Re: [Tsv-art] Tsvart last call review of draft-ietf-detnet-architecture-08

["Black, David" <[e-mail]>]

Trying to help with these Transport issues.  I generally agree with Michael Sharf's
first two concerns.  Like Michael, I will defer to the IESG on appropriate use of
RFC 7426.

In each case, I've extracted a key piece of text to try to make the point crisp ...

[1] Minimum required level of DetNet support.

> Does "some of the DetNet capabilities
> (not necessarily all)" include e.g. vanilla traffic engineering? For instance, if a
> network node can ensure that the QoS requirements are met by leveraging
> existing standardized TE without any additional DetNet awareness, is this
> enough for deploying DetNet? Or does this node have to implement additional
> DetNet-specific functionality? Specifically, please explain if an existing DetNet-
> unaware router on the path taken by DetNet traffic can be used to forward
> DetNet traffic, or if it is mandatory that any router forwarding DetNet traffic
> indeed implements new capabilities specific to DetNet.

It's the former, as indicated by this text in Section 4.2.2 (I'm working from the -09 draft):

    In general, if a DetNet flow passes through one or more DetNet-
   unaware network nodes between two DetNet nodes providing the DetNet
   transport sub-layer for that flow, there is a potential for
   disruption or failure of the DetNet QoS.  A network administrator
   needs to ensure that the DetNet-unaware network nodes are configured
   to minimize the chances of packet loss and delay, and provision
   enough extra buffer space in the DetNet transit node following the
   DetNet-unaware network nodes to absorb the induced latency
   variations.

Perhaps a forward reference to Section 4.2.2 from the new text in the Introduction would help.

[2] Protecting the network from detnet problems, failures, etc.

> > * It is surprising that there is hardly any discussion on network robustness
> > and safety; this probably also relates to security. For instance,
> > misconfiguration or errors of functions performing packet replication could
> > severely and permanently congest a network and cause harm.
[... snip ...]
> NEW:
> Robust real-time systems require to reduce the number of possible
> failures. Filters and policers should be used in a DetNet network to
> detect if DetNet packets are received on the wrong interface, or at
> the wrong time, or in too great volume. Furthermore, filters and
> policers can take action to discard offending packets or flows, or
> trigger shutting down the offending DetNet flow, or shutting down
> the offending interface.
> 
> [ms] That does not address my concern. As I wrote already, I believe this
> document has to discuss the applicability of circuit breakers according to RFC
> 8084.

I concur with Michael's implied concern that the "should" in the NEW text
is woefully inadequate.  The provisions to prevent escaped non-elastic traffic
from causing damage elsewhere need to be mandatory particularly if DetNet
and non-DetNet networks are interconnected.  The above
text on filters doesn't appear to cover escape from the DetNet network,
and needs to do so.  Deployment of filters at DetNet nodes that drop all
DetNet traffic that tries to escape, and that drop all misdirected DetNet
traffic should address most of this problem.

Once that is done, I think a discussion of circuit breakers is needed, in the
form of Detnet service mechanisms that can shut down (cease transmission
of) a flow that is being sent entirely into the bit-bucket by a downstream filter,
due to misconfiguration, failure, etc.

