Static vs SLAAC - Static expected to be preferred?

[Mark Smith]

Hi,

In this earlier discussion in November 2010 -

http://www.gossamer-threads.com/lists/nsp/ipv6/26148

Geert Hendrickx described issues with SLAAC addresses being used
in preference to static addresses for outbound source addresses under
Linux. It seems that when there are multiple "non-deprecated" addresses,
i.e. those with non-zero preferred lifetimes, Linux is preferring the
most recently configured ones. The SLAAC addresses would usually be the
ones chosen as RAs would continue to periodically "re-configure" them.

RFC3484, "Default Address Selection for Internet Protocol version 6
(IPv6)", doesn't seem to specify this sort of behavior. It would seem
to be covered by the following statement - "If the eight rules fail to
choose a single address, some unspecified tie-breaker should be used."

RFC3484 is being revised, and I've suggested the following, which would
ultimately result in a static address with infinite lifetimes being
preferred over SLAAC addresses (or any other type address) with
non-infinite lifetimes -

"I'd like to suggest the a few rules to be inserted between Rules 3 and
4 of RFC3484 -

   Rule 3:  Avoid deprecated addresses.
   The addresses SA and SB have the same scope.  If one of the two
   source addresses is "preferred" and one of them is "deprecated" (in
   the RFC 2462 sense), then prefer the one that is "preferred."

   Rule 4:  Prefer home addresses.
   If SA is simultaneously a home address and care-of address and SB is
   not, then prefer SA.  Similarly, if SB is simultaneously a home
   address and care-of address and SA is not, then prefer SB.
   If SA is just a home address and SB is just a care-of address, then
   prefer SA.  Similarly, if SB is just a home address and SA is just a
   care-of address, then prefer SB.

Suggested rule -

Rule X: Prefer greatest preferred lifetime
   If the addresses SA and SB both have non-zero value preferred
   lifetimes (are "non-deprecated"), prefer the address with the
   greatest value preferred lifetime."

Here is the discussion thread  -

http://www.ietf.org/mail-archive/web/ipv6/current/msg13866.html

It has been suggested that in this SLAAC vs static address scenario,
it's not a common expectation that static addresses (with the default
infinite preferred and valid lifetimes) would always be used in
preference to the SLAAC addresses. 

I disagree with that, as when I specifically choose to go to the effort
of overriding an automated configuration mechanism with a static one,
then I intend for and expect the static configuration to be used in
preference to the automated configuration. Static routes in a
router overriding dynamic routing protocol routes by default would be
a similar example where this is standard and expected behaviour. 

I said I'd ask on this mailing list what the common expectation is. So
if a host has addresses it has acquired via SLAAC (or less likely
stateful DHCPv6), with non-infinite lifetimes, and you've
administratively configured a static IPv6 address on the same interface
(with default infinite preferred and valid lifetimes), would you expect
the static address to always be used as a source address for outbound
connections?



Thanks very much,
Mark.