:19, Jeffrey L. Taylor wrote: > Quoting Ulrich Windl <[EMAIL PROTECTED]>: > > Hi, > > > > I came across a very strange problem, which I'll describe like this: > > On a new machine with Nvidia Nforce4 chipset, AMD Athlon 64 X2 3800+ > > (dual core), and a SATA2-2 disk Windows/XP stops booting after > > _significant_ harddisk IO (as from the perspective of the HD > > LED). "Stops" means the rotating progress bar freezes after some > > time, and I#ll have to press reset. > > > > Now what's interesting: It only happens when I had booted Linux > > before. I can only suspect that it's either related to Linux doing > > something with the SATA controller that BIOS & Windows don't undo, > > or it is the activation of the second core of the CPU that makes > > Windows freeze on boot, or whatever. > > > > Windows event log is reporting "disk" errors after a successful boot > > then > > > > My BIOS is completely up to date. > > > > What makes things worse is the fact that neither hdparam nor > > smartctl seems to work with SATA (only tried 10.0). Any progress in > > thios area? > > > > Any ideas what the reason of the problems could be, or what to do > > against it? It's bit annoying. > > > > Please no suggestions like "remove Windows". > > > > Are you switching OSs while one or both are in hibernate/suspend? It
No, not that I could remember. Also I found out that resume from LVM swap doesn't work anyway (in 10.0). > seems like you should be able to do this, but I had to re-install > Windows XP four times and fix filesystem corruption on Linux from the > rescue disk twice in a 1-2 month period while switch OS during > hibernate/suspend to disk. The problem has not happened once in the You were not sharing the Windows swap file with Linux, were you? ;-) > 6+ months since I starting always doing full shutdowns before > switching OS. I'd be surprised if Windows/XP would swap outside its partitions (just as I'd be surprised if Linux did). After all, it wouldn't explain why a boot after power on or after hardware reset always succeeds. Regards, Ulrich --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
