<?xml version="1.0" encoding="utf-8"?>
<!-- name="GENERATOR" content="github.com/mmarkdown/mmark Mmark Markdown Processor - mmark.miek.nl" -->
<rfc version="3" ipr="trust200902" docName="draft-yorgos-dnsop-dry-run-dnssec-04" submissionType="IETF" category="std" xml:lang="en" xmlns:xi="http://www.w3.org/2001/XInclude" indexInclude="true">

<front>
<title abbrev="dry-run-dnssec">dry-run DNSSEC</title><seriesInfo value="draft-yorgos-dnsop-dry-run-dnssec-04" stream="IETF" status="standard" name="Internet-Draft"></seriesInfo>
<author initials="Y." surname="Thessalonikefs" fullname="Yorgos Thessalonikefs"><organization>NLnet Labs</organization><address><postal><street>Science Park 400</street>
<city>Amsterdam</city>
<code>1098 XH</code>
<country>Netherlands</country>
</postal><email>yorgos@nlnetlabs.nl</email>
</address></author><author initials="W." surname="Toorop" fullname="Willem Toorop"><organization>NLnet Labs</organization><address><postal><street>Science Park 400</street>
<city>Amsterdam</city>
<code>1098 XH</code>
<country>Netherlands</country>
</postal><email>willem@nlnetlabs.nl</email>
</address></author><author initials="R." surname="Arends" fullname="Roy Arends"><organization>ICANN</organization><address><postal><street></street>
</postal><email>roy.arends@icann.org</email>
</address></author><date year="2025" month="June" day="17"></date>
<area>Internet</area>
<workgroup>DNSOP Working Group</workgroup>

<abstract>
<t>This document describes a method called &quot;dry-run DNSSEC&quot; that allows for
testing DNSSEC deployments without affecting the DNS service in case of DNSSEC
errors.
It accomplishes that by introducing new DS Type Digest Algorithms that when
used in every record of a DS RRset, referred to as dry-run DS, signal to
validating resolvers that dry-run DNSSEC is used for the zone.
DNSSEC errors are then reported with DNS Error Reporting, but any bogus
responses to clients are withheld.
Instead, validating resolvers fallback from dry-run DNSSEC and provide the
response that would have been answered without the presence of the dry-run DS.
A further EDNS option is presented for clients to opt-in for dry-run DNSSEC
errors and allow for end-to-end DNSSEC testing.</t>
</abstract>

</front>

<middle>

<section anchor="introduction"><name>Introduction</name>
<t>DNSSEC was introduced to provide DNS with data origin authentication and data
integrity.
This brought quite an amount of complexity and fragility to the DNS which in
turn still hinders general adoption.
When an operator decides to adopt DNSSEC on an existing insecure zone there is
no way to realistically check that DNS resolution will not break for the zone.</t>
<t>Recent efforts that improve troubleshooting DNS and DNSSEC include Extended DNS
Errors <xref target="RFC8914"></xref> and DNS Error Reporting <xref target="RFC9567"></xref>.
The former defines error codes that can be attached to a response as EDNS
options.
The latter introduces a way for resolvers to report those error codes to the
zone operators.</t>
<t>This document describes a method called &quot;dry-run DNSSEC&quot; that builds upon the
two aforementioned efforts and provides measurable feedback about DNSSEC
resolution health to operators by enabling production testing of a DNSSEC zone.
This is accomplished by introducing new DS Type Digest Algorithms.
The zone operator signs the zone and makes sure that every DS record in the
published DS RRset on the parent side use dry-run DS Type Digest Algorithm(s).</t>
<t>Validating resolvers that don't support the DS Type Digest algorithms ignore it
as per <xref target="RFC6840" sectionFormat="comma" section="5.2"></xref>.
Validating resolvers that do support dry-run DNSSEC make use of <xref target="RFC8914"></xref> and
<xref target="RFC9567"></xref> to report any DNSSEC errors to the zone operator.
If a DNSSEC validation error was due to dry-run DNSSEC, validation falls back
to insecure as the reply to the client.</t>
<t>This allows real world testing with resolvers that support dry-run DNSSEC
by reporting DNSSEC feedback, without breaking DNS resolution for the domain
under test.</t>
</section>

