On Thu, Apr 19, 2018 at 7:47 AM, Frank Brockners (fbrockne)
<fbroc...@cisco.com> wrote:
> Tom,
>
> -----Original Message-----
> From: Tom Herbert <t...@herbertland.com>
> Sent: Donnerstag, 19. April 2018 16:32
> To: Frank Brockners (fbrockne) <fbroc...@cisco.com>
> Cc: Tianran Zhou <zhoutian...@huawei.com>; Shwetha Bhandari (shwethab)
> <shwet...@cisco.com>; Mickey Spiegel <mspie...@barefootnetworks.com>; NVO3
> <nvo3@ietf.org>; Service Function Chaining IETF list <s...@ietf.org>; IETF
> IPPM WG <i...@ietf.org>
> Subject: Re: [ippm] [Int-area] encapsulation of IOAM data in various
> protocols - follow up from WG discussion in London
>
> On Wed, Apr 18, 2018 at 12:51 AM, Frank Brockners (fbrockne)
> <fbroc...@cisco.com> wrote:
>> Tom,
>>
>> inline... ("...FB")
>>
>> -----Original Message-----
>> From: Tom Herbert <t...@herbertland.com>
>> Sent: Dienstag, 17. April 2018 16:23
>> To: Frank Brockners (fbrockne) <fbroc...@cisco.com>
>> Cc: Tianran Zhou <zhoutian...@huawei.com>; Shwetha Bhandari (shwethab)
>> <shwet...@cisco.com>; Mickey Spiegel <mspie...@barefootnetworks.com>;
>> NVO3 <nvo3@ietf.org>; Service Function Chaining IETF list
>> <s...@ietf.org>; IETF IPPM WG <i...@ietf.org>
>> Subject: Re: [ippm] [Int-area] encapsulation of IOAM data in various
>> protocols - follow up from WG discussion in London
>>
>> On Tue, Apr 17, 2018 at 12:51 AM, Frank Brockners (fbrockne)
>> <fbroc...@cisco.com> wrote:
>>>
>>> Hi Tianran,
>>>
>>> Tom's note already includes the hint: You'll add IOAM data to the
>>> protocol/layer that you're interested in monitoring. Again using Geneve
>>> over IPv6 as an example:
>>> * If you're interested in the overlay, i.e. Geneve (e.g. timestamping
>>> the packet when it enters and exists the tunnel) - you'd add IOAM
>>> data to Geneve
>>> * If you're interested in the underlay, i.e. IPv6 (e.g. you'd like to
>>> understand which path packets take in the v6 network) - you'd add
>>> IOAM data to IPv6
>>> * If you're interested in both, then you'd add IOAM data to Geneve
>>> and
>>> IPv6
>>>
>> Frank,
>>
>> In that case why not just use a hop-by-hop option for measuring the underlay
>> and a destination option for measuring the overlay? The advantage is that
>> this works _any_ IP encapsulation method or any IP protocol for that matter.
>> I don't believe adding ippm to every encapsulation protocol is
>> straightforward: e.g.
>> draft-brockners-ippm-ioam-geneve describe but notes the limited size of
>> header, draft-weis-ippm-ioam-gre states that a new EtherType would be needed
>> just for this purpose. This also entails additional encapsulation-specific
>> HW support also, whereas support destination and hbh options could be more
>> generic.
>>
>> ...FB: There are quite a few deployment examples, such as overlay VPN
>> services, where you don't have access to the underlay (e.g. IPv6) - but do
>> control the overlay and desire insights into your overlay using IOAM. Hence
>> the need carry IOAM data along the overlay encapsulation.
>>
> Frank,
>
> I'm not sure I follow your argument. Maybe examining some of the scenarios
> would help (here in in the draft):
>
> Consider that an end host sources a simple TCP packet. So headers look like
>
> IP | TCP
>
> If they wish to do ippm for end-to-end measure, a destination option can be
> used, so packet looks like
>
> IP | ippm_DO | TCP
>
> Similarly, if they want to do per hop measurement they that could also use
> and appropiate HBH option
>
> IP | ippm_HBH | ippm_DO | TCP
>
> Now consider that a device in the network is tunneling the packet with
> encapsulation.
