Re: [netatalk-admins] Samba/Unix File Management vs. Resource Forks?


Subject: Re: [netatalk-admins] Samba/Unix File Management vs. Resource Forks?
From: Clemens Anhuth (clemens.anhuth@snap.de)
Date: Fri Jul 16 1999 - 04:09:07 EDT


a sun wrote:

> samba has a --with-netatalk option that causes samba to treat
> .AppleDouble directories in a special fashion. i suggest that you
> use it.

ah, a compile time option.

i found this in samba-2.0.4/cvs.log:

"netatalk.c contains the meat of the code required to move Macintosh
resource
forks around when Windows users manipulate files created by a mac.
This
allows for greater integration between netatalk and Samba, but is,
of course,
potentially a recipe for disaster. All the netatalk related code is
insulated
by #ifdefs from the standard build."

makes me go "hm" and <gulp>. :)

and leaves "manipulate files" undefined, but i think this is
important to know in order to judge how far and tight the
integration actually can go.

> in the future, samba and netatalk will cooperate even more.

how far will this go technically?

will making use of windows explorer and the other standard tools
provided with windows 9x/nt (copy, xcopy, etc.) be pure in this
respect and not cause problems?

what about the mentioned case of programs making use of unorthodox
file-management?

i guess the main question is, how bullet-proof will it become with
whatever can be done about this problem?

if you find the time, don't hesitate throwing all the dirty bits and
bytes regarding this at us, i guess some are interested in this.

as far as i can see it comes down to an individual decision per
company, business unit, etc. on wether to integrate samba and
netatalk.

we decided against sharing the same directories via samba and
netatalk for now, because we want to completely rule out problems of
that kind.

(the per default enabled crlf translation of ethershare and suse's
6.1 netatalk package have caused some irritation and consumed quite
some developer time already in our company.)

again, sorry for being resistive and elaborative, i just try to
understand this matter as completely as i can.

thank you for your work and your responsiveness, adrian (and to the
others participating in this thread and on this mailing list of
course).



This archive was generated by hypermail 2b28 : Sat Dec 18 1999 - 16:16:58 EST