<section anchor="terminology"><name>Terminology</name>
<t>The key words &quot;<bcp14>MUST</bcp14>&quot;, &quot;<bcp14>MUST NOT</bcp14>&quot;, &quot;<bcp14>REQUIRED</bcp14>&quot;,
&quot;<bcp14>SHALL</bcp14>&quot;, &quot;<bcp14>SHALL NOT</bcp14>&quot;, &quot;<bcp14>SHOULD</bcp14>&quot;, &quot;<bcp14>SHOULD NOT</bcp14>&quot;,
&quot;<bcp14>RECOMMENDED</bcp14>&quot;, &quot;<bcp14>NOT RECOMMENDED</bcp14>&quot;, &quot;<bcp14>MAY</bcp14>&quot;, and
&quot;<bcp14>OPTIONAL</bcp14>&quot; in this document are to be interpreted as described in
BCP 14 <xref target="RFC2119"></xref><xref target="RFC8174"></xref> when, and only when, they appear in all
capitals, as shown here.</t>

<dl spacing="compact">
<dt>dry-run DS</dt>
<dd>The DS RRset with dry-run DS Type Digest Algorithm(s) that signals dry-run
DNSSEC for the delegation. All DS records in the RRset MUST use a dry-run DS
Type Digest Algorithm.</dd>
<dt>dry-run zone</dt>
<dd>A zone that is DNSSEC signed but uses a dry-run DS to signal the use of the
dry-run DNSSEC method.</dd>
<dt>dry-run parent zone</dt>
<dd>A zone that supports dry-run DNSSEC for its delegation; that is support for
publishing the dry-run DS.</dd>
<dt>dry-run resolver</dt>
<dd>A validating resolver that supports dry-run DNSSEC.</dd>
<dt>wet-run client</dt>
<dd>A client that has opted-in to receive the actual DNSSEC errors from the
upstream validating resolver instead of the insecure answers.</dd>
</dl>
</section>

<section anchor="overview"><name>Overview</name>
<t>Dry-run DNSSEC builds upon three previous experiences namely DMARC <xref target="RFC7489"></xref>,
Root Key Trust Anchor Sentinel <xref target="RFC8509"></xref> and Signaling Trust Anchor Knowledge
<xref target="RFC8145"></xref>.
The former enabled email operators to verify their configuration with real
email servers by getting DMARC reports and understanding the impact on email
delivery their configuration would have before committing to enable DMARC.
Experience with the latter two showed that with only a small, up to date
resolver population, the signaling is already quite substantial.</t>
<t>Dry-run DNSSEC offers zone operators the means to test newly signed zones and
a turn-key action to conclude testing and commit to the tested DNSSEC records.
Operators that want to use dry-run DNSSEC SHOULD support <xref target="RFC9567"></xref> and have a
reporting agent in place to receive the error reports.</t>
<t>The only change from normal operations when signing a zone with dry-run
DNSSEC is to not publish the real DS RRset on the parent but publish the
dry-run DS instead.
See <xref target="signaling"></xref> for more information on the dry-run DS itself, and
<xref target="provisioning"></xref> on the parent-child communication for the dry-run DS.</t>
<t><xref target="RFC9567"></xref> is used for invalid answers and it can generate reports
for errors in dry-run DNSSEC zones.
This helps with monitoring potential DNS breakage when testing a DNSSEC
configuration for a zone.
This is also the main purpose of dry-run DNSSEC.</t>
<t>The newly signed zone is publicly deployed but DNSSEC configuration errors
cannot break DNS resolution yet.
DNS Error Reports can pinpoint potential issues back to the operator.
When the operator is confident that the DNSSEC configuration under test does
not introduce DNS breakage, the turn-key action to conclude testing and commit
to the signed zone is to replace the dry-run DS with the real DS RRset on the
parent zone.</t>

