[OpenAFS-devel] RxOSD configure option

Hartmut Reuter reuter@rzg.mpg.de
Mon, 20 Jul 2009 16:13:34 +0200


Felix Frank wrote:
> Hi,
> 
> RxOSD is currently ifdef'ed away and enabled by configuring with
> --enable-object-storage.
> 
> Is this structure desirable for the main tree? Or should OSD support be
> compiled in at all times?
> 
> The latter would save ifdefs and make libuafs Makefiles a lot more simple.
> 
> Note that from outside, servers build with and without OSD support
> initially act the same. The OSD-capable fileserver will change behaviour
> if and when one or more of its volumes are set to use OSD.
> However, the fileserver flavors are not interoperable wrt. the on-disk
> data, so binaries cannot be replaced at will. As an OpenAFS-server with
> full RxOSD support will thus be incompatible to earlier versions, it
> might after all be desirable to continue building non-OSD binaries.
> 
> Thanks in advance for you opinions.
>  - Felix

I think the client should be compiled without ifdefs because it may have
to deal with different cells among which may be cells deploying OSDs.
There should be no overhead for non-osd AFS traffic even if the client
contains the OSD extensions. And on the client use of the osd extensions
can be controlled by the "fs protocol" command (e.g. at start time).

The server side, however, should depend on the configure.

Hartmut


> _______________________________________________
> OpenAFS-devel mailing list
> OpenAFS-devel@openafs.org
> https://lists.openafs.org/mailman/listinfo/openafs-devel


-- 
-----------------------------------------------------------------
Hartmut Reuter                  e-mail 		reuter@rzg.mpg.de
			   	phone 		 +49-89-3299-1328
			   	fax   		 +49-89-3299-1301
RZG (Rechenzentrum Garching)   	web    http://www.rzg.mpg.de/~hwr
Computing Center of the Max-Planck-Gesellschaft (MPG) and the
Institut fuer Plasmaphysik (IPP)
-----------------------------------------------------------------