← Back to Blog
Phil Smith, CTO
Founder & Chief Technology Officer
Technical GuideNovember 12, 2026

Securing Inactive & Parked Domains: Defensive SPF and DMARC Null Records

ENGINEERING BLUEPRINT // ARCHITECTURE GUIDEAUTHOR: PHIL SMITH, CTO
Securing Inactive & Parked Domains: Defensive SPF and DMARC Null Records
domainaftercare.com/blogSPEC: MULTI-TENANT DNS & TLS HYGIENE

1. The Forgotten Attack Surface of Dormant & Defensive Domains

Agencies and enterprise security teams routinely register dozens of defensive domains for every primary brand they manage: common keyboard typos, country-code top-level domains (.co.uk, .de, .io), legacy product names, and campaign microsites. Because these domains are never intended to send or receive production email, engineers typically configure an HTTP 301 redirect at the edge and leave the DNS zone's mail records completely blank.

Unfortunately, the Simple Mail Transfer Protocol (SMTP) and modern inbox authentication standards do not interpret the absence of DNS records as a denial of service. When a domain lacks explicit MX, SPF, and DMARC records, receiving Mail Transfer Agents (MTAs) treat the domain as an unauthenticated "open identity." Any external threat actor can spin up an unauthorized relay and transmit spear-phishing messages claiming to originate from billing@yourclient-support.com or payroll@clientbrand-uk.com.

Even worse, RFC 5321 Section 5.1 specifies an implicit fallback rule: if a domain has no MX record at all, a sending MTA is permitted to query the domain's apex A or AAAA record and attempt to deliver mail directly to port 25 of your web server. Leaving parked domains unconfigured wastes network resources, invites silent spoofing, and erodes parent brand trust.

2. Protocol Architecture: The 4-Layer Null Mail Shield

To cryptographically lock a dormant domain against both inbound mail delivery queues and outbound identity spoofing, engineers must publish four explicit "Null" DNS resource records at the authoritative zone level.

ARCHITECTURAL INFOGRAPHICDOMAINAFTERCARE ENGINEERING

Defensive 4-Layer Null Mail Shield for Parked Domains

RFC 7505, RFC 7208, RFC 6376 & RFC 7489 Zero-Trust Enforcement

1
Explicit Null MX (RFC 7505)@ IN MX 0 .

Instructs remote MTAs that the domain accepts zero incoming mail and aborts port 25 A-record fallback.

2
Strict SPF Hardfailv=spf1 -all

Declares zero authorized IPv4/IPv6 egress hosts; any envelope sender immediately fails SPF.

3
Wildcard DKIM Key Revocation*._domainkey TXT "v=DKIM1; p="

Publishes an empty public key across all wildcard selectors to invalidate forged signatures.

4
Enforced DMARC Reject Policyv=DMARC1; p=reject; sp=reject

Mandates 100% rejection across both the organizational root domain and all unmapped subdomains.

Spoofing Surface Reduction
100%
Blocks apex and wildcard subdomain impersonation
Recursive DNS Lookups
0 / 10
Static null policies consume zero SPF lookup budget
MTA Retry Queue Overhead
0 Retries
RFC 7505 triggers immediate 550 5.1.10 permanent rejection

Layer 1: The RFC 7505 "Null MX" Resource Record

Before RFC 7505 was standardized, administrators tried to block inbound mail by pointing MX records to 0.0.0.0 or localhost. That approach forced sending mail servers to queue messages and retry delivery for up to 5 days before timing out. RFC 7505 solves this by introducing a single dot (.) as the target hostname with a preference of 0. When a compliant MTA resolves MX 0 ., it immediately terminates the delivery attempt with a permanent 550 5.1.10 error code.

Layer 2: Restrictive SPF Hardfail (v=spf1 -all)

Sender Policy Framework (RFC 7208) evaluates the RFC5321.MailFrom (Return-Path) domain. By publishing v=spf1 -all with zero ip4, ip6, or include mechanisms, you instruct every receiving server on the internet that no host is authorized to emit mail on behalf of this zone.