<section anchor="dnssec-validation"><name>DNSSEC validation of a dry-run zone</name>
<t>Validating resolvers that don't support the DS Type Digest algorithm ignore it
as per <xref target="RFC6840" sectionFormat="comma" section="5.2"></xref>.
Dry-run resolvers are signaled to treat the zone as a dry-run zone.
Dry-run resolvers SHOULD support <xref target="RFC9567"></xref> in order to report possible errors
back to the operators.</t>
<t>Valid answers as a result of dry-run validation yield authentic data (AD)
responses and clients that expect the AD flag can already benefit from the
transition.</t>
<t>Invalid answers yield the insecure response that would have been answered when
no dry-run DS would have been present in the parent, instead of SERVFAIL.
This is not proper data integrity but the delegation SHOULD NOT be considered
DNSSEC signed at this point.</t>
<t>Dry-run resolvers MAY store the dry-run validation status if they want to
support end-to-end testing as discussed in <xref target="opt-in"></xref>.</t>

<section anchor="dry-run-ds-inconsistencies"><name>Inconsistencies in the dry-run DS</name>
<t>If a dry-run DS consists of multiple DS records and not all of them use a
dry-run DS Type Digest algorithm, the DS records with a dry-run DS Type Digest
algorithm MUST be ignored by dry-run resolvers.
This means that in this case, DNSSEC validation continues only with the non
dry-run DS records.</t>
</section>

<section anchor="negative-caching"><name>Use of aggressive negative caching</name>
<t>Aggressive negative caching <xref target="RFC8198"></xref> needs an explicit mention since DNSSEC
faults there can lead to valid answers that could potentially mask underlying
NSEC(3) issues.</t>
<t>Dry-run resolvers that support aggressive negative caching, upon synthesizing
an answer in a dry-run zone, SHOULD hold off using the synthesized answer and
instead issue an explicit query for the record in question.
If the reply that comes back is different, i.e., the synthesized answer would
prove that the record does not exist whereas the explicit query comes back with
the record itself, this means that there lies an issue with negative records.
A report SHOULD be generated using the Extended DNS Error code TBD_nsec and the
answer to the explicit upstream query SHOULD be used instead of the synthesized
one.</t>
</section>
</section>

<section anchor="fallback"><name>Fallback behavior</name>
<t>In case of validation errors with the dry-run DS, dry-run resolvers fallback
to the insecure state of the zone.
Dry-run resolvers MAY store the dry-run validation status if they want to
support end-to-end testing as discussed in <xref target="opt-in"></xref>.</t>
</section>

<section anchor="no-error"><name>NOERROR report</name>
<t>Dry-run DNSSEC relies on DNS Error Reporting <xref target="RFC9567"></xref> to report resolution
errors back to the zone operators.
DNS Error Reporting solely addresses the reporting of DNS errors but it does
not give any guarantees that DNS Error Reporting aware resolvers are resolving
the zone.
This raises a concern especially for dry-run DNSSEC where absence of error
reports needs to translate to a positive signal that no DNSSEC errors were
encountered.</t>
<t>To solve this, dry-run DNSSEC introduces the NOERROR report.
The NOERROR report is sent from the resolver when no error was encountered
during dry-run DNSSEC validation and notifies the reporting agent of the
resolver's presence.</t>
<t>As with <xref target="RFC9567" sectionFormat="comma" section="4"></xref> the resolver will cache the reporting agent
reply and dampen the number of NOERROR report queries.</t>
<t>The NOERROR report is using the Extended DNS Error code TBD_no.</t>

<section anchor="no-error-query"><name>Constructing the NOERROR Query</name>
<t>The QNAME for the NOERROR report query follows the same semantics as with
<xref target="RFC9567" sectionFormat="comma" section="6.1.1"></xref> and is constructed by concatenating the
following elements:</t>

<ul>
<li><t>A label containing the string &quot;_er&quot;.</t>
</li>
<li><t>The decimal value &quot;0&quot; in a single DNS label as the QTYPE is not relevant for
the NOERROR report.</t>
</li>
<li><t>The list of non-null labels representing the apex of the query name that
triggered this report.</t>
</li>
<li><t>The decimal value of TBD_no in a single DNS label as the Extended DNS Error.</t>
</li>
<li><t>A label containing the string &quot;_er&quot;.</t>
</li>
<li><t>The agent domain. The agent domain as received in the EDNS0 Report-Channel
option set by the authoritative server.</t>
</li>
</ul>
<t>As with <xref target="RFC9567" sectionFormat="comma" section="6.1.1"></xref> if the QNAME of the report query
exceeds 255 octets, it MUST NOT be sent.</t>
<t>The apex is specifically used as the query name for resolvers to only send one
NOERROR report (if applicable) per zone and for the monitoring agents to
differentiate between different zones they are configured with.</t>
</section>
</section>

