Re: [v6ops] Flow Label Load Balancing

[Brian E Carpenter <[e-mail]>]

I'm not Joel, but I did once spend some time grepping RFCs to find out whether
"flow" or "microflow" was the preferred term. In RFC2474, which is normative,
we have:

   Microflow: a single instance of an application-to-application flow of
   packets which is identified by source address, destination address,
   protocol id, and source port, destination port (where applicable).

But in the flow label work we explicitly avoided being that precise, and
did not use the term "microflow". There might be some load balancing
scenarios where you want a broader definition, even including bidirectional
flows. There are expired drafts on that topic:
draft-tarreau-extend-flow-label-balancing
draft-wang-6man-flow-label-reflection

Regards
   Brian

On 26-Nov-20 11:13, Tom Herbert wrote:
> Joel, is there a normative definition of a flow?
> 
> On Wed, Nov 25, 2020, 1:27 PM Joel M. Halpern <[e-mail] <mailto:[e-mail]>> wrote:
> 
>     No, as Brian says, there are escape clauses in the flow definitions.
> 
>     But changing the flow label due to traffic problems does not correspond
>     to packets being in actually different flows.
>     If one were using UDP, and mixing loss sensitive packets with loss
>     insensitive packets for a special application, sure, one could use two
>     flow labels.  But that is not what you are describing.
> 
>     Yours,
>     Joel
> 
>     On 11/25/2020 3:05 PM, Tom Herbert wrote:
>     > Joel,
>     >
>     > Is there an RFC that clearly and unambiguously states that a host MUST
>     > use the same flow label for the lifetime _and_ clearly defines exactly
>     > what a flow is with respect to such a requirement (for instance, how
>     > would you define a flow and enforce such a requirement in UDP? IPsec?
>     > other encapsulations?). If there is such a requirement then we'll
>     > change the code to be conformant.
>     >
>     > Tom
>     >
>     > On Wed, Nov 25, 2020 at 12:39 PM Joel M. Halpern <[e-mail] <mailto:[e-mail]>> wrote:
>     >>
>     >> This kind of thing is why, as I understand it, MPTCP has discovery
>     >> mechanisms ot know if both sides use it, and can select alternative
>     >> addresses for communication.
>     >>
>     >> Trying to guess flow labels that might avoid a problem because it might
>     >> be an ECMP problem, is just flailing about.  Not a good design for
>     >> operational protocols.
>     >>
>     >> And in general, designing protocols around "I know exactly what is going
>     >> on"  (the requirement for what you describe that goes well beyond just
>     >> "limited domains") is also a recipe for failure.
>     >>
>     >> The Flow Label RFCs are actually very explicit that a flow label is
>     >> supposed to be stable for the life of the flow.  Otherwise, it isn't a
>     >> flow label.
>     >>
>     >> Yours,
>     >> Joel
>     >>
>     >> On 11/25/2020 2:35 PM, Tom Herbert wrote:
>     >>> Hi Fernando, comments in line...
>     >>>
>     >>> On Wed, Nov 25, 2020 at 12:13 AM Fernando Gont <[e-mail] <mailto:[e-mail]>> wrote:
>     >>>>
>     >>>> Hi, Tom,
>     >>>>
>     >>>> On 24/11/20 16:43, Tom Herbert wrote:
>     >>>> [....]
>     >>>>> Modulating the flow label is a means to affect the routing of packets
>     >>>>> through the network that uses flow labels as input to the ECMP hash.
>     >>>>
>     >>>> What's the point?
>     >>>>
>     >>>> 1) You cannot tell *if* the FL is being used.
>     >>>>
>     >>> Generally true, but in a limited domain this information could be
>     >>> discerned. I'd note that it's also generally true that we don't know
>     >>> if there is a load balancer or stateful firewall in the path that
>     >>> requires consistent routing, but in a limited domain we could know
>     >>> that also.
>     >>>
>     >>>> 2) Changing the FL does not necessarily mean that packets will employ a
>     >>>> different link.
>     >>>
>     >>> It's an opportunistic mechanism. If a connection is failing and we get
>     >>> a better path that fixes it by simply changing the flow label then
>     >>> what's the harm?
>     >>>