Layer 3: Wildcard DKIM Selector Revocation

Sophisticated attackers sometimes attach syntactically valid DKIM headers referencing arbitrary selectors (e.g., s=default or s=mail) hoping poorly configured filters will treat the presence of a signature header favorably. Publishing a wildcard TXT record at *._domainkey with an empty p= tag explicitly tells verifiers that any selector key under this zone is revoked.

Layer 4: Strict DMARC Rejection with Subdomain Inheritance

Even with v=spf1 -all in place, an attacker can still spoof the visible RFC5322.From header while using their own Return-Path domain—unless DMARC (RFC 7489) enforces identifier alignment. Setting p=reject, sp=reject, adkim=s, and aspf=s guarantees that both the root domain and every possible non-existent subdomain (*.parked-domain.com) are locked under strict alignment.

3. Production DNS Zone File & Terraform Implementation

Below is the complete, production-ready BIND zone configuration and record matrix for locking down any parked, defensive, or non-mailing domain:

Record HostTypeTTLCanonical Production ValueSecurity Function
@MX36000 .Prevents inbound mail queueing and A-record port 25 fallback (RFC 7505)
@TXT3600"v=spf1 -all"Hardfails all outbound envelope senders at the root zone
*TXT3600"v=spf1 -all"Hardfails outbound envelope senders on any arbitrary subdomain
*._domainkeyTXT3600"v=DKIM1; p="Revokes all cryptographic DKIM selectors across the zone
_dmarcTXT3600"v=DMARC1; p=reject; sp=reject; adkim=s; aspf=s; rua=mailto:dmarc@agency.com"Enforces strict rejection and routes spoofing telemetry to your agency
text
; =====================================================================
; Defensive Null Mail Zone Template (BIND / Cloudflare Import Format)
; Target Zone: parked-brand-typo.com.
; =====================================================================
$TTL 3600
@               IN  MX   0 .
@               IN  TXT  "v=spf1 -all"
*               IN  TXT  "v=spf1 -all"
*._domainkey    IN  TXT  "v=DKIM1; p="
_dmarc          IN  TXT  "v=DMARC1; p=reject; sp=reject; adkim=s; aspf=s; fo=1; rua=mailto:dmarc-rua@yourdomain.com"

4. Command-Line Verification & Edge-Case Auditing

After deploying defensive records across your client portfolio, verify that authoritative nameservers return the expected payloads for both apex and random wildcard queries:

bash
# 1. Verify RFC 7505 Null MX returns "0 ."
dig +short MX parked-brand-typo.com @1.1.1.1
# Expected output: 0 .

# 2. Verify apex and wildcard SPF hardfail
dig +short TXT parked-brand-typo.com @1.1.1.1
dig +short TXT random-sub.parked-brand-typo.com @1.1.1.1
# Expected output: "v=spf1 -all"

# 3. Verify DMARC policy enforcement and subdomain policy (sp=reject)
dig +short TXT _dmarc.parked-brand-typo.com @1.1.1.1
# Expected output: "v=DMARC1; p=reject; sp=reject; adkim=s; aspf=s; ..."

Important Edge Case — External RUA Authorization: If you route aggregate DMARC reports (rua=mailto:dmarc@agency.com) from parked-brand-typo.com to a central agency reporting domain (agency.com), RFC 7489 Section 7.1 requires you to publish an authorization record on the receiving zone: parked-brand-typo.com._report._dmarc.agency.com IN TXT "v=DMARC1". Without this record, external mailbox providers will silently drop your DMARC XML reports.

5. Agency Production Verification Checklist

  • ☐ Inventory all parked, typo-defensive, and redirect-only domains across every client registrar account.
  • ☐ Assign the Parked / Defensive operational profile in DomainAfterCare so automated scanners enforce Null MX and v=spf1 -all compliance.
  • ☐ Publish @ IN MX 0 . and verify no legacy third-party MX records remain in the zone.
  • ☐ Deploy v=DMARC1; p=reject; sp=reject; adkim=s; aspf=s and verify external _report._dmarc authorization if aggregating RUA telemetry centrally.
  • ☐ Enable continuous drift alerting so any accidental removal of defensive TXT records triggers an immediate Slack notification.