Re: iSCSI: AHS drop bit

["Julian Satran" <Julian_Satran@il.ibm.com>]

Mallikarjun,

The bit was really meant to enable dropping the additional header and not
the whole request if ignored but rejecting the whole request.
We don't have any use for it but some "vendor specific" extensions may.

I suggest we change the wording of non-ISCSI non-iSCSI/vendor specific

Julo


                                                                                                  
                    "Mallikarjun                                                                  
                    C."                  To:     ips <ips@ece.cmu.edu>                            
                    <cbm@rose.hp.c       cc:                                                      
                    om>                  Subject:     iSCSI: AHS drop bit                         
                    Sent by:                                                                      
                    owner-ips@ece.                                                                
                    cmu.edu                                                                       
                                                                                                  
                                                                                                  
                    14-12-01 22:48                                                                
                    Please respond                                                                
                    to cbm                                                                        
                                                                                                  
                                                                                                  



Julian,

Section 3.2.2.1 says the following on the AHS drop bit -
           bit 7 - Drop Bit - if set to 1 this AHS may be ignored if not
           understood; if set to 0 this AHS must be rejected if not
           understood.

- I suspect it is the entire command that was meant to be failed, not
  just the AHS...

- If it is a SCSI operation that is being failed, I suggest that
  the target should end the command with an appropriate SCSI CHECK
  CONDITION - so the command sequence window can advance at the
  transport level (and the inability of the target to support say
  extended CDB is known end-to-end, to prevent fruitless SCSI retries).

- This bit may be useful for AHS code "Non-iSCSI extensions", but
  I could not see how it can be used for the defined iSCSI operations.
  If it is indeed so, I would actually suggest dropping this feature.

Regards.
--
Mallikarjun


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