RE: ISCSI: Marking PDUs.

["Douglas Otis" <dotis@sanlight.net>]

David,

If it is easy to sync with the PDU, then there is less of an argument for
having a Urgent Pointer for an analyzer.  Should there be a mode where PDU
headers are compressed?

Doug

> Doug,
>
> > A protocol that allows a simple random scan to locate critical
> information
> > is less secure than one that remains obscure.
>
> I really wish you'd take hints ... a simple random scan
> will have no problem picking SCSI CDBs and PDUs out of
> an iSCSI stream without the Urgent mechanism, and
> figuring out what the data is.  This particular obscurity
> property of not knowing the alignment is so weak as to
> be not worth the time spent discussing it (e.g., after
> a pause in traffic, it's a good bet that the next TCP
> segment has an iSCSI CDB or data PDU at its start),
> and I again suggest leaving this rathole alone.
>
> --David
>
> ---------------------------------------------------
> David L. Black, Senior Technologist
> EMC Corporation, 42 South St., Hopkinton, MA  01748
> +1 (508) 435-1000 x75140     FAX: +1 (508) 497-8500
> black_david@emc.com       Mobile: +1 (978) 394-7754
> ---------------------------------------------------
>
>