[AFS3-std] Standardization of the AFS3 protocol

[Jeffrey Hutzelman]

On Wed, 22 Feb 2006, Jeffrey Altman wrote:

> However, this does not mean that developers can go off willy nilly
> making changes to RX, adding new service RPCs or altering the behavior
> of existing service RPCs.  There needs to be a process setup by which
> a rough consensus can be achieved on proposed changes in a forum that
> will be archived so that crucial knowledge will not be lost.

Traditionally, discussion of these sorts of changes has happened via email
and/or zephyr, with occasional face-to-face meetings in the form of
hackathons which have been held periodically since 2001.  For example, the
current mechanism for coordinating RPC and pioctl numbers exists because
of a need recognized at the first hackathon, and the ways in which those
numbers are chosen (assigning new RPC numbers starting at 0x10000, having
separate pioctl namespaces for each implementation) are the result of
discussions at the same meeting.

A large part of what we're trying to accomplish now is a process for
designing and specifying protocol changes which is more open, both to
observers and would-be participants, and which ultimately results in
any protocol changes being well-defined, so that we all know how things
are expected to behave.  This is critical in orde to know when we make a
change in the future that it will not have bad effects.  For example, the
behavior of the VolSetInfo RPC has now changed more than once in ways that
are not backward-compatible, such that older clients may change fields
they did not intend to change.  Part of the goal is to avoid this sort of
lossage in the future.

> The afs3-standardization@openafs.org mailing list has been created to
> be the discussion forum.  At the present time a web interface for
> subscribing can be found at
>
> http://michigan-openafs-lists.central.org/mailman/listinfo/afs3-standardization

Once the mailing list transition is complete, michigan-openafs-lists will
become lists.openafs.org, and this list will appear alongside all the
others.  In the meantime, please bear with us.

-- Jeff