On 25 March 2012 18:16, Bill Gurley <gur...@ion.chem.utk.edu> wrote: > On 03/24/2012 07:49 PM, jumpingspo...@hushmail.com wrote: >> Thanks for your input Bill! Now that you mention it, there could easily >> be other explanations for what I'm experiencing. For example, remember >> what I said about Windows 7 dividing the hard drive into 2 partitions? >> If you do an installation of Windows 7 on an empty hard drive (without >> multi booting), it will by default create 2 partitions: the first one is >> the boot partition (about 100 MB) and the second one takes up the rest >> of your drive. >> >> When you said the partitions could actually be LARGER than the source >> partitions, that got my attention. It is possible that my guess about >> the smaller partition being exactly 100 MiB was wrong. The only reason I >> tried to create partitions that were exactly the same size is that >> Steven's document said the file geometry had to match exactly. If I >> still have time, I will make partitions on the destination drive that >> are slightly larger than the partitions on the source drive, then see if >> files still get corrupted. >> >> There could also be a number of other problems that I don't know about >> (but others might). Given all the problems I've been having, I'm quite >> impressed that you have the courage to actually use the source drive's >> image. To be honest it would be much easier to simply do a clean install >> of Windows! All this testing has set my work back by a week. > > > I still have not read that document by Stephen (will try to look at it > soon), so I'm not sure about the "geometry" statement. That word is > normally used in the context of "disk geometry": cylinders, heads, > sector size, etc. This becomes critical when creating partitions. > i.e., a disk partitioning program such as gparted "knows" what the disk > geometry is, and uses that when creating the partitions and the > partition table. In my mind, the only relevance of this to clonezilla > is when using the "disk" cloning instead of "parts" cloning. I hope > that Stephen will correct me if I'm wrong, but I think that when you > restore to a drive using an image that was a *disk* image, it is best to > use an identical drive. I believe that it may still work on a different > drive but it would possibly be forcing a disk geometry to that drive > which might not be its "natural" drive geometry.
In recent LInux with libata there is no notion of disk geometry anymore. Linux just assigns the maximum possible number of sectors and heads. If that does not match what your disk reports your restored image does not work. > > So, if you are dealing with a "parts" image, you can restore to ANY > drive that has sufficient space, as long as you first manually create > the necessary partitions using gparted or similar program. This way you > will be honoring the actual native drive geometry for the target drive. > I don't think that the clonezilla "parts" image really cares then > about drive geometry. It will simply restore to the partitions that you > created before starting the restoration process. > > Does that make sense? No. Any Linux created partition table has the same geometry. Regards Michal ------------------------------------------------------------------------------ This SF email is sponsosred by: Try Windows Azure free for 90 days Click Here http://p.sf.net/sfu/sfd2d-msazure _______________________________________________ Clonezilla-live mailing list Clonezilla-live@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/clonezilla-live