>
> Basic encapsulated packet looks like
>
> IP | encap_hdr | IP | TCP
>
> Measurements across the tunnel (ingress to egress point) could be done
>
> IP | ippm_DO | encap_hdr | IP | TCP (#1)
>
> Measuring underday then would be
>
> IP | ippm_HBH | encap_hdr | IP | TCP
>
> And again both can be done simulataneously
>
> IP | ippm_HBH | ippm_DO | encap_hdr | IP | TCP
>
> Note also that in simple ipip encasulation the encap_hdr would be null when
> the DO EH hae next protocol as IP. So that case would look like:
>
> IP | ippm_HBH | ippm_DO | IP | TCP
>
> Now if the ippm is in the encapsulation, then to measure across the tunnel
> packet (ingress to egress) looks something like
>
> IP | encap_hdr | ippm_data | IP | TCP (#2)
>
> Where ippm_data is either a field or TLV of the encapsulation (like would be
> in Geneve) or it's a layer in a header chain (like in the GRE proposal).
>
> If underlay is also measured also then that gives
>
> IP | ippm_HBH | encap_hdr | ippm_data | IP | TCP '
> #1 and #2 above are the comparable cases.
>
> #2 (ippm in the encapsulation) measures from encapsulation to decapsulation
> of the tunnel. This corresponds to the tunnel ingress and egress points. In
> #1, the DO is used to measure between tunnel ingress and egress. Processing
> of the encapsulation header is also done at tunnel ingress and egress so that
> coincides with where DO is processed. In an implementation, the destination
> option can even be passed the encpasulation layer if desired so that it can
> be processed as though it were part of the encapsulation.
>
> So from this, I don't see any functional difference between using the ippm DO
> in the outer IP header of an encapsulation and putting the ippm data in the
> encapsulation. There might be an agument that this is needed to support IPv4,
> but even so I'm not sure that justifies retrofitting every IETF defined
> encapsulation protocol to support ippm.
>
> Also, I think how ippm works with simple IPIP encapsulation should be thought
> out since that is still probably the most common encapsulation deployed.
> There is no way to extend IPIP (doesn't even have encap
> header) and adding a new IP protocol code point (similar to
> draft-weis-ippm-ioam-gre where a new EtherType for this purpose is
> proposed) is pretty unlikely to be accepted. So in this case of IPIP, it
> seems like the only answer is to use DO or HBH options, but as I mentioned I
> don't see that is any disadvantage or less functional of a solution (except
> for lacking IPv4 support).
>
> ...FB: Thanks. I do follow your logic, though ignoring v4 isn't really an
> option - hence the need to get IOAM data into the tunnel encaps - at least
> into the most popular ones.
>
Frank,
It's not a matter of ignoring IPv4, it's that IPv4 doesn't have the
capabilities that seem to be needed. Unless the intent is to resurrect
IPv4 options, there is no way to add ippm to IP packets other than by
encapsulation. That means that the cases I listed above that don't
involve encapsulation don't exist in IPv4. Also, again assuming IPv4
options are off the table, there is no way in IPv4 to perform ippm
hop-by-hop measurement since there is no allowance in IP that
intermediate devices can change IP payloads (see the UDP port number
misinterpretation problem for example). Maybe it's time to sunset IPv4
:-)
As for "popular encapsulations", I still don't see any feasible way to
extend IPIP for ippm. Also, I believe the attempt to make GRE
extensible as described in draft-weis-ippm-ioam-gre is problematic in
several ways-- I will send comments on that draft.
Tom
> Frank
>
> Tom
>
>> Frank
>>
>> Tom
>>
>>> Draft draft-ietf-ippm-ioam-data-02 already mentions layering (see section
>>> 3):
>>> "Layering: If several encapsulation protocols (e.g., in case of tunneling)
>>> are stacked on top of each other, IOAM data-records could be present at
>>> every layer. The behavior follows the ships-in-the-night model."
>>>
>>> Given the discussion here, we'll add some additional text in the next
>>> revision to make things crisper (e.g. adding an example might help).
>>>
>>> Frank
>>>
>>> -----Original Message-----
>>> From: Tianran Zhou <zhoutian...@huawei.com>
>>> Sent: Dienstag, 17. April 2018 03:18
>>> To: Tom Herbert <t...@herbertland.com>
>>> Cc: Shwetha Bhandari (shwethab) <shwet...@cisco.com>; Frank Brockners
>>> (fbrockne) <fbroc...@cisco.com>; Mickey Spiegel
>>> <mspie...@barefootnetworks.com>; NVO3 <nvo3@ietf.org>; Service
>>> Function Chaining IETF list <s...@ietf.org>; IETF IPPM WG
>>> <i...@ietf.org>
>>> Subject: RE: [ippm] [Int-area] encapsulation of IOAM data in various
>>> protocols - follow up from WG discussion in London
>>>
>>> I think it's better that Frank or Shwetha can explain the multi-layer use
>>> case in detail.
