Re: iSCSI: error recovery

[David Peterson <dap@cisco.com>]

Couple of comments below...dap

> /<JS>
>
> The whole point about data numbering is to allow a target to discard
> read-data when read-data is hard to recover (as in tapes).  Once data
is
> acked it is discarded. A target getting a command in restart mode will

> resend whatever data it has still buffered (not acked) and continue
from
> there. The initiator would not e any wiser because it is not supposed
to
> scoreboard.
>
> <JS>/

Don't believe issuing a NOP to ack the read data is useful using a
guaranteed
delivery such as TCP, it only adds overhead.
For a current tape target implementation I would expect it to hold on to
the
current record until the next command is received which implies the
status was
received at the host. Given the existing iSCSI tools, for the retried
read
command scenario I would expect the target to resend the data at offset
0 (i.e.
the start of the record).

>
> I assume by the time iSCSI is implemented, tapes will be behave just
like
> discs.  Tapes are already moving to that model and there is a question
of
> whether the "fc-tape" error recovery in FCP2 will be needed when it is

> completed.
>

It is true we are moving/reviewing the explicit state tape drive model,
but
there will still be plenty of legacy tape devices around.









begin:vcard 
n:Peterson;David
tel;cell:612-802-3299
tel;work:763-398-1007
x-mozilla-html:FALSE
org:Cisco Systems, Inc.
adr:;;6450 Wedgwood Road;Maple Grove;Minnesota;55311;
version:2.1
email;internet:dap@cisco.com
title:Lead Architect - Standards Development
fn:David Peterson
end:vcard