<section anchor="opt-in"><name>Opt-in end-to-end DNSSEC testing</name>
<t>For further end-to-end DNS testing, a new EDNS0 option code TBD_w (Wet-Run
DNSSEC) is introduced that a client can send along with a query to a validating
resolver.
This signals dry-run resolvers that the client has opted-in to DNSSEC errors
for dry-run zones.
Dry-run resolvers that support opt-in MUST respond with the dry-run DNSSEC
error, if any, and MUST attach the same EDNS0 option code TBD_w in the response
to mark the error response as coming from a dry-run zone.</t>
<t>Dry-run resolvers that support opt-in MUST cache the DNSSEC status of the
dry-run validation next to the actual DNSSEC status.
This enables cached answers to both regular and opt-in clients, similar to
cached answers to clients with and without the CD flag set.</t>
<t>Additional Extended DNS Errors can still be attached in the error response by
the validating resolver as per <xref target="RFC8914"></xref>.</t>
<t>Dry-run resolvers that do not support opt-in MUST ignore the TBD_w EDNS0
option and MUST NOT attach the TBD_w EDNS0 option code in their replies.</t>
</section>
</section>

<section anchor="signaling"><name>Signaling</name>
<t>Signaling to dry-run resolvers that a delegation uses dry-run DNSSEC happens
naturally with the DS RRset returned from the parent zone by specifying new
DS Digest Type Algorithm(s).</t>
<t>Each real algorithm has a potential dry-run equivalent.
Since this is an attribute for all available DS Digest Type Algorithms, the
most significant bit of the DS Digest Type Algorithm is used to signal dry-run
when that bit is set.</t>
<t>Resolvers that do not support dry-run DNSSEC and have no knowledge of the
introduced DS Digest Type Algorithms ignore them as per
<xref target="RFC6840" sectionFormat="comma" section="5.2"></xref>.</t>

<section anchor="discussion-from-ietf-114"><name>Discussion from IETF 114</name>
<t><strong>Note to the RFC Editor</strong>: please remove this entire section before publication.</t>
<t>This is addressed feedback as a result of IETF 114. We keep it here for future
reference while the document is advancing.</t>

<section anchor="burn-a-bit-for-dry-run-ds-digest-type-algorithms"><name>Burn a bit for dry-run DS Digest Type Algorithms</name>

<ul>
<li><t>Viktor Dukhovni:</t>

<ul spacing="compact">
<li>Saner than variable variant.</li>
<li>Hash algorithms are introduced exceedingly rarely, symmetric hashes are
very stable.</li>
<li>No evidence that SHA2 will be compromised in the next 100 years; we may
have SHA3 at some point but little demand.</li>
</ul></li>
<li><t>Peter Thomassen:</t>

<ul spacing="compact">
<li>Better to sacrifice a bit than variable length. Also for post quantum
crypto, in response to Paul Hoffman below, even if keys are large the hash
value will have a constant length.</li>
</ul></li>
<li><t>Libor Peltan: (mailing list)</t>

<ul spacing="compact">
<li>Only a few code points in use now, it seems viable.</li>
</ul></li>
</ul>
</section>

<section anchor="use-a-single-ds-digest-type-algorithm-for-dry-run"><name>Use a single DS Digest Type Algorithm for dry-run</name>

<ul>
<li><t>Need to encode the actual algorithm and data in the DS record; results in
variable length DS record for a single algorithm.</t>
</li>
<li><t>May hinder adoption due to EPP checks/requirements (known record length for
each algorithm).</t>
</li>
<li><t>Mark Andrews:</t>

<ul spacing="compact">
<li>Variable length will be needed for private algorithm types so we may as
well support it here.</li>
</ul></li>
<li><t>Paul Hoffman:</t>

<ul spacing="compact">
<li>Recommends going to variable length to pave the way for post quantum crypto
and the surprising length it may need.</li>
</ul></li>
</ul>
</section>
</section>
</section>