>>>
>>> Tianran
>>>> -----Original Message-----
>>>> From: Tom Herbert [mailto:t...@herbertland.com]
>>>> Sent: Monday, April 16, 2018 10:40 PM
>>>> To: Tianran Zhou <zhoutian...@huawei.com>
>>>> Cc: Shwetha Bhandari (shwethab) <shwet...@cisco.com>; Frank
>>>> Brockners
>>>> (fbrockne) <fbroc...@cisco.com>; Mickey Spiegel
>>>> <mspie...@barefootnetworks.com>; NVO3 <nvo3@ietf.org>; int-area
>>>> <int-a...@ietf.org>; Service Function Chaining IETF list
>>>> <s...@ietf.org>; IETF IPPM WG <i...@ietf.org>
>>>> Subject: Re: [ippm] [Int-area] encapsulation of IOAM data in various
>>>> protocols - follow up from WG discussion in London
>>>>
>>>> On Mon, Apr 16, 2018 at 6:31 AM, Tianran Zhou <zhoutian...@huawei.com>
>>>> wrote:
>>>> > Hi Shwetha,
>>>> >
>>>> > You are talking about the outer encapsution. It is straight
>>>> > forward for the underlay to record by the header. But what about
>>>> > the overlay, i.e., inner encapsulation(e.g. geneve)? Without
>>>> > special configuration, intermediate node will not read the inner
>>>> > header, hence not be able to process IOAM.e
>>>>
>>>> Hi Tianran,
>>>>
>>>> I believe that is also not protocol conformant. Intermediate nodes
>>>> should not be processing transport layer data as this can lead to
>>>> misinterpretation and possibly silent data corruption.
>>>>
>>>> For instance, Geneve is a UDP encapsulation protocol with assigned port
>>>> 6081.
>>>> In order for an intermediate device to process the Geneve
>>>> encapsulation header it would need to look for packets with
>>>> destination port of 6081 since that is the only possible
>>>> discriminator. However, transport port numbers do not have global
>>>> meaning and hosts may use port numbers for other purposes (RFC7605
>>>> describes this). So a packet to port 6081 might be something other
>>>> than Geneve and may be misinterpreted. If a misinterpreted packet is
>>>> changed (like ippm data is written) then that would be systematic silent
>>>> data corruption.
>>>>
>>>> As far as I know, hop-by-hop options is the only protocol confirming
>>>> mechanism that allows an intermediate note to change data of packet in
>>>> flight.
>>>> Encpasulation is the only conforming mechanism that allows an
>>>> intermediate node to add data (like extension headers) to a packet in
>>>> flight.
>>>>
>>>> Tom
>>>>
>>>> > Maybe we are not synced by this overlay/underlay use case. :-)
>>>> >
>>>> > Tianran
>>>> >
>>>> >
>>>> >
>>>> > ________________________________
>>>> > Sent from WeLink
>>>> >
>>>> > 发件人： Shwetha Bhandari (shwethab)
>>>> > 收件人： Tianran Zhou<zhoutian...@huawei.com>;Frank Brockners
>>>> > (fbrockne)<fbroc...@cisco.com>;Mickey
>>>> > Spiegel<mspie...@barefootnetworks.com>;Tom
>>>> > Herbert<t...@herbertland.com>
>>>> > 抄送： NVO3<nvo3@ietf.org>;int-area<int-a...@ietf.org>;Service
>>>> > Function Chaining IETF list<s...@ietf.org>;IETF IPPM
>>>> > WG<i...@ietf.org>
>>>> > 主题： Re: [ippm] [Int-area] encapsulation of IOAM data in various
>>>> > protocols - follow up from WG discussion in London
>>>> > 时间： 2018-04-16 18:17:01
>>>> >
>>>> > Hi Tianran,
>>>> >
>>>> >> If I recall right, it is not written in the ioam data draft.
>>>> >
>>>> > Data draft is defining the data to be carried in IOAM in an
>>>> > encapsulation agnostic way, it does not specify how the
>>>> > encapsulation protocol is configured.
