RE: iSCSI: aborting an immediate command with ABORT TASK

["Tony Battersby" <tonyb@cybernetics.com>]

> 1) The possible out-of-order arrival of an abort and the immediate command
upon which
> it is supposed to act
> 2) The race between an immediate command finishing normally (causing the
target
> to free the command and forget its associated ITT) and the initiator
aborting it
> 3) The immediate command to be aborted may have been discarded due to a
header digest
> error

Replying to myself, #1 should actually not happen because of the requirement
for an ABORT TASK to be issued on the same connection to which the task
being acted upon is currently allegiant.  Points #2 and #3 are still valid
though.  Another possibility is that the target is in the process of sending
a Reject PDU for the immediate command to be aborted when it receives the
ABORT TASK command, and the target implementation doesn't check the ITT from
a PDU that it has chosen to reject.

Tony