[SAMBA4][PATCH] dd client for samba 4

[James Peach]

On Tue, 17 Jan 2006 09:26 pm, Andrew Bartlett wrote:
> On Tue, 2006-01-17 at 16:18 +1100, James Peach wrote:
> > On Tue, 17 Jan 2006 03:16 pm, James Peach wrote:
> > > On Tue, 17 Jan 2006 03:06 pm, tridge at samba.org wrote:
> >
> > [snip]
> >
> > > > Having code like sdd.c in Samba that re-invents popt, and avoids
> > > > using all the existing command line handling code for options
> > > > (especially auth options) is something that I'm very reluctant to do,
> > > > so it would require some very strong justifications!
> > >
> > > I guess I don't feel as strongly about this. I expect to use this for
> > > generating client load. Strictly speaking I don't need _any_
> > > authentication.
> > >
> > > Maybe I could add the popt handling prior to looking at any dd
> > > arguments? sdd [common samba options] [dd-style options]
> >
> > The attached patch implements this ...
>
> You still don't use cmdline_credentials.  This global variable is what
> the POPT_COMMON_CREDENTIALS creates.

Ok, I didn't understand that this was what you meant. IIRC, the reason I
wanted to use the SMB URL for the "if" and "of" options is the case where
you have 2 connections do different servers and you need to have different
authentication information for each. This will probably be a rare case,
however, so I can certainly use cmdline_credentials until it's needed.

On the issue raised on IRC of there already being a dd clone called
"sdd", I'm more than happy to change the name. cifsdd?

> I think tridge has a point about making this an smbclient command.  Once
> this is all done, what will be the difference between the operation of
> the modified sdd and smbclient -c 'dd .....'?

First, the actual syntax of smbclient is smbclient -c 'foo' //host/share. This
creates ambiguity in the dd syntax.

Where does the data go in this case?
    smbclient -c 'dd if=/dir/file1 of=/dev/null" //server/share

What about this case?
    smbclient -c 'dd if=//server1/share/dir/file1 
of=/dir/file1" //server/share

The dd subcommand certainly wouldn't operate like any of the other smbclient
subcommands.

Second, the smbclient syntax is incompatible with existing test harnesses that
drive dd. The usage I'm planning for this is something like:
    DD=/usr/samba/bin/sdd SRC=//server/share/foo DST=/dev/null ./test.sh

Third, the sdd command is carefully constructed to do exactly the IO
you tell it to as far as possible. If you specify a 32KiB block size,
then you get a 32KiB read on the wire. To move this code into smbclient
would require moving its IO path. smbclient would then contain 2 parallell
IO paths. I don't see how making smbclient more complicated in this way
is helpful.

> Why the change to popt.h?

There was a gcc warning; when returning const char * const, the second const
has no effect.

-- 
James Peach | jpeach at sgi.com | SGI Australian Software Group
I don't speak for SGI.