Thanks, --David

> -----Original Message-----
> From: Tsv-art [mailto:[e-mail]] On Behalf Of Scharf, Michael
> Sent: Monday, October 22, 2018 5:44 PM
> To: 'János Farkas'
> Cc: [e-mail]; DetNet WG; [e-mail]
> Subject: Re: [Tsv-art] Tsvart last call review of draft-ietf-detnet-architecture-08
> 
> 
> [EXTERNAL EMAIL]
> Please report any suspicious attachments, links, or requests for sensitive
> information.
> 
> 
> Hi Janos,
> 
> Thanks for the reply. My comments are inline [ms].
> 
> I have omitted all proposed changes that work for me.
> 
> Michael
> 
> 
> -----Original Message-----
> From: János Farkas <[e-mail]>
> Sent: Wednesday, October 17, 2018 10:56 AM
> To: Scharf, Michael <[e-mail]>
> Cc: [e-mail]; [e-mail]; DetNet WG
> <[e-mail]>
> Subject: Re: Tsvart last call review of draft-ietf-detnet-architecture-08
> 
> Hi Michael,
> 
> Thank you very much for the thorough review and good comments!
> 
> Please find below in-line responses and how we plan to update the draft to
> resolve your comments.
> 
> Best regards,
> Janos
> 
> 
> On 9/29/2018 12:24 AM, Michael Scharf wrote:
> > Reviewer: Michael Scharf
> > Review result: Ready with Issues
> >
> > The document "Deterministic Networking Architecture"
> > (draft-ietf-detnet-architecture-08) defines an overall framework for
> > Deterministic Networking.
> >
> > As TSV-ART reviewer, I believe that this document has issues as
> > detailed below.
> >
> > Michael
> >
> > Major issues:
> 
> [...]
> 
> > DetNet
> > basically requires support in (almost) all network devices
> > transporting DetNet traffic. That assumption should be explicitly
> > spelt out early in the document, e.g., in the introduction.
> 
> Actually, some form of DetNet support is required from each node that
> transports DetNet traffic. For instance, DetNet flows have to be recognized in
> order not to spoil their QoS at the minimum.
> 
> [ms] I look for an clear explanation in the architecture document of what "some
> form of DetNet support" exactly means. For instance, is standard MPLS TE
> without any DetNet support good enough for an LSR on the IP/MPLS path used
> by DetNet traffic, or not? As an example, when I read the following paragraph in
> draft-ietf-detnet-dp-sol-mpls-01 ...
> 
>    Transit nodes are normal MPLS Label Switching Routers (LSRs).  They
>    are generally unaware of the special requirements of DetNet flows,
>    although they need to provide traffic engineering services and proper
>    QoS to the LSPs associated with DetNet flows to enhance the prospect
>    of the LSPs meeting the DetNet service requirements.  Some
>    implementations of transit nodes may be DetNet aware, but such nodes
>    just support the DetNet transport layer.
> 
> ... I could maybe assume that actually there can be "transit nodes" that do not
> require *any* sort of DetNet support or DetNet awareness. Is that correct? I
> think the architecture document has to be clear on such fundamental questions.
> 
> The Introduction could be extended to clarify, e.g., the third paragraph:
> 
> OLD:
> A goal of DetNet is a converged network in all respects. That is, the presence of
> DetNet flows does not preclude non-DetNet flows, and the benefits offered
> DetNet flows should not, except in extreme cases, prevent existing QoS
> mechanisms from operating in a normal fashion, subject to the bandwidth
> required for the DetNet flows. A single source-destination pair can trade both
> DetNet and non-DetNet flows. End systems and applications need not
> instantiate special interfaces for DetNet flows. Networks are not restricted to
> certain topologies; connectivity is not restricted. Any application that generates
> a data flow that can be usefully characterized as having a maximum bandwidth
> should be able to take advantage of DetNet, as long as the necessary resources
> can be reserved. Reservations can be made by the application itself, via network
> management, by an application’s controller, or by other means, e.g., a dynamic
> control plane (e.g., [RFC2205]).
> 
> NEW:
> A goal of DetNet is a converged network in all respects. That is, the presence of
> DetNet flows does not preclude non-DetNet flows, and the benefits offered
> DetNet flows should not, except in extreme cases, prevent existing QoS
> mechanisms from operating in a normal fashion, subject to the bandwidth
> required for the DetNet flows. A single source-destination pair can trade both
> DetNet and non-DetNet flows. End systems and applications need not
> instantiate special interfaces for DetNet flows. Networks are not restricted to
> certain topologies; connectivity is not restricted. Any application that generates
> a data flow that can be usefully characterized as having a maximum bandwidth
> should be able to take advantage of DetNet, as long as the necessary resources
> can be reserved. Reservations can be made by the application itself, via network
> management, by an application’s controller, or by other means, e.g., a dynamic