<section anchor="provisioning"><name>Provisioning</name>
<t>This section discusses the communication between a dry-run DNSSEC zone and the
parent domain and the procedures that need to be in place in order for the
parent to publish a dry-run DS for the delegation.
Most of the burden falls with the parent zone since they have to understand the
delegation's intent for use of dry-run DNSSEC.
If the parent does not accept DS records, they need to provide a means so that
the child can mark the provided DNSKEY(s) as dry-run DNSSEC.
This can be achieved either by a flag on the parent's interface, or their
willingness to accept and inspect DS records, that accompany DNSKEY records,
for use of the DRY-RUN DS Type Digest Algorithm.
The case of CDS/CDNSKEY is discussed below.</t>

<section anchor="parent-zone-records"><name>Parent zone records</name>
<t>The only change that needs to happen for dry-run DNSSEC is for the parent to be
able to publish the dry-run DS.
If the parent accepts DS records from the child, the child needs to provide the
dry-run DS.
If the parent does not accept DS records and generates the DS records from the
DNSKEY, support for generating the dry-run DS record, when needed, should be
added to the parent if dry-run DNSSEC is a desirable feature.</t>
<t>When the child zone operator wants to complete the DNSSEC deployment, the
parent needs to be notified for the real DS RRset publication.</t>

<section anchor="cds-cdnskey-consideration"><name>CDS and CDNSKEY Consideration</name>
<t>CDS works as expected by providing the dry-run DS content for the CDS record.
CDNSKEY cannot work by itself; it needs to be accompanied by the aforementioned
CDS to signal dry-run DNSSEC for the delegation.
Thus, parents that rely only on CDNSKEY need to add support for checking the
accompanying CDS record for the DRY-RUN DS Type Digest Algorithm and generating
a dry-run DS if such a record is encountered.</t>
<t>Operators of a dry-run child zone are advised to publish both CDS and CDNSKEY
so that both cases above are covered.</t>
</section>
</section>
</section>

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

<section anchor="security-dnssec"><name>DNSSEC status</name>
<t>For the use case of DNSSEC adoption, dry-run DNSSEC disables one of the
fundamental guarantees of DNSSEC, data integrity.
Bogus answers for expired/invalid data will become insecure answers providing
the potentially wrong information back to the requester.
This is a feature of this proposal but it also allows forged answers by third
parties to affect the zone.</t>
<t>This should be treated as a warning that dry-run DNSSEC is not an end solution
but rather a temporarily intermediate test step of a zone going secure.</t>
<t>Thus, a dry-run zone (only dry-run DS on the parent) SHOULD NOT be
considered as DNSSEC signed since it does not offer all the DNSSEC guarantees.</t>
</section>

<section anchor="security-error-report"><name>Error reporting</name>
<t>Since dry-run DNSSEC relies heavily on DNS Error Reporting <xref target="RFC9567"></xref>, the
same security considerations about the generated error reports apply here as
well.
Especially the use of TCP or DNS Cookies for the reports, which can be enforced
by the monitoring agent to make it harder to falsify the source address of
error reports.</t>
</section>
</section>

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

<section anchor="dry-run-ds-type-digest-algorithm"><name>DRY-RUN DS Type Digest Algorithm</name>
<t>This document defines a new entry in the &quot;Digest Algorithms&quot; registry in the
&quot;Delegation Signer (DS) Resource Record (RR) Type Digest Algorithms&quot; registry
at <eref target="https://www.iana.org/assignments/ds-rr-types">https://www.iana.org/assignments/ds-rr-types</eref> :</t>
<table>
<thead>
<tr>
<th align="right">Value</th>
<th>Description</th>
<th>Status</th>
<th>Reference</th>
</tr>
</thead>

<tbody>
<tr>
<td align="right">128-255</td>
<td>Dry-run DNSSEC</td>
<td>OPTIONAL</td>
<td>[this document]</td>
</tr>
</tbody>
</table></section>

<section anchor="noerror-extended-dns-error"><name>NOERROR Extended DNS Error</name>
<t>This document defines a new entry in the &quot;Extended DNS Error Codes&quot;
registry in the &quot;Domain Name System (DNS) Parameters&quot; registry group at
<eref target="https://www.iana.org/assignments/dns-parameters">https://www.iana.org/assignments/dns-parameters</eref> :</t>
<table>
<thead>
<tr>
<th align="right">INFO-CODE</th>
<th>Purspose</th>
<th>Reference</th>
</tr>
</thead>

