> No [copying files with metadata] can't [be done in userspace]. Example:
> /home/bob is a Windows share, mounted from an NT server.
> Bob does "cp foo bar" to make a copy of his file.
> That file has:
> * ACLs
> * multiple forks
> * extended attributes
> * several gigabytes of data
> * holes (compression)
> Now, you want to haul several gigabytes of data over 10baseT
Bob obviously does. Or are you suggesting cp should know what kind of
file system it's using, from what kind of clients, and use some kind of
RPC call when possible?
> and lose all the information that Linux can't understand???
It depends on how that information is presented to/by Linux. If Linux
supports the particular (meta)information, tools should handle them
transparently. Stuff like "forks" could be handled by presenting them
as directories.
> You'd have the server uncompress and recompress all the data too.
Why does this follow? How could it be different?
> That looks like a 30-minute operation, plus data loss
> which causes a security hole. No, that is not OK.
Well, the solution is of course to provide complete support for all
kinds of remote file systems and operating systems, including all the
semantic implications. Perhaps it can be done.
I think its better, if for no other reason than it being a lot simpler,
to learn to live in a Unix paradigm, and map foreign attributes to that
paradigm as good as possible.
> Perhaps Bob should bypass the kernel filesystem. He could do raw
> network I/O to reach the server using a setuid /bin/cp.
Perhaps Bob should put his data on a real system :-)
~kzm
-- If I haven't seen further, it is by standing in the footprints of giants- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html