> control plane (e.g., [RFC2205]). Network nodes supporting DetNet flows have to
> implement some of the DetNet capabilities (not necessarily all) in order to treat
> DetNet flows such that their QoS requirements are met.
> 
> [ms] To me, the last sentence is not clear. Does "some of the DetNet capabilities
> (not necessarily all)" include e.g. vanilla traffic engineering? For instance, if a
> network node can ensure that the QoS requirements are met by leveraging
> existing standardized TE without any additional DetNet awareness, is this
> enough for deploying DetNet? Or does this node have to implement additional
> DetNet-specific functionality? Specifically, please explain if an existing DetNet-
> unaware router on the path taken by DetNet traffic can be used to forward
> DetNet traffic, or if it is mandatory that any router forwarding DetNet traffic
> indeed implements new capabilities specific to DetNet. In the latter case, I think
> the document would need to discuss incremental deployment strategies as well.
> 
> > * It is surprising that there is hardly any discussion on network robustness
> > and safety; this probably also relates to security. For instance,
> > misconfiguration or errors of functions performing packet replication could
> > severely and permantly congest a network and cause harm. How does the DetNet
> > architecture ensure that a network stays fully operational e.g. if the topology
> > changes or there are equipment failures? Probably this can be solved by
> > implementations (e.g., dynamic control plane), but why are corresponding
> > requirements not spelt out? Section 3.3.2 speculates that filters and policers
> > can help, and that may be true, but that probably still assumes consistently
> > and correctly configured (and well-behaving) devices. And Section 3.3.2 is
> > vague and mentions a "infinite variety of possible failures" without stating
> > any requirements or recommendations. There may be further solutions, such as
> > circuit breakers and the like. Why are such topics not discussed?
> Security issues and considerations are addressed by the DetNet Security
> draft: https://datatracker.ietf.org/doc/draft-ietf-detnet-security.
> Reference to the  DetNet Security draft will be added.
> 
> There will be a document dedicated to Operations, Administration and
> Maintenance (OAM), which will address operational, misconfiguration, and
> failure detection issues.
> 
> The topology changes have to be solved by the control plane, either
> centralized or distributed.
> 
> The filters and policers described in Section 3.3.2 also provide
> robustness by eliminating/mitigating traffic coming from malfunctioning
> devices.
> 
> RFC2475 is recommended for traffic policing functions to increase
> robustness.
> 
> The text  in Section 3.3.2 could be made clearer, e.g., by updating the
> first paragraph to:
> 
> OLD:
> One key to building robust real-time systems is to reduce the
> infinite variety of possible failures to a number that can be
> analyzed with reasonable confidence. DetNet aids in the process by
> allowing for filters and policers to detect DetNet packets received
> on the wrong interface, or at the wrong time, or in too great a
> volume, and to then take actions such as discarding the offending
> packet, shutting down the offending DetNet flow, or shutting down the
> offending interface.
> 
> NEW:
> Robust real-time systems require to reduce the number of possible
> failures. Filters and policers should be used in a DetNet network to
> detect if DetNet packets are received on the wrong interface, or at
> the wrong time, or in too great volume. Furthermore, filters and
> policers can take action to discard offending packets or flows, or
> trigger shutting down the offending DetNet flow, or shutting down
> the offending interface.
> 
> [ms] That does not address my concern. As I wrote already, I believe this
> document has to discuss the applicability of circuit breakers according to RFC
> 8084.
> 
> > * Somewhat related, the document only looks at impact of failures to the QoS of
> > DetNet traffic. What is missing is a discussion how to protect non-DetNet parts
> > of a network from any harm caused by DetNet mechanisms. Solutions to this
> > probably exist. But why is the impact on non-DetNet traffic (e.g., in case of
> > topology changes or failures of DetNet functions) not discussed at all in the document?
> Actually the regulation of DetNet traffic by polices and filters
> protects both other DetNet traffic and non-DetNet traffic.
> 
> Section 3.3.1 could be extended to make it clear, e.g., by extending the
> last paragraph:
> 
> OLD:
> Ideally, the net effect of the presence of DetNet flows in a network
> on the non-DetNet packets is primarily a reduction in the available
> bandwidth.
> 
> NEW:
> Traffic policing functions (e.g. [RFC2475]) are used to avoid the
> starvation of non-DetNet traffic. Thus, the net effect of the presence
> of DetNet flows in a network on non-DetNet flows is primarily a