<tbody>
<tr>
<td align="right">TBD_no</td>
<td>NOERROR reporting</td>
<td>[this document]</td>
</tr>

<tr>
<td align="right">TBD_nsec</td>
<td>Broken negative cache</td>
<td>[this document]</td>
</tr>
</tbody>
</table></section>

<section anchor="wet-run-edns0-option"><name>Wet-Run EDNS0 Option</name>
<t>This document defines a new entry in the &quot;DNS EDNS0 Option Codes (OPT)&quot;
registry in the &quot;Domain Name System (DNS) Parameters&quot; registry group at
<eref target="https://www.iana.org/assignments/dns-parameters">https://www.iana.org/assignments/dns-parameters</eref> :</t>
<table>
<thead>
<tr>
<th align="right">Value</th>
<th>Name</th>
<th>Status</th>
<th>Reference</th>
</tr>
</thead>

<tbody>
<tr>
<td align="right">TBD_wet</td>
<td>Wet-Run DNSSEC</td>
<td>Optional</td>
<td>[this document]</td>
</tr>
</tbody>
</table></section>
</section>

<section anchor="acknowledgements"><name>Acknowledgements</name>
<t>The authors would like to thank the following people, in no particular order,
who contributed into shaping this document with their feedback: Libor Peltan,
Dave Lawrence, Paul Wouters, Tedius Schrijven, Mats Dufberg, Petr Špaček, Marco
Davids, Mark Andrews, Ben Schwartz, Peter Thomassen, Gavin Brown, Nils Wisiol,
Viktor Dukhovni, Paul Hoffman and Lars-Johan Liman.</t>
</section>

</middle>

<back>
<references><name>Normative References</name>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.6840.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.7489.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8145.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8198.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8509.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.8914.xml"/>
<xi:include href="https://bib.ietf.org/public/rfc/bibxml/reference.RFC.9567.xml"/>
</references>

<section anchor="implementation-status"><name>Implementation Status</name>
<t><strong>Note to the RFC Editor</strong>: please remove this entire section before publication.</t>
<t>In the following implementation status descriptions, &quot;dry-run DNSSEC&quot; refers
to dry-run DNSSEC as described in this document.</t>
<t>None yet.</t>
</section>

<section anchor="change-history"><name>Change History</name>
<t><strong>Note to the RFC Editor</strong>: please remove this entire section before publication.</t>

<ul spacing="compact">
<li>draft-yorgos-dnsop-dry-run-dnssec-00</li>
</ul>
<blockquote><t>Initial public draft.</t>
</blockquote>
<ul spacing="compact">
<li>draft-yorgos-dnsop-dry-run-dnssec-01</li>
</ul>
<blockquote><t>Document restructure and feedback incorporation from IETF 113.</t>
</blockquote>
<ul spacing="compact">
<li>draft-yorgos-dnsop-dry-run-dnssec-02</li>
</ul>
<blockquote><t>Document restructure and feedback incorporation from IETF 114; mainly:</t>
<t>Use explicit dry-run algorithm types for DS.</t>
<t>Introduce NOERROR reporting.</t>
</blockquote>
<ul spacing="compact">
<li>draft-yorgos-dnsop-dry-run-dnssec-03</li>
</ul>
<blockquote><t>Shape up NOERROR reporting.</t>
<t>No need for exclusive NOERROR signal from upstream; existence of dry-run suffices.</t>
<t>Ask for NOERROR Extended DNS Error.</t>
<t>Remove most IETF 114 feedback sections for better flow of the document; kept the discussion about signaling.</t>
<t>Add security considerations for increased validation workload.</t>
<t>Add an explicit fallback behavior section.</t>
</blockquote>
<ul spacing="compact">
<li>draft-yorgos-dnsop-dry-run-dnssec-04</li>
</ul>
<blockquote><t>Dry-run only specified for insecure zones.</t>
<t>Add complete acknowledgement section covering all feedback thus far.</t>
<t>Add explicit section about negative caching.</t>
<t>Burn a bit in the DS Digest Type Algorithm for dry-run.</t>
<t>Specify that all DSes in the DS set must be dry-run.</t>
</blockquote></section>

</back>

</rfc>
