ISC DHCP 3.0.4rc1 is now available!

David W. Hankins David_Hankins at isc.org
Wed Apr 19 19:53:34 UTC 2006


On Wed, Apr 19, 2006 at 08:53:50PM +0200, Joerg Mayer wrote:
> On Wed, Apr 19, 2006 at 08:41:03AM -0700, David W. Hankins wrote:
> > As for documenting changes to ISC DHCP failover implementation that
> > represent protocol changes, there are changes queued for feature release
> > (and it looks like I'm going to get to commit them to HEAD today as
> > the review process has completed) which you'll see in 3.1.0.  They
> > are not backwards compatible to any 3.0.x version by necessity (even
> > the CONNECT message protocol number has changed!).
> > 
> > But they're already documented in draft revision 12 of the failover
> > draft...we're just syncing up with that.
> 
> In case you have some traces with the new version that you can share
> with me, I can update the Ethereal dissector for the dhcp-failover
> protocol.

Oh, you're THAT guy.  That's going to be problematic.

One of the errors in draft revision 12 is that it documents a
'recover-wait' state, but it doesn't number this state in the table
of state->protocol-numbers (in section 12.24).

So I'd like there to be a -13, so we know what that number is.  I've
been meaning to write the authors mail but I keep finding something
else to do instead.  I need to send mail about a lot more than that
is part of the trouble.

I think it's unlikely the other states will be renumbered when the
author allocates a number to it.  But since the message numbers
were alphabetized and renumbered as a consequence, I have to accept
that as a possibility.

So that's one issue: to update the dissector now without that
number, means we know we're going to modify that yet again.

I don't know about you, but I like to do things once.  Fewer context
switches.  So it's up to you there.


The other is that it's just not going to be easy to tell which draft
revision the implementation is written to for someone from the outside.

As I said the CONNECT/CONNECTACK messages were renumbered...this is
why we can't be backwards compatible even if we wanted to, the protocol
version number is contained within the connect messages!

If you come into the conversation midway, you have no frame of
reference to tell...you have to see that the first message was
numbered as a draft-7 connect or a draft-12 connect.

So I'm not sure where to start except to pick one of the two.  And
since -7 is presently more widely deployed than -12, maybe it should
wait?  Your call again, but the issue disturbs me.


The final one is that although we've completed review process, we
haven't completed testing.

So I admit to not having a system I can go packet trace right now.

But I'll record your email address in the ticket as being interested
for this purpose when we get to that point again.

-- 
David W. Hankins		"If you don't do it right the first time,
Software Engineer			you'll just have to do it again."
Internet Systems Consortium, Inc.		-- Jack T. Hankins


More information about the dhcp-users mailing list