> reduction in the available bandwidth.
> 
> [ms] I believe a stronger wording is needed, e.g. along the lines of "a DetNet
> deployment must avoid starvation of non-DetNet traffic ...".
> 
> [...]
> 
> > * Section 4.4 refers to RFC 7426, which is an informational RFC on IRTF stream,
> > and the document uses the concepts introduced there (e.g., "planes"). This is
> > very confusing. First, an IETF Proposed Standard should probably refer to
> > documents having IETF consensus. An example would be RFC 7491, albeit there is
> > other related work as well, e.g., in the TEAS WG. Second, Section 4.4 is by and
> > large decoupled from the rest of the document and not specific to DetNet.
> > Neither do other sections of the document refer to the concepts introduced in
> > Section 4.4, nor does Section 4.4 use the DetNet terminology or discuss
> > applicability to DetNet. Section 4.4 even mentions explicitly at the end that
> > it discusses aspects that are orthogonal to the DetNet architecture. It is not
> > at all clear why Section 4.4 is in this document. Section 4.4 could be removed
> > from the document without impacting the rest of the document.
> The document should point out to traffic engineering and centralized
> control plane aspects, so it is better to keep Section 4.4.
> 
> RFC 7426 provides the full picture of SDN architecture, which needs to
> be referred from the DetNet architecture. No other RFC provides such
> detailed picture. Therefore, we need to keep the reference to RFC 7426,
> which is an informative reference.
> 
> [ms] I'll leave that up to the IESG to decide whether an IETF proposed standard
> should build an architecture based on research work instead of, say, documents
> with IETF consensus. As a side note, I believe the understanding in industry what
> "SDN architecture" really means and how it is implemented may have evolved
> since publication of IRTF RFC 7426, which was published in Jan. 2015. For
> instance, IRTF RFC 7426 does not discuss the implications of segment routing,
> BGP-LS, and other recent IETF techniques. And, based on what I read in the
> DetNet architecture, I believe DetNet could actually be implemented without
> any "SDN architecture", e.g. by distributed traffic engineering. As a result, I don't
> even understand the need to discuss "SDN".
> 
> [ms] For the record: As TSV-ART reviewer, I question whether IRTF RFC 7426 is
> indeed an appropriate reference, I have doubts that IRTF RFC 7426 is really up-
> to-date, and I am concerned about the content quality of section 4.4, e.g.,
> regarding the lack of alignment with the rest of the document. The IESG will
> certainly have a better understanding that than me how to deal with such topics.
> 
> The related work in TEAS is already referred to.
> 
> [...]
> 
> > Minor issues:
> >
> > * Terminology "DetNet transport layer"
> >
> > The term "transport layer" has a well-defined meaning in the IETF, e.g.
> > originating from RFC 1122. While "transport" and e.g. "transport network" is
> > used in the IETF for different technologies in different areas, I think the
> > term "transport layer" is typically understood to refer to transport
> > protocols such as TCP and UDP. As such, I personally find the term "DetNet
> > transport layer" misleading and confusing. The confusion is easy to see e.g.
> > in Figure 4, where UDP (which is a transport protocol as per RFC 1122) sits
> > on top of "transport".
> >
> > Based on the document it also may be solution/implementation specific
> whether
> > the "DetNet transport layer" is actually a separate protocol layer compared
> > to the "DetNet service layer". Thus it is not clear to me why the word
> > "layer" has to be used, specifically in combination "transport layer".
> >
> > To me as, the word "transport layer" (and "transport protocol") should be
> > used for protocols defined in TSV area, consistent with RFC 1122. But this is
> > probably a question to be sorted out by the IESG.
> 
> "transport" is used here as in "packet transport networks" not as OSI L4
> transport.
> 
> [ms] At the risk of stating the obvious: The relevant section of RFC 1122 defining
> "transport layer" is entitled by "The Internet Architecture". And, according e.g.
> to RFC 5921, "packet transport network" is a term defined by ITU-T. In an IETF
> document, I believe IETF terminology should be used.
> 
> I suggest to use the terms "DetNet transport sub-layer" and "DetNet
> service sub-layer" throughout the document, which would hopefully avoid
> confusion.
> 
> [ms] "DetNet transport sub-layer" somewhat solves my concern of avoiding the
> expression "transport layer". To really avoid confusion, I believe it would be
> better to avoid the ambiguous term "transport" altogether.  But I will be fine if
> all DetNet documents consistently avoid the term "transport layer".
> 
> Michael
> _______________________________________________
> Tsv-art mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/tsv-art