Re: draft-mmm-bfd-on-lags: IPv4 _and_ IPv6 ?

[Marc Binderberger <[e-mail]>]

Hello Jeff et al.,

is the following text achieving what we want?


----snip----snap----snip----snap----
5.  Detecting a member link failure

   When a micro BFD session goes down then this member link MUST be
   taken out of the LAG L2 load balance table(s).

   In case an implementation has separate load balance tables for IPv4
   and IPv6 then if both an IPv4 and IPv6 micro session exist for a
   member link an implementation MAY remove the member link from the
   load balance table only that matches the address family of the
   failing BFD session.  If for example the IPv4 micro session fails but
   the IPv6 micro session stays up then the member link MAY be removed
   from the IPv4 load balance table only but remains forwarding in the
   IPv6 load balance table.
----snip----snap----snip----snap----


And similar for the part where we describe adding a member link to the table.


Regards, Marc




On 2012-09-12, at 16:04 , Jeffrey Haas wrote:

> On Wed, Sep 12, 2012 at 03:57:14PM +0200, Marc Binderberger wrote:
>>> The implication is that if IPv4 drops, but IPv6 stays up, that the LMM
>>> should remove *only* the impacted address family from use.
>> 
>> ... if the implementation can differentiate between IPv4 and IPv6 in the load balance. I have no intention to make such an AF-aware load-balance table a requirement. The fact that either an IPv4 or an IPv6 BFD failure can remove the link for all traffic is understood by (my) customers and acceptable.
> 
> Agreed - we don't want to force implementations to have unnecessary
> abstractions.  (We're not OSI.)  But...
> 
>> But with such an "if an implementation's distribution algorithm can differentiate between IPv4 and IPv6 then ..." condition I can add something to the draft.
> 
> Exactly.  Let's be thorough.
> 
>> Has anyone implemented such an AF-ware load-balancer?  Just for my curiosity.
> 
> FWIW, this is one of those "I'm not concerned about this year's
> implementations.  It's next year's that I'm worried about" issues.
> 
> Prophylactic standardization, if you will.
> 
> -- Jeff
> 

--
Marc Binderberger           <[e-mail]>