<?xml version="1.0" encoding="UTF-8"?>
  <?xml-stylesheet type="text/xsl" href="rfc2629.xslt" ?>
  <!-- generated by https://github.com/cabo/kramdown-rfc version 1.6.34 (Ruby 3.0.2) -->


<!DOCTYPE rfc  [
  <!ENTITY nbsp    "&#160;">
  <!ENTITY zwsp   "&#8203;">
  <!ENTITY nbhy   "&#8209;">
  <!ENTITY wj     "&#8288;">

<!ENTITY RFC4301 SYSTEM "https://bib.ietf.org/public/rfc/bibxml/reference.RFC.4301.xml">
<!ENTITY RFC7296 SYSTEM "https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7296.xml">
]>


<rfc ipr="trust200902" docName="draft-mglt-ipsecme-dscp-np-00" category="std" consensus="true" submissionType="IETF" updates="RFC4301" tocInclude="true" sortRefs="true" symRefs="true">
  <front>
    <title abbrev="DSCP Notify Payload">Differentiated Services Field Codepoints Internet Key Exchange version 2 Notification</title>

    <author initials="D." surname="Migault" fullname="Daniel Migault">
      <organization>Ericsson</organization>
      <address>
        <email>daniel.migault@ericsson.com</email>
      </address>
    </author>
    <author initials="J." surname="Halpern" fullname="Joel Halpern">
      <organization>Ericsson</organization>
      <address>
        <email>joel.halpern@ericsson.com</email>
      </address>
    </author>
    <author initials="U." surname="Parkholm" fullname="U. Parkholm">
      <organization>Ericsson</organization>
      <address>
        <email>ulf.x.parkholm@ericsson.com</email>
      </address>
    </author>
    <author initials="D." surname="Liu" fullname="Daiying Liu">
      <organization>Ericsson</organization>
      <address>
        <email>harold.liu@ericsson.com</email>
      </address>
    </author>

    <date year="2023" month="October" day="06"/>

    <area>Security</area>
    <workgroup>IPsecme</workgroup>
    <keyword>Internet-Draft</keyword>

    <abstract>


<?line 42?>

<t>IPsec supports "classifier" mechanism to send traffic with specific Differentiated Services Field Codepoints (DSCP) over different tunnels. However, such classification is not explicitly notified to the other peer.
This document specifies the DSCP Notification Payload, which, in a CREATE_CHILD_SA Exchange, explicitly mentions which DSCP code points will be tunneled in the newly created tunnel.</t>



    </abstract>



  </front>

  <middle>


<?line 48?>

<section anchor="introduction"><name>Introduction</name>

<t><xref section="4.1" sectionFormat="comma" target="RFC4301"/> acknowledges that aggregating traffic with multiple DSCP over the same SA may result in inappropriate discarding of lower priority packets due to the windowing mechanism used by this feature.
To address such concern, <xref section="4.1" sectionFormat="comma" target="RFC4301"/> recommends the sender implements a "classifier" mechanism which dispatches the traffic over multiple SAs.
However, the peer is not able to indicate the other peer it is classifying traffic according to its DSCP values, nor how DSCP values are classified.</t>

<t>While <xref section="4.4.2.1" sectionFormat="comma" target="RFC4301"/> mentions the "DSCP values" fields in the Security Association Database (SAD), <xref target="RFC7296"/> does not provides a way to for peers to indicate which "DSCP values" are associated to the created SA.
This document fills that gap and specifies the DSCP Notification Payload, which, in a CREATE_CHILD_SA Exchange, explicitly mentions which DSCP code points will be tunneled in the newly created tunnel.</t>

</section>
<section anchor="sec-rfc4301"><name>RFC4301 clarification</name>

<t><xref section="4.4.2.1" sectionFormat="comma" target="RFC4301"/> mentions</t>

<figure><artwork><![CDATA[
    o DSCP values -- the set of DSCP values allowed for packets carried
      over this SA.  If no values are specified, no DSCP-specific
      filtering is applied.  If one or more values are specified, these
      are used to select one SA among several that match the traffic
      selectors for an outbound packet.  Note that these values are NOT
      checked against inbound traffic arriving on the SA.
]]></artwork></figure>

<t>The text does not clearly specify what happens when the DSCP of a packet does not match any of the corresponding DSCP values.
This document proposes the following text:</t>

