Re: DRAFT Nashua interim meeting minutes

["Mallikarjun C." <cbm@rose.hp.com>]

David,

I just have two comments on this note.

>Within-connection recovery reissues command on same connection to fill
>	CmdSN holes.  Has to be on same connection due to connection
>	allegiance.  

If there is a CmdSN hole, the command should be unacknowledged causing
the initiator to re-use the same CmdSN on retry.  I don't see connection
allegiance here since the CmdSN was never ack'ed.  IMHO, technically 
it should be able to be re-issued on any connection in the session
(though don't know why an initiator would do that).  

>These errors require framing support over TCP -
>	in the absence of framing, the connection has to
>	be terminated.

This is generally true, since most likely it would be a header 
digest error.  If the CmdSN hole was due to a data digest error on
immediate data, original framing is still intact.
--
Mallikarjun 


Mallikarjun Chadalapaka
Networked Storage Architecture
Network Storage Solutions Organization
MS 5668	Hewlett-Packard, Roseville.
cbm@rose.hp.com