>>>> >
>>>> >
>>>> >
>>>> >> Yes, node by node configuration is an easy way.
>>>> >
>>>> > While it is, it does not have to be a node by node configuration.
>>>> > It can be part of the encapsulation definition.
>>>> >
>>>> > For e.g. If the encapsulation is IPv6 and if we define the data to
>>>> > be carried as HbH options, then based on the Option Type with
>>>> > highest order 2 bits set to 00 then the v6 nodes that implement
>>>> > IOAM will process the option and others will skip over.
>>>> >
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > Thanks,
>>>> >
>>>> > Shwetha
>>>> >
>>>> >
>>>> >
>>>> > From: ippm <ippm-boun...@ietf.org> on behalf of Tianran Zhou
>>>> > <zhoutian...@huawei.com>
>>>> > Date: Monday, April 16, 2018 at 2:36 PM
>>>> > To: "Frank Brockners (fbrockne)" <fbroc...@cisco.com>, Mickey
>>>> > Spiegel <mspie...@barefootnetworks.com>, Tom Herbert
>>>> > <t...@herbertland.com>
>>>> > Cc: NVO3 <nvo3@ietf.org>, "int-a...@ietf.org" <int-a...@ietf.org>,
>>>> > Service Function Chaining IETF list <s...@ietf.org>, IETF IPPM WG
>>>> > <i...@ietf.org>
>>>> > Subject: Re: [ippm] [Int-area] encapsulation of IOAM data in
>>>> > various protocols - follow up from WG discussion in London
>>>> >
>>>> >
>>>> >
>>>> > Hi Frank,
>>>> >
>>>> >
>>>> >
>>>> > If I recall right, it is not written in the ioam data draft.
>>>> >
>>>> > Yes, node by node configuration is an easy way. In the
>>>> > draft-zhou-ippm-ioam-yang, we have the “protocol-type” to indicate
>>>> > the layering.
>>>> >
>>>> > +--rw ioam
>>>> >
>>>> > +--rw ioam-profiles
>>>> >
>>>> > +--rw enabled? boolean
>>>> >
>>>> > +--rw ioam-profile* [profile-name]
>>>> >
>>>> > +--rw profile-name string
>>>> >
>>>> > +--rw filter
>>>> >
>>>> > | +--rw filter-type? ioam-filter-type
>>>> >
>>>> > | +--rw acl-name? -> /acl:acls/acl/name
>>>> >
>>>> > +--rw protocol-type? ioam-protocol-type
>>>> >
>>>> > +--rw incremental-tracing-profile {incremental-trace}?
>>>> >
>>>> > | ...
>>>> >
>>>> > +--rw preallocated-tracing-profile {preallocated-trace}?
>>>> >
>>>> > | ...
>>>> >
>>>> > +--rw pot-profile {proof-of-transit}?
>>>> >
>>>> > | ...
>>>> >
>>>> > +--rw e2e-profile {edge-to-edge}?
>>>> >
>>>> > ...
>>>> >
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > Tianran
>>>> >
>>>> > From: Frank Brockners (fbrockne) [mailto:fbroc...@cisco.com]
>>>> > Sent: Monday, April 16, 2018 4:51 PM
>>>> > To: Tianran Zhou <zhoutian...@huawei.com>; Mickey Spiegel
>>>> > <mspie...@barefootnetworks.com>; Tom Herbert <t...@herbertland.com>
>>>> > Cc: NVO3 <nvo3@ietf.org>; int-a...@ietf.org; Service Function
>>>> > Chaining IETF list <s...@ietf.org>; IETF IPPM WG <i...@ietf.org>
>>>> > Subject: RE: [ippm] [Int-area] encapsulation of IOAM data in
>>>> > various protocols - follow up from WG discussion in London
>>>> >
>>>> >
>>>> >
>>>> > Hi Tianran,
>>>> >
>>>> >
>>>> >
>>>> > IOAM is a domain specific feature (see also
>>>> > draft-ietf-ippm-ioam-data-02 sections 3 and 4), which allows an
>>>> > operator to control by means of configuration where and for which
>>>> > traffic IOAM data fields are added/updated/removed from the
>>>> > customer traffic. Using your example of Geneve over IPv6 – with
>>>> > IOAM data in both the Geneve and the IPv6 protocol, one would
>>>> > expect that the operator configures the endpoints of the Geneve
>>>> > tunnel to operate on the IOAM data in Geneve, and the IPv6 routers
>>>> > that the Geneve tunnel
>>>> traverses to operate on the IOAM data in IPv6.
