Re: [PATCH 0/2] Evaluate TPGS setting for ALUA device handler

["Nicholas A. Bellinger" <nab@linux-iscsi.org>]

On Thu, 2009-03-19 at 09:05 +0100, Hannes Reinecke wrote:
> The ALUA device handler should only be attached if the 'TPGS' setting
> in the inquiry structure indicates so. So we should be evaluating the
> 'tpgs' setting when trying to attach the ALUA device handler.
> This way we can get rid of the explicit device table in scsi_dh_alua
> and will support any ALUA capable device out-of-the-box.
> 
> This is a repost of the patchset send yesterday, including
> the suggestions from James B. to just pull the tpgs setting
> off the inquiry data instead of adding a new field in the
> scsi_device structure.

Hannes and co, these patches looks fine to me..   The plan on my side to
is start testing the scsi_dh_alua client changes againt the
lio-core-2.6.git implict ALUA logical unit group and target port group
bits for v2.6.30.  Getting the initial optimized <-> non optimize ALUA
state transitions up should be done in the next weeks for you to poke at
seriously against your scsi_dh_alua nodes using Open/iSCSI.  Also from
there the first explict ALUA using MO_SET_TARGET_PGS.

>From there target_core_mod PR and ALUA fabric against OpenFCoE target
ports and all sorts of Linux Cluster clients is not too far away.

:-)

--nab

> 
> Signed-off-by: Hannes Reinecke <hare@suse.de>