On Tue, Mar 11, 2003 at 09:25:06AM -0800, Bill Studenmund wrote:
> On Mon, 10 Mar 2003, Manuel Bouyer wrote:
> 
> > On Mon, Mar 10, 2003 at 01:31:30AM +0100, Christoph Badura wrote:
> > > Actually, what business do we have resetting the SCSI busses?
> > >
> > > 1) this wreaks havoc with scsibusses with multiple initiators on the bus
> > > (vulgo: HA clusters).  We'll have to deal with that anyway at one point.
> >
> > Other HBAs on the bus should also be able to deal with a reset
> 
> I think the point is what if the other initiator is in the middle of a
> transaction? i.e. we'er the main server rebooting, and failover has
> already happened. In that case, it's rude to just reset the bus.

The HBA sould support this and restart the transactions. The scispi mid-layer
has support for this, and {,e}siop use it.

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 24 ans d'experience feront toujours la difference
--