[openib-general] [SRP] [RFC] Needed changes to support fail-over drivers

[Ishai Rabinovitz]

On Tue, Jul 25, 2006 at 04:58:48PM +0300, rdreier at cisco.com wrote:
> [CC'ing linux-scsi as well -- I think we'll get better insight from =
> there]
> 
> OK, but is this a valid assumption?  What happens for iSCSI and/or iSER?

>From Mike's response I understand that it is a reasonable behavior to keep the
host (at least for a period of time) and let the userspace daemon be
responsible to the reconnection or deallocating of that host.

> 
> How does the daemon know when something is gone for good vs. when it
> might come back?
> 

I think we should use a time out in the daemon.

> 
> Why does userspace need to be able to disconnect a connection?
> 

There are two options on who will initiate the disconnection: the userspace
daemon or the ib_srp module.  I considered both options and I was not sure
which one is better.  I choose to do it in userspace because it looks a good
symmetry that both the disconnection and reconnection will be initiate in the
same place.  I will accept your comment and change it to the kernel.


> 
> Why the asymmetry here?  In other words, why does anything work for
> reconnect_target but only the literal "erase" work for erase_target?
> 

Because erase_target is a destructive command that can not be reversed I think
it should use a more safe approach.

-- 
Ishai Rabinovitz