[Css-csts] URGENT: Discrepancy in the SFW regarding the NOTIFY invocation

Ray, Timothy J. (GSFC-5830) timothy.j.ray at nasa.gov
Fri Feb 27 15:44:12 UTC 2015


Dear Wolfgang and John,

Yes, what you propose makes sense to me, and has my approval.

Best regards,
Tim

On 2/27/15, 4:30 AM, "Wolfgang Hell" <Wolfgang_._Hell at t-online.de> wrote:

>Dear CSTS colleagues,
>
>While trying to close all RIDs and other comments regarding the SFW,
>John and I detected a discrepancy between some requirements in the main
>part of the SFW document and the NOTIFY invocation type specification in
>annex E. The SFW contained in some place requirements postulating that a
>notification shall report the service that triggered the reported event.
>Taking a look at the type specification, it turns out that this
>information cannot be conveyed except if one would introduce a related
>parameter either by changing the invocation type or by means of
>extension. However, in general the reporting of the service is not
>needed as the user can be safely assumed to know which service is being
>used when the notification comes in. There might be the case that a
>service has to send notifications that actually triggered due to other
>services being used in which case one would need the extra parameter
>identifying the service with which the notification is associated. It
>turns out that not even the MD CSTS has such need. Therefore John and I
>favor the following approach:
>
>- Any requirement postulating that the service triggering an event is
>reported in the notification are removed from the SFW.
>- The type specifications associated with the NOTIFY invocation are left
>alone, i.e. no additional parameter is introduced.
>- Should at some point in the future the need arise to accommodate an
>additional parameter reporting the service triggering the event, this
>can be achieved by means of extension.
>
>Given that we are proposing to remove an originally specified
>capability, we are seeking the WG consent on this point. A reply at your
>earliest convenience will be appreciated since John and I are trying to
>wrap up the SFW asap.
>
>Best regards,
>Wolfgang
>
>_______________________________________________
>Css-csts mailing list
>Css-csts at mailman.ccsds.org
>http://mailman.ccsds.org/cgi-bin/mailman/listinfo/css-csts





More information about the CSS-CSTS mailing list