>     >>>>
>     >>>> 3) If the network is failing, shouldn't you handle this via routing?
>     >>>>
>     >>> Sure, but then that requires an out of band feedback loop from a TCP
>     >>> implementation to the network infrastructure to indicate there is a
>     >>> problem and then the network needs to respond. That's significant
>     >>> infrastructure and higher reaction time than doing something in TCP
>     >>> and IP. Think of modulating the flow label is an inexpensive form of
>     >>> source routing within a limited domain that doesn't need any
>     >>> infrastructure or heavyweight protocols or something like segment
>     >>> routing.
>     >>>
>     >>>>
>     >>>>
>     >>>>> The basic idea is that the flow label associated with a connection is
>     >>>>> randomly changed when the stack observes that the connection is
>     >>>>> failing (e.g. and an RTO). There is nothing in the specs that prevents
>     >>>>> this since the source is at liberty to set the flow label as it sees
>     >>>>> fit.
>     >>>>
>     >>>> The FL is expected to remain constant for the life of a flow. A
>     >>>> retransmitted packet is part of the same flow as the
>     >>>> originally-transmitted packet. So this seems to be contradicting the
>     >>>> very specification of the FL.
>     >>>>
>     >>>> For instance, If a RTO for a flow causes the FL to change, then one may
>     >>>> possibly argue that the FL is not naming/labeling what is said/expected
>     >>>> to be anming/labeling.
>     >>>
>     >>> Specifically, RFC6437 states:
>     >>>
>     >>> "It is therefore RECOMMENDED that source hosts support the flow label
>     >>> by setting the flow label field for all packets of a given flow to the
>     >>> same value chosen from an approximation to a discrete uniform
>     >>> distribution."
>     >>>
>     >>> So that is clearly a just recommendation, and not a requirement (and
>     >>> definitely not a MUST). Furthermore, RFC6437 states:
>     >>>
>     >>> "A forwarding node MUST either leave a non-zero flow label value
>     >>> unchanged or change it only for compelling operational security
>     >>> reasons as described in Section 6.1."
>     >>>
>     >>> So there's no guarantee in the protocol specs that flow labels are
>     >>> consistent for the life of the connection, which means that the
>     >>> network cannot assume that and thus it would be incorrect if the
>     >>> network tried to enforce flow label consistency as a protocol
>     >>> requirement. As I said, it is prudent to try to be consistent with
>     >>> flow labels and the default behavior in Linux should be changed,
>     >>> however I do not believe there's a valid claim of non-conformance that
>     >>> motivates removal of the feature that is already deployed.
>     >>>
>     >>> Tom
>     >>>
>     >>>
>     >>>
>     >>>
>     >>>>
>     >>>>
>     >>>>
>     >>>>> The feature is useful in large datacenter networks, like
>     >>>>> pparently Facebook where the patches originate, since information
>     >>>>> discerned by TCP can opportunistically be applied to route selection.
>     >>>>> The practical issue is that there are stateful devices like firewalls
>     >>>>> that require consistent routing in the network in which case changing
>     >>>>> the flow label can confuse them. As I mentioned, the original intent
>     >>>>> was that the flow label randomization feature should be opt-in instead
>     >>>>> of on by default.
>     >>>>
>     >>>> So... where is the "source" of the packet that would be "modulating" the FL?
>     >>>>
>     >>>> Thanks,
>     >>>> --
>     >>>> Fernando Gont
>     >>>> e-mail: [e-mail] <mailto:[e-mail]> || [e-mail] <mailto:[e-mail]>
>     >>>> PGP Fingerprint: 7809 84F5 322E 45C7 F1C9 3945 96EE A9EF D076 FFF1
>     >>>>
>     >>>>
>     >>>>
>     >>>
>     >>> _______________________________________________
>     >>> v6ops mailing list
>     >>> [e-mail] <mailto:[e-mail]>
>     >>> https://www.ietf.org/mailman/listinfo/v6ops
>     >>>
> 
> 
> _______________________________________________
> v6ops mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/v6ops
>