RE: iSCSI : Digest Error recovery causes data corruption

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

Pierre,

A digest error is not seen by TCP.  TCP knows nothing about the content of
the byte stream and depends only on the TCP 16-bit checksum.  As such, TCP
will not retransmit.  There are two aspects to the iSCSI transport layer
added to TCP.  It should allow a mechanism to deal with transport related
errors independent of SCSI and it allows multiple TCP connections to be
shared as a single connection.  As the client or initiator share a single
sequence count, the handshake and selective acknowledgement is effectively
independent of any connection.  With an assumption the server or target
response does not require strict ordering over multiple connections, each
server connection enjoys an independent sequence.  It should be through
these sequences digest or header errors are recovered.  The iSCSI header
should be seen separate from SCSI.  Presently this is not the case and the
cause of complexity in my opinion.

Doug


> Hello,
>
>
> I would invite you to revisit the Mike Krause mail about
> "iSCSI Data Integrity - Digests".
>
> If Alternative 1 or 2 is adopted, when a digest error (CRC)
> occurs, the segment could be discarded, and TCP will
> retransmit. For software implementation with an hardware
> assist to calculate the CRC, (same kind of the one that exists
> now to calculate the checksum now), if a CRC error is detected
> it could be assimilated as a checksum error.
> Hence there is no need to overload the iSCSI protocol
> to deal with digest errors, because the iSCSI layer could not
> see them.
>
> Why to do complicate when simple an efficient can be achieved?
>
> Regards,
>
> Pierre
>
>