<t><list style="symbols">
  <t>DSCP values -- the set of DSCP values allowed for packets carried
   over this SA. If no values are specified, no DSCP-specific
   filtering is applied.  If one or more values are specified, these
   are used to select one SA among several that match the traffic
   selectors for an outbound packet. In case of multiple matches a preference to the most selective list DSCP value could be implemented by the peer's policy. 
   If the DSCP value of the packet does not match any of the DSCP values provided by the associated matching SAs and there is at least one SA with no DSCP-specific filtering, then, one of these SA SHOULD be selected. On the other hand, if all SAs have a DSCP filtering, then, any of the matching SAs can be selected. Note that these values MUST NOT be checked against inbound traffic arriving on the SA.</t>
</list></t>

</section>
<section anchor="protocol-overview"><name>Protocol Overview</name>

<t>The illustrative example of this section considers Expedited Forwarding (EF) with low latency traffic has its own IPsec tunnel, Assured Forward (AF) classes with different drop precedence and may take a different route have their own tunnel and all remaining DSCP values are put in another tunnel.<br />
This section details how a peer uses the DSCP Notify Payload to classify traffic carrying the DSCP values AF11 or AF3 in one tunnel, traffic carrying a DSCP value of EF in another tunnel and traffic with other DSCP values a third tunnel.
This latest SA is designated as the default no-DSCP specific SA. It is RECOMMENDED to configure the Security Policy Data Base (SPD), so that such a default no-DSCP specific SA is created and it is RECOMMENDED its creation happens prior the SA with specific DSCP values. 
Note that according to <xref target="sec-rfc4301"/>, there is no specific ordering, but starting with the no-DSCP specific SA ensures compatibility with IPsec implementation that would for example discard or create a new SA when the DSCP do not match.</t>

<t>Generally, it is recommended that the outer DSCP value matches the inner DSCP value so that the tunneled packet be treated similarly to the inner packet. Such behavior is provided by setting the Bypass DSCP to True.
If the initiator prefers for example every tunneled packet being treated similarly, then, an explicit mapping needs to be indicated. Typically, the initiator may be willing to prevent reordered traffic to fall outside the anti-replay windows.
Note that such policy is implemented by each peer.</t>

<figure><artwork><![CDATA[
   Initiator                         Responder
   -------------------------------------------------------------------
   HDR, SK {IDi, [CERT,] [CERTREQ,]
       [IDr,] AUTH, SAi2,
       TSi, TSr}  -->

                                <--  HDR, SK {IDr, [CERT,] AUTH,
                                         SAr2, TSi, TSr}
]]></artwork></figure>

<t>Once the no-DSCP specific SA is created, all traffic with any DSCP value is steered to that SA. The initiator then creates the child SA associated with specific DSCP values. In this example, it creates the SA associated with the DSCP value AF11 or AF3, followed by the one associated with  value of EF, but this does not follow any specific ordering.
The initiator specifies the DSCP values being classified in that SA with a DSCP Notify Payload that carries the DSCP values.</t>

<t>If the responder supports the DSCP Notify Payload, it SHOULD respond with a Notify Payload that indicates the DSCP values selected for that tunnel. 
By default these values SHOULD be the ones specified by the initiator, but the responder's policy MAY select other values. If the responder does not want to perform DSCP filtering, the responder SHOULD send a empty DSCP Notify Payload in order to at least indicate the support of the DSCP Notify Payload.</t>

<t>As specified in <xref section="3.10.1" sectionFormat="comma" target="RFC7296"/>, Notify Payload with status type MUST be ignored if not recognized.
The absence of a DSCP Notify Payload by the responder may be due to the responder not supporting the notification or not advertising the application of DSCP filtering. 
We do not consider that the absence of classification by the responder  prevents the SA to be created. The classification is at least performed for the outbound stream, which is sufficient to justify the creation of the additional SA.
Note also that DSCP values are not agreed, and the responder cannot for example narrow down the list of DSCP values being classified.
If that would cause a significant issue, the responder can create another SA with the narrow down list of DSCP values. The responder may also REKEY_SA the previously SA to redefine the DSCP values to be considered.</t>

<t>When multiple DSCP values are indicated, and the initiator is mapping the outer DSCP value, the outer DSCP value is expected to be one of these values.</t>

<figure><artwork><![CDATA[
   Initiator                         Responder
   -------------------------------------------------------------------
   HDR, SK {SA, Ni, KEi, N(DSCP, AF11, AF3)} -->

                                <--  HDR, SK {SA, Nr, KEr, 
                                          N(DSCP, AF11, AF3)}
]]></artwork></figure>

