|
From: Szakacsits S. <sz...@si...> - 2004-03-05 17:14:43
|
On Fri, 5 Mar 2004, Anton Altaparmakov wrote: > I can see that but I still do not see it as a showstopper. I think > people would rather have a tool that had bad error code reporting than > not have the tool at all... IMHO it's a potentially great benefit for minimal investeded time. I just need to categorise the 89 failure points into 6-10 exit codes. > > Seems to be unneeded. If user boots Windows then it's unavoidable. If they > > want to run ntfsresize or ntfsclone then any action will be refused if the > > fs is inconsistent even by using the --force option. And at present I > > can't see any other way to damage or lose data using Windows or released > > Linux-NTFS tools. > > I guess that is true but for example you could mount the resized > ntfspartition in Linux and then you could get data corruption (maybe, I > don't know if you actually could) if $Bitmap is inconsistent. I though about that also and I think that's also impossible today. Using a write enhanced driver in the _future_ it could be possible _if_ the driver allows to mount a dirty volume read-write. It should never allow, perhaps only read-only mount. > Well how about releasing it as it is now as 1.9.0? Additional features > like better error reporting could wait for 1.9.1, 1.9.2, etc, > releases... What do you think? Sounds like a plan, I especially like the "0" indicator :) But let's wait to Monday, I try to arrange some time to get some worth and risk free issues to be done in the weekend. Szaka |