>>>> >
>>>> >
>>>> >
>>>> > Frank
>>>> >
>>>> >
>>>> >
>>>> > From: Tianran Zhou <zhoutian...@huawei.com>
>>>> > Sent: Montag, 16. April 2018 10:37
>>>> > To: Frank Brockners (fbrockne) <fbroc...@cisco.com>; Mickey
>>>> > Spiegel <mspie...@barefootnetworks.com>; Tom Herbert
>>>> > <t...@herbertland.com>
>>>> > Cc: NVO3 <nvo3@ietf.org>; int-a...@ietf.org; Service Function
>>>> > Chaining IETF list <s...@ietf.org>; IETF IPPM WG <i...@ietf.org>
>>>> > Subject: RE: [ippm] [Int-area] encapsulation of IOAM data in
>>>> > various protocols - follow up from WG discussion in London
>>>> >
>>>> >
>>>> >
>>>> > Hi Frank,
>>>> >
>>>> >
>>>> >
>>>> > How does a forwarder know when and where to insert the data?
>>>> >
>>>> > In the case of Geneve over IPv6, do you mean the device need to
>>>> > scan all the protocol stack? Or just the outer encapsulation?
>>>> >
>>>> >
>>>> >
>>>> > Tianran
>>>> >
>>>> >
>>>> >
>>>> > From: ippm [mailto:ippm-boun...@ietf.org] On Behalf Of Frank
>>>> > Brockners
>>>> > (fbrockne)
>>>> > Sent: Monday, April 16, 2018 3:08 PM
>>>> > To: Mickey Spiegel <mspie...@barefootnetworks.com>; Tom Herbert
>>>> > <t...@herbertland.com>
>>>> > Cc: NVO3 <nvo3@ietf.org>; int-a...@ietf.org; Service Function
>>>> > Chaining IETF list <s...@ietf.org>; IETF IPPM WG <i...@ietf.org>
>>>> > Subject: Re: [ippm] [Int-area] encapsulation of IOAM data in
>>>> > various protocols - follow up from WG discussion in London
>>>> >
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > Tom,
>>>> >
>>>> >
>>>> >
>>>> > a quick addition to what Mickey mentioned below: What you seem to
>>>> > have in mind is what draft-ietf-ippm-ioam-data-02 refers to as “layering”
>>>> > (see section 3.), i.e. if you’re running for example Geneve over
>>>> > IPv6, then IOAM data could be encapsulated in both protocols,
>>>> > Geneve and
>>>> > IPv6 – giving you visibility into the “underlay” (IPv6) and the “overlay”
>>>> (Geneve).
>>>> >
>>>> >
>>>> >
>>>> > Frank
>>>> >
>>>> >
>>>> >
>>>> > From: ippm <ippm-boun...@ietf.org> On Behalf Of Mickey Spiegel
>>>> > Sent: Freitag, 13. April 2018 20:22
>>>> > To: Tom Herbert <t...@herbertland.com>
>>>> > Cc: NVO3 <nvo3@ietf.org>; int-a...@ietf.org; Service Function
>>>> > Chaining IETF list <s...@ietf.org>; IETF IPPM WG <i...@ietf.org>
>>>> > Subject: Re: [ippm] [Int-area] encapsulation of IOAM data in
>>>> > various protocols - follow up from WG discussion in London
>>>> >
>>>> >
>>>> >
>>>> > Tom,
>>>> >
>>>> >
>>>> >
>>>> > On Thu, Apr 12, 2018 at 10:17 PM, Tom Herbert <t...@herbertland.com>
>>>> > wrote:
>>>> >
>>>> > Mickey,
>>>> >
>>>> > Looking at these ippm drafts more closely, I have a much more
>>>> > fundamental concern.
>>>> >
>>>> > In draft-brockners-ippm-ioam-geneve-00 for instance, there is the
>>>> > text in the introduction:
>>>> >
>>>> > "In-situ OAM (IOAM) records OAM information within the packet
>>>> > while the packet traverses a particular network domain. The term
>>>> > "in-situ"
>>>> > refers to the fact that the IOAM data fields are added to the data
>>>> > packets rather than is being sent within packets specifically
>>>> > dedicated to OAM. This document defines how IOAM data fields are
>>>> > transported as part of the Geneve [I-D.ietf-nvo3-geneve]
>>>> > encapsulation."