<t>The initiator may then create additional child SAs specifying other DSCP values.</t>

<figure><artwork><![CDATA[
   Initiator                         Responder
   -------------------------------------------------------------------
   HDR, SK {SA, Ni, KEi, N(DSCP, EE)} -->

                                <--  HDR, SK {SA, Nr, KEr}
]]></artwork></figure>

</section>
<section anchor="protocol-description"><name>Protocol Description</name>

<t>During the CREATE_CHILD_SA exchange, the initiator or the responder MAY indicate to the other peer the DSCP filtering policy applied to the SA. This is done via the DSCP Notify Payload indicating the DSCP values being considered for that SA.</t>

<t>The initiator MAY send an empty DSCP Notify Payload to indicate support of the DSCP Notify Payload as well as an indication the negotiated SA as a no-DSCP specific SA. This SA MAY be followed by the creation of  DSCP specific SA.</t>

<t>Upon receiving a DSCP Notify Payload, if the responder supports the notification it SHOULD respond with a DSCP Notify Payload. The value indicated SHOULD be the one selected by the initiator.</t>

<t>There is no specific error handling.</t>

</section>
<section anchor="payload-description"><name>Payload Description</name>

<t>The DSCP Notify Payload is based on the format of the Notify Payload as described in <xref section="3.10" sectionFormat="comma" target="RFC7296"/> and represented in <xref target="fig-np"/>.</t>

<figure title="Notify Payload" anchor="fig-np"><artwork><![CDATA[
                     1                   2                   3
 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Next Payload  |C|  RESERVED   |         Payload Length        |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|  Protocol ID  |   SPI Size    |      Notify Message Type      |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|                                                               |
~                       Notification Data                       ~
|                                                               |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
]]></artwork></figure>

<t>The fields Next Payload, Critical Bit, RESERVED, and Payload Length are defined in <xref target="RFC7296"/>.  Specific fields defined in this document are:</t>

<t><list style="symbols">
  <t>Protocol ID (1 octet):  Set to zero.</t>
  <t>Security Parameter Index (SPI) Size (1 octet):  Set to zero.</t>
  <t>Notify Message Type (2 octets):  Specifies the type of notification message.  It is set to DSCP_VALUES (see <xref target="sec-iana"/>).</t>
  <t>Notification Data (variable length): lists the DSCP values that are considered for the SA. Each value is encoded over a single byte.</t>
</list></t>

</section>
<section anchor="sec-iana"><name>IANA Considerations</name>

<t>IANA is requested to allocate one value in the "IKEv2 Notify Message Types - Status Types" registry:
(available at https://www.iana.org/assignments/ikev2-parameters/ikev2-parameters.xhtml#ikev2-parameters-16) with the following definition:</t>

<figure><artwork><![CDATA[
  Value    Notify Messages - Status Types
-----------------------------------------
  TBD      DSCP 
]]></artwork></figure>

</section>
<section anchor="security-considerations"><name>Security Considerations</name>

<t>As the DSCP value field is already defined by <xref target="RFC4301"/> in the SA structure. As such security considerations of <xref target="RFC4301"/> apply.
The DSCP Notification Payload communicates clearly the DSCP value field to the responder.</t>

<t>When the tunnel mode is used, the communication of the DSCP value field could be easily interpreted by monitoring the received DSCP values of the inner traffic when that traffic is encapsulated, and so no secret information is revealed.
When the transport mode is used, that value may be changed by the network and eventually, the value of the field could be unknown to the other peer. However, this cannot be considered as a protection mechanism, and the communication of the DSCP value cannot be considered as revealing information that was previously not revealed.</t>

<t>The ability to indicate the other peer to set the DSCP value does not lead to the consumption of additional resources nor additional constraints. First, these SAs are anyway created either with DSCP values or without. Then, the responder may also ignore the DSCP Notification Payload and the fact that the SA has set a DSCP value field to specific values does not prevent the responder to send traffic with a different DSCP value over that SA.</t>

</section>
<section anchor="acknowledgements"><name>Acknowledgements</name>

<t>We would like to thank Scott Fluhrer for his useful comments; Valery Smyslov, Tero Kivinen for their design suggestions we carefully followed.</t>

</section>


  </middle>

  <back>


    <references title='Normative References'>

&RFC4301;
&RFC7296;


    </references>



<?line 207?>



  </back>

<!-- ##markdown-source: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-->

</rfc>

