[Css-csts] FW: missing negative Bind-Return

[Martin Götzelmann]

Dear WG members,
 
I cannot recall any discussion related to a negative BIND return and I do not believe that removal of this from the state table has been intentional. Obviously there are a number of good reasons to respond with a negative BIND return as identified in section 3.4.2.2.2. What we did agree on is that a negative UNBIND return must never be sent, and that is correctly reflected in the state table and in section 3.5.1.
 
However, when reviewing section 3.5.1, I noticed that the reason for rejecting an UNBIND invocation is limited to the detection of a duplicate Invoke-ID (see 3.5.1.2: "In case the Service Provider cannot accept the UNBIND invocation (i.e. 'duplicate Invoke-ID'), it shall peer-abort the association"). I think this is too limited as an UNBIND invocation will have to be rejected also when the service instance sate is BOUND.ACTIVE. It might be better to add a note on when UNBIND is not acceptable instead of just mentioning 'duplicate Invoke-ID'. As a minimum, "i.e." would have to be changed to "e.g.".

Regards, Martin
 
 
________________________________

From: css-csts-bounces at mailman.ccsds.org [mailto:css-csts-bounces at mailman.ccsds.org] On Behalf Of John Pietras
Sent: 01 March 2010 18:17
To: Ray, Timothy J. (GSFC-5830); css-csts at mailman.ccsds.org
Subject: RE: [Css-csts] FW: missing negative Bind-Return



Tim,

I did a quick check, and the negative Bind return disappeared at or shortly after the Berlin meeting. That is, it *was* in version 0.15 (the version used as input in Berlin), but it was removed in version 0.16 (which was issued in early December). The minutes of the Berlin meeting don't mention this particular change: for Association Control, the minutes say only "detailed review". 

 

I wasn't present for that part of the Framework review, but perhaps someone who was might remember why the negative Bind return was deleted from the state table.

 

John

 

________________________________

From: css-csts-bounces at mailman.ccsds.org [mailto:css-csts-bounces at mailman.ccsds.org] On Behalf Of Ray, Timothy J. (GSFC-5830)
Sent: Monday, March 01, 2010 11:17 AM
To: css-csts at mailman.ccsds.org
Subject: [Css-csts] FW: missing negative Bind-Return

 

Dear WG members,

 

I don't want the issue raised in this email (below) to fall through the cracks, so I'm resending it.  

 

Best regards,

Tim

 

From: Ray, Timothy J. (GSFC-5830) 
Sent: Tuesday, December 01, 2009 11:29 AM
To: css-csts at mailman.ccsds.org
Subject: missing negative Bind-Return

 

Hello all,

 

Section 3.4.2.2.2 of our Framework specification contains a list of diagnostics that can be associated with a negative Bind-Return.  However, the Provider state table for the Association Control procedure does not include any possibility of sending a negative Bind-Return.  

 

Best regards,

Tim


______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email 
______________________________________________________________________