>>>> >
>>>> > I assume this means that as packets with Geneve encapsulation
>>>> > traverse the network they are interpreted by intermediate nodes as
>>>> > being Geneve. Since Geneve is a UDP encapsulation, then the
>>>> > destination UDP port number would be used to identify packets as
>>>> > being Geneve. So an intermediate device might be looking for UDP
>>>> > packets destined to port
>>>> > 6081 (the assigned UDP port for Geneve). If my understanding is
>>>> > correct, then this is a problem.
>>>> >
>>>> > UDP port numbers do not have global meaning. An intermediate
>>>> > device may very well see UDP packets destined to port 6081 that
>>>> > are not actually Geneve. This scenario is discussed in RFC7605:
>>>> >
>>>> > "...intermediate device interprets traffic based on the port number.
>>>> > It is important to recognize that any interpretation of port
>>>> > numbers
>>>> > -- except at the endpoints -- may be incorrect, because port
>>>> > numbers are meaningful only at the endpoints."
>>>> >
>>>> > If the UDP data is modified, as the draft would imply, then
>>>> > misinterpretation may also mean silent data corruption of packets.
>>>> > A protocol that would allow this seems pretty incorrect! Note that
>>>> > this would be true also for any UDP encapsulation that the network
>>>> > tries to interpret.
>>>> >
>>>> >
>>>> >
>>>> > The intention is to allow for multiple nodes that a packet
>>>> > traverses
>>>> >
>>>> > to be able to insert IOAM node information in the same trace
>>>> > option,
>>>> >
>>>> > but leave some flexibility regarding which nodes actually do the
>>>> >
>>>> > IOAM processing and the node information. This may vary
>>>> >
>>>> > depending on the transport.
>>>> >
>>>> >
>>>> >
>>>> > In case of a tunneled encapsulation such as Geneve or VXLAN,
>>>> >
>>>> > there may still be multiple hops. For example a network may use
>>>> >
>>>> > Geneve or VXLAN, but only do L2 processing at ToRs, with L3
>>>> >
>>>> > processing done at aggregation or core switches. In this case
>>>> >
>>>> > many packets would do 2 Geneve or VXLAN hops, so the packet
>>>> >
>>>> > would contain IOAM node information from two nodes.
>>>> >
>>>> >
>>>> >
>>>> > Another example is service function chaining using Geneve or
>>>> >
>>>> > VXLAN rather than NSH.
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > I am also wondering if hop-by-hop options been considered for this
>>>> > application? Their interpretation in the network is unabiguous and
>>>> > they also have the advantage that the work with any IP protocol or
>>>> > encapsulation.
>>>> >
>>>> >
>>>> >
>>>> > IPv6 hop-by-hop options has been considered. See
>>>> >
>>>> > draft-brockners-inband-oam-transport-05. This has not yet been
>>>> >
>>>> > broken out into a separate draft.
>>>> >
>>>> >
>>>> >
>>>> > Mickey
>>>> >
>>>> >
>>>> >
>>>> >
>>>> > Thanks,
>>>> > Tom
>>>> >
>>>> >
>>>> > On Thu, Apr 12, 2018 at 3:31 PM, Mickey Spiegel
>>>> > <mspie...@barefootnetworks.com> wrote:
>>>> >
>>>> >> Tom,
>>>> >>
>>>> >> On Thu, Apr 12, 2018 at 2:46 PM, Tom Herbert <t...@herbertland.com>
>>>> >> wrote:
>>>> >>>
>>>> >>> On Thu, Apr 12, 2018 at 9:54 AM, Greg Mirsky
>>>> >>> <gregimir...@gmail.com>
>>>> >>> wrote:
>>>> >>> > Hi Frank,
>>>> >>> > thank you for sharing your points. Please find my notes
>>>> >>> > in-line and tagged
>>>> >>> > GIM>>. I believe that this is very much relevant to work of
>>>> >>> > GIM>>other
>>>> >>> > working
>>>> >>> > groups that directly work on the overlay encapsulations in the
>>>> >>> > center of the discussion and hence I've added them to the list.
>>>> >>> > Hope we'll have more opinions to reach the conclusion that is
>>>> >>> > acceptable to all.
