RE: description of recovery mechanisms

[julian_satran@il.ibm.com]

David,

If you mean a complete formal description (including state charts) the last
complete description I've seen was the SNA manual and that was many years
ago (and included hundreds of pages on recovery alone).  I've tried to go
into enough detail so that an implementation should be possible and given
the amount of options possible I am afraid that either we give them all and
that is an inordinate (and wasteful) amount of work. None of the major
TCP/IP protocols have it and they have all added details in a pragmatic
fashion.

The same thing can be claimed for the whole protocol  not only for
recovery.

If anybody has a specific suggestion about something that is missing or a
another structure for the recovery chapter (in addition or instead the
two-dimensional approach that I have attempted) and would not require me to
write detailed recovery pseudocode (I could be amenable to write a rough
pseudocode skeleton for the recovery) I am ready to listen.

Regards,
Julo

Black_David@emc.com on 09/03/2001 02:33:52

Please respond to Black_David@emc.com

To:   someshg@yahoo.com, ips@ece.cmu.edu
cc:
Subject:  RE: description of recovery mechanisms




What Somesh is asking for is necessary.  We can
either put it in now, or discover it the hard
way later in interoperability testing with
considerably greater expenditure of time and
effort.

--David

> -----Original Message-----
> From:   Somesh Gupta [SMTP:someshg@yahoo.com]
> Sent:   Wednesday, March 07, 2001 10:24 PM
> To:     ips@ece.cmu.edu
> Subject:     description of recovery mechanisms
>
> I hope David and Julian will excuse me for using the following
> sentences from the Orlando minutes.
>
> -----------  start of quote -------------------------------
>
> - There will be a significant connection recovery write-up,
>    including details, procedures and examples added to the draft.
>
> -----------  end of quote ---------------------------------
>
>
> As an engineer, I believe that we do need detailed and thorough
> description of the usage of all the recovery tools in the
> protocol. This ensures
>
> 1. Determination that there are no holes. Presence of "holes"
>    will lead to the mechanisms not being used (but implemented)
>
>    or determine that there are no holes which will lead
>    to testing nightmares.
>
> 2. Ensure interoperability among implementations
>
> I hope this does not sound like I am asking Julian to do
> my work for me. But it is better hashed out and debated
> in one place.
>