[Sls-slp] FW: [EXTERNAL] RE: Is AOS 5 backward compatible with AOS 4?

Tue Jul 28 13:52:46 EDT 2026

Dear SLP WG,

A sticky backward compatibility issues has raised its head, concerning the Version 5 AOS Blue book (732.0-B-5) and previous versions (see thread below concerning LNIS).
Attached is a proposed NOTE I would like to amend to the AOS Version 5 blue book as a technical corrigendum to clarify the meaning of 'all ones' and 'all ones minus one' values of the First Header Pointer (FHP) when dealing with implementations that need to talk to previous (V4) and the current version.

Your comments are most welcome.

Best regards
Greg
SLP WG chair

From: KAZZ, GREG J. (MSFC-HP27)[MOSSI2]
Sent: Tuesday, July 28, 2026 10:08 AM
To: Conway, Adrian E. (GSFC-452.0)[MTI SYSTEMS, INC.] <adrian.e.conway at nasa.gov>; Singh, Somendra {Simon} (GSFC-5340) <simon.singh at nasa.gov>; Daniel Pettitt <Daniel.Pettitt at esa.int>; Andrews, Kenneth S (JPL-361A)[JPL Employee] <kenneth.s.andrews at jpl.nasa.gov>; Lux, Jim (JPL-430E)[JPL Employee] <james.p.lux at jpl.nasa.gov>; Matt Cosby <matt.cosby at goonhilly.org>; Nicola Maturo <nicola.maturo at esa.int>
Cc: Parker, Joel J K (HQ-CG000) <joel.j.k.parker at nasa.gov>; Stefano Petri <Stefano.Petri at esa.int>; Verville, Jonathan P. (GSFC-5310) <jonathan.p.verville at nasa.gov>; McGraw, Shawn J (GSFC-450.0)[MTI SYSTEMS, INC.] <shawn.mcgraw at nasa.gov>
Subject: RE: [EXTERNAL] RE: Is AOS 5 backward compatible with AOS 4?

Hi Adrian.,

Good catch. Sad that it wasn't caught before in CCSDS Agency Review.

In fact, both 'all ones' as well as 'all ones minus 1' for Only Idle Data (AOS 4.1.4.2.3.4.5) are interpreted differently depending on the AOS Version Number.

Here is a suggestion: Implementors could consider creating a MIB (Managed Information Base) parameter called AOS Version Number. It's the Version 5 assets that need to know when they communicate with Version 4 assets. V5 assets could then know when they are talking to a V4 or V5 asset.

At a minimum, I will recommend that CCSDS generate a NOTE ahead of 4.1.4.2.3.4 to make the user aware of the difference in interpretation of the FHP set to 'all ones' as well as 'all ones minus 1' based upon AOS Version Number.

Regards,
Greg

From: Conway, Adrian E. (GSFC-452.0)[MTI SYSTEMS, INC.] <adrian.e.conway at nasa.gov>
Sent: Monday, July 27, 2026 1:01 PM
To: Singh, Somendra {Simon} (GSFC-5340) <simon.singh at nasa.gov>; KAZZ, GREG J. (MSFC-HP27)[MOSSI2] <greg.j.kazz at nasa.gov>; Daniel Pettitt <Daniel.Pettitt at esa.int>; Andrews, Kenneth S (JPL-361A)[JPL Employee] <kenneth.s.andrews at jpl.nasa.gov>
Cc: Parker, Joel J K (HQ-CG000) <joel.j.k.parker at nasa.gov>; Stefano Petri <Stefano.Petri at esa.int>; Verville, Jonathan P. (GSFC-5310) <jonathan.p.verville at nasa.gov>; McGraw, Shawn J (GSFC-450.0)[MTI SYSTEMS, INC.] <shawn.mcgraw at nasa.gov>
Subject: RE: [EXTERNAL] RE: Is AOS 5 backward compatible with AOS 4?

Greg,

Recently, Shawn and I had been looking for differences between AOS Issue 4 and Issue 5 (in connection with the LNIS AD6 Data Services document that we have been working on in the DS WG) and found the following counterexample to V5 backwards compatibility:

If an AOS v4 M_PDU payload only contains a middle part of an EPP packet, then the v4 M_PDU header will be composed of 5 zeros (the RSVD Spare bits)  + 11 ones (the V4 header pointer) since AOS 4.1.4.2.3.4 says that "If no Packet starts in the M_PDU Packet Zone, the First Header Pointer shall be set to 'all ones'.).  But V5 would look at this 16-bit V4 M_PDU header, interpret it as a V5 16 bit First Header Pointer,  and say that the first octet of the next EPP packet is at position 2047, which does not agree with what the AOS v4 M_PDU payload contains.

Adrian.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ccsds.org/pipermail/sls-slp/attachments/20260728/e927bbeb/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: FHP_all_ones_all_ones_minus_one.docx
Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document
Size: 14760 bytes
Desc: FHP_all_ones_all_ones_minus_one.docx
URL: <http://mailman.ccsds.org/pipermail/sls-slp/attachments/20260728/e927bbeb/attachment-0001.docx>