>>>> >>> >
>>>> >>> > Regards,
>>>> >>> > Greg
>>>> >>> >
>>>> >>> > On Wed, Apr 11, 2018 at 12:02 PM, Frank Brockners (fbrockne)
>>>> >>> > <fbroc...@cisco.com> wrote:
>>>> >>> >>
>>>> >>> >> Back at the IPPM meeting in London, we discussed several
>>>> >>> >> drafts dealing with the encapsulation of IOAM data in various
>>>> >>> >> protocols (draft-brockners-ippm-ioam-vxlan-gpe-00,
>>>> >>> >> draft-brockners-ippm-ioam-geneve-00,
>>>> >>> >> draft-weis-ippm-ioam-gre-00). One discussion topic that we
>>>> >>> >> decided to take to the list was the question on whether
>>>> >>> >> draft-ooamdt-rtgwg-ooam-header could be leveraged.. After
>>>> >>> >> carefully considering draft-ooamdt-rtgwg-ooam-header, I came
>>>> >>> >> to the conclusion that the “OOAM header” does not meet the
>>>> >>> >> needs of
>>>> >>> >> IOAM:
>>>> >>> >>
>>>> >>> >> * Efficiency: IOAM adds data to live user traffic. As such,
>>>> >>> >> an encapsulation needs to be as efficient as possible. The “OOAM
>>>> >>> >> header”
>>>> >>> >> is 8
>>>> >>> >> bytes long. The approach for IOAM data encapsulation in the
>>>> >>> >> above mentioned drafts only requires 4 bytes. Using the OOAM
>>>> >>> >> header approach would add an unnecessary overhead of 4 bytes
>>>> >>> >> – which is significant.
>>>> >>> Greg,
>>>> >>>
>>>> >>> I'm missing something here. I looked at the drafts you
>>>> >>> referenced and each of them looks like the overhead for OAM is
>>>> >>> greater that four bytes. In each there is some overhead
>>>> >>> equivalent to type/length, for instance in Geneve four bytes are
>>>> >>> needed for option class, type, and length. Unless the the OAM
>>>> >>> data is zero length, I don't see how this adds up to only four bytes
>>>> >>> of overhead.
>>>> >>
>>>> >>
>>>> >> The four versus eight bytes just refers to the fields in the four
>>>> >> bytes of IOAM info, that is common to all IOAM options. Beyond
>>>> >> that, there are IOAM option specific fields. For example if doing
>>>> >> one of the IOAM trace options, there are four bytes of trace
>>>> >> option header, including the IOAM-trace-type, NodeLen, Flags, and
>>>> >> RemainingLen fields. These are followed by the node data list
>>>> >> containing the per hop IOAM information.
>>>> >>
>>>> >> In looking at the OOAM header content, it has nothing to do with
>>>> >> any of the IOAM information after the first four bytes. It
>>>> >> contains another variant of the information in the first four
>>>> >> bytes of IOAM info, spread out over eight bytes.
>>>> >>
>>>> >>>
>>>> >>> Tom
>>>> >>>
>>>> >>> >
>>>> >>> > GIM>> The difference in four octets is because OOAM Header:
>>>> >>> >
>>>> >>> > provides more flexibility, e.g. Flags field and Reserved
>>>> >>> > fields;
>>>> >>
>>>> >>
>>>> >> The flags field only has one defined flag at the moment, for a
>>>> >> timestamp block. For IOAM trace we need per hop timestamps, which
>>>> >> the timestamp block cannot address, i.e. the timestamp block is
>>>> >> redundant for
>>>> IOAM.
>>>> >>
>>>> >>>
>>>> >>> > supports larger OAM packets than iOAM header;
>>>> >>
>>>> >>
>>>> >> For IOAM purposes, 1020 octets is more than enough.
>>>> >>
>>>> >>>
>>>> >>> > is future proof by supporting versioning (Version field).
>>>> >>
>>>> >>
>>>> >> IMO, taking the first two bits of the IOAM-Type to define a
>>>> >> Version field would be a good thing. This does not require adding
>>>> >> four more bytes of overhead. 64 IOAM-Types is more than enough.
>>>> >>
>>>> >>>
>>>> >>> >>
>>>> >>> >> * Maturity: IOAM has several implementations, which were also
>>>> >>> >> shown at recent IETF hackathons – and we’re expecting
>>>> >>> >> additional implementations to be publicized soon.
