Copied file vanishes when client is OS X

Ed Holden eholden at mclean.harvard.edu
Wed Nov 26 18:48:46 GMT 2003


Benny,

This was above and beyond, to say the least.  Thank you very much for 
working on this, and I'm eager to see what Apple has to say about it.

The FAT issue is baffling.  I'm using ext3 on my target partition, and 
the differences between it and FAT should be transparent from the client 
side.  I wonder if permissions are part of the issue?  That's one major 
difference between ext3 and FAT that would be noticeable via Samba.

Thanks again,
Ed

:: Ed Holden
:: Administrator, Research Information Systems
:: McLean Hospital
:: Tel: (617) 855-2822
:: Web: http://research.mclean.org/ris

Benjamin Riefenstahl wrote:
> Hi Ed,
> 
> 
> Ed Holden <eholden at mclean.harvard.edu> writes:
> 
>>If you want to compare using the same file, it's the PDF file on
>>http://www.edholden.com/linux.
> 
> 
> I can now reproduce it with your file if there is no other file in the
> directory that has Mac OS metadata.  Reading the log at debug levels 2
> and 3 makes for some interesting reading ;-).
> 
> It looks like the Finder does it like this:
> 
> - Create the files for actual datafork and for Mac OS metadata.
> 
> - Copy the data. 
> 
> - It looks at *some* file in that directory with metadata and resource
>   fork.  If there is some other file with metadata, it looks at that
>   and all is well.  If it only finds the metadata file that it has
>   created before for gnu-linux.pdf than it deletes both files again.
> 
> What it *should* do in this instance, is to realize that the metadata
> file it has created only contains default information and discard it,
> but leave the actual data file alone.  Maybe at this point it hasn't
> yet written the metadata file correctly and so it thinks something
> went wrong.  Or maybe the routine that reads the metadata file for
> verification recognizes that the data in there is just defaults and
> thinks that *that* is an error, because such a file shouldn't actually
> exist.
> 
> Interestingly I can not reproduce the problem with a virtual FAT disk
> on my Mac, although that has the same on-disk representation for Mac
> OS metadata, and so it should in theory be handled in the same way by
> the Finder.  But maybe there is something more going on when there is
> a network involved.
> 
> I created a bug report for Apple.  They will probably have to fix this
> in the Finder or at least they should be able to tell us what goes
> wrong here.  We'll see if we get feedback...
> 
> 
> benny



Any information, including protected health information (PHI), transmitted
in this email is intended only for the person or entity to which it is
addressed and may contain information that is privileged, confidential and or
exempt from disclosure under applicable Federal or State law. Any review,
retransmission, dissemination or other use of or taking of any action in
reliance upon, protected health information (PHI) by persons or entities other
than the intended recipient is prohibited. If you received this email in error,
please contact the sender and delete the material from any computer.


More information about the samba-technical mailing list