>>>> >>> >> Interoperable implementations need timely specifications.
>>>> >>> >> Despite the question being asked, the recent thread on OOAM
>>>> >>> >> in the NVO3 list hasn’t revealed any implementation of the OOAM
>>>> >>> >> header.
>>>> >>> >> In
>>>> >>> >> addition, the thread revealed that several fundamental
>>>> >>> >> questions about the OOAM header are still open, such as
>>>> >>> >> whether or how active OAM mechanisms within protocols such as
>>>> >>> >> Geneve would apply to the OOAM header. This ultimately means
>>>> >>> >> that we won’t get to a timely specification.
>>>> >>> >
>>>> >>> > GIM>> May I ask which encapsulations supported by the
>>>> >>> > GIM>> implementations
>>>> >>> > you
>>>> >>> > refer to. Until very recently all iOAM proposals were to use
>>>> >>> > meta-data TLV in, e.g. Geneve and NSH. And if these or some of
>>>> >>> > these implementations already updated to the newly proposed
>>>> >>> > iOAM shim, I don't see problem in making them use OOAM Header.
>>>> >>> > Would you agree?
>>>> >>> >
>>>> >>> >>
>>>> >>> >> * Scope: It isn’t entirely clear to which protocols the OOAM
>>>> >>> >> header would ultimately apply to. The way the OOAM header is
>>>> >>> >> defined, OOAM uses a 8-bit field for “Next Prot”, the next
>>>> >>> >> protocol. Some protocols that IOAM data needs to be
>>>> >>> >> encapsulated into use 16-bits for their next protocol code points.
>>>> >>> >> See e.g.
>>>> >>> >> the GRE encapsulation – as specified in
>>>> >>> >> draft-weis-ippm-ioam-gre-00.
>>>> >>> >
>>>> >>> > GIM>> The first paragraph of the Introduction section states:
>>>> >>> > New protocols that support overlay networks like VxLAN-GPE
>>>> >>> > [I-D.ietf-nvo3-vxlan-gpe], GUE [I-D.ietf-nvo3-gue], Geneve
>>>> >>> > [I-D.ietf-nvo3-geneve], BIER
>>>> >>> > [I-D.ietf-bier-mpls-encapsulation],
>>>> and
>>>> >>> > NSH [I-D.ietf-sfc-nsh] support multi-protocol payload, e.g.
>>>> >>> > Ethernet, IPv4/IPv6, and recognize Operations, Administration, and
>>>> >>> > Maintenance (OAM) as one of distinct types. That ensures that
>>>> >>> > Overlay OAM (OOAM)packets are sharing fate with Overlay data
>>>> >>> > packet
>>>> >>> > traversing the underlay.
>>>> >>> > I'm updating the OOAM Header draft and along with cleaning
>>>> >>> > nits will update reference to GUE. I think that the list and
>>>> >>> > the statemnt are quite clear in identifying the scope of
>>>> >>> > networks that may benefit from using not only common OOAM
>>>> >>> > Header but common OOAM mechanisms, e.g. Echo Request/Reply.
>>>> >>> >
>>>> >>> >> With the above in mind, I’d suggest that the WG moves forward
>>>> >>> >> with specific definitions for encapsulating IOAM data into
>>>> >>> >> protocols – per the above mentioned drafts.
>>>> >>> >>
>>>> >>> >>
>>>> >>> >>
>>>> >>> >> Regards, Frank
>>>> >>> >>
>>>> >>> >>
>>>> >>> >> _______________________________________________
>>>> >>> >> ippm mailing list
>>>> >>> >> i...@ietf.org
>>>> >>> >> https://www.ietf.org/mailman/listinfo/ippm
>>>> >>> >>
>>>> >>> >
>>>> >>> >
>>>> >>> > _______________________________________________
>>>> >>> > Int-area mailing list
>>>> >>> > int-a...@ietf.org
>>>> >>> > https://www.ietf.org/mailman/listinfo/int-area
>>>> >>> >
>>>> >>>
>>>> >>> _______________________________________________
>>>> >>> ippm mailing list
>>>> >>> i...@ietf.org
>>>> >>> https://www.ietf.org/mailman/listinfo/ippm
>>>> >>
>>>> >>
>>>> >
>>>> >
_______________________________________________
nvo3 mailing list
nvo3@ietf.org
https://www.ietf.org/mailman/listinfo/nvo3
