[ SPF ]

SPF failures concentrated in one recipient network

Compare the actual SMTP path, evaluated identity and DNS evidence before changing SPF to address failures at one recipient network.

By the EmailAuth research team · 3 min read

[ Key points ]

  • Compare the actual source, identity and policy before changing authorization.
On this page

An SPF incident that appears limited to one recipient network is a comparison problem. First establish whether that network evaluated the same sending IP, envelope domain and DNS policy as your successful controls. A broad authorisation change can hide the difference while increasing who may send using your domain.

Build a matched comparison

Select a failed transaction and a successful transaction from the same application and approximate time. Record the UTC time, connecting IPv4 or IPv6 address, SMTP MAIL FROM, HELO name, recipient network, exact SMTP reply, and trusted authentication result if the message was accepted. A message rejected during SMTP may never have a delivered header to inspect. Preserve the sender's transaction log instead. Treat identifiers and message contents as restricted operational evidence.

SPF normally evaluates the MAIL FROM domain; for a null reverse path, its MAIL FROM identity is constructed using postmaster and the HELO domain. The visible From address is not the lookup instruction. A forwarding service can also change the connecting IP seen by the final receiver. RFC 7208, sections 2.3–2.4 and 10.3.

Separate the failure classes

An explicit fail points toward an evaluated path that did not authorise the client. Permerror calls for inspection of policy syntax, multiple SPF records, recursive dependencies and evaluation limits. Temperror calls for investigation of temporary lookup or processing problems. None means no usable evaluated domain or no applicable SPF record was found, rather than an explicit prohibition. Neutral means the policy makes no definite authorization assertion; softfail expresses a weaker negative assertion than fail. Neither is an SPF pass. Capture the result before interpreting a generic “authentication problem” ticket.

Compare authoritative DNS answers with recursive answers and their timestamps. Query the exact owner name and every dependency reached for that IP. SPF's ten-term limit covers evaluated include, a, mx, ptr and exists mechanisms and the redirect modifier, including nested evaluation, rather than every DNS packet or TXT fragment. The ip4, ip6 and all mechanisms do not consume that term budget; separate limits can still apply to other DNS work. A static tool scan and a receiver's actual path can therefore disagree. RFC 7208, sections 4.5–4.6.4 and 8.

Work through a controlled example

Illustrative scenario: the sender reaches network A over an authorised IPv4 address but reaches network B over an unlisted IPv6 address. The application's name and From address look identical. The useful intervention is to verify ownership and intended use of the IPv6 route, then authorise it narrowly if appropriate or stop using it. Adding the recipient network to SPF would address the wrong side of the connection.

A second possibility is forwarding: network B receives from an intermediary rather than your original outbound host. Do not add arbitrary forwarders to your policy. Check the forwarding path, surviving aligned DKIM and the receiver's policy. A passing SPF result can still be unaligned for DMARC. RFC 9989, section 4.4.

Close with reproducible evidence

Make one controlled change with an owner, saved prior value and rollback condition. Repeat the matched tests through the affected route, including both address families where used. Confirm SPF, aligned authentication and SMTP handling separately. If the same inputs and policy still yield different results, prepare a redacted evidence packet for the receiving operator. Record the unresolved implementation difference instead of labelling the whole network broken.

Use a matched evidence worksheet

  • Failed/control pair: record UTC times, application and recipient network for both.
  • Connection: compare the actual IPv4 or IPv6 source and any forwarding hop.
  • Identity: compare MAIL FROM, HELO and the domain the receiver actually checked.
  • Policy: retain dated TXT answers and the evaluated dependency path.
  • Result: preserve fail, softfail, neutral, none, temperror or permerror exactly, plus the SMTP reply.
  • Action: repair only a verified difference; repeat the same route and retain unresolved differences for escalation.

[ Takeaway ]

Compare the actual SMTP path, evaluated identity and DNS evidence before changing SPF to address failures at one recipient network.

[ By role ]

What this means for you

Choose your role for practical next steps.

IT / Mail Admin

Capture SMTP IP and envelope identity.

Keep a paired transaction record with IP, MAIL FROM, HELO, timestamps and DNS snapshots. A successful dashboard check is not a substitute for replaying the actual failing route.

Next steps

  1. Capture SMTP IP and envelope identity.
  2. Compare IPv4, IPv6 and forwarding paths.
  3. Retest the exact failed route after the change.

CISO / SecOps

Verify ownership of newly authorised addresses.

Require a named owner for any new address range and retain the pre-change policy. Receiver concentration is useful incident evidence, but it does not justify authorising unrelated networks.

Next steps

  1. Verify ownership of newly authorised addresses.
  2. Retain the previous permission set.
  3. Document unresolved receiver differences.

Founder / Non-technical

Identify affected customer workflows.

Ask which business messages are affected and whether the failing route differs from the successful route. Approve a narrow recovery change rather than a blanket relaxation of sender permissions.

Next steps

  1. Identify affected customer workflows.
  2. Request one failed and one successful comparison.
  3. Approve only a justified narrow recovery change.

[ FAQs ]

Frequently asked questions

Why does one recipient see failure while another sees pass?

They may see different connecting IPs, forwarding paths, identities, cached DNS data or evaluation behaviour. Compare these inputs before drawing a conclusion.

Should I add the recipient network to SPF?

No. SPF describes authorised sending hosts for the evaluated domain. A recipient network is not automatically an outbound sender.

Can IPv6 explain the difference?

Yes. A route may send over IPv6 while the policy authorises only its IPv4 source. Verify the actual connection address and intended service before editing.

What if the failed message has no headers?

Use the SMTP transaction logs and exact rejection response. A message rejected before delivery may have no recipient-side copy to inspect.

Does permerror mean the source IP was unauthorised?

It means the policy could not be interpreted or evaluated successfully. Diagnose its cause rather than treating it as a normal fail.

Will a current DNS lookup prove what happened yesterday?

No. Keep contemporaneous answers and configuration history. Current DNS helps reproduce the issue but may differ from the receiver’s earlier view.

Can DMARC pass despite SPF failure?

Yes, when a valid DKIM signature aligns with the Author Domain. Receiver handling and other requirements remain separate.

What evidence should accompany an escalation?

Provide redacted times, connection addresses, evaluated identities, exact result or SMTP reply, policy snapshots and a matched successful control.

[ Sources ]

Sources

  1. RFC7208

    Sections §2.3–2.4; 4.5–4.6.4; 8; 10.3

  2. RFC8601

    Sections §1.2; 7.1

  3. RFC9989

    Sections §4.4

[ Glossary ]

Glossary

SPF
A check of whether a connecting IP address is authorised to use the evaluated SMTP domain.
MAIL FROM
The SMTP reverse-path identity used for handling delivery errors and normally for SPF evaluation. It differs from the visible From header.
DMARC alignment
The required relationship between the Author Domain in From and a successfully authenticated SPF or DKIM domain.

Enterprise provenance

Why top financial institutions and governments retired their email security dashboards.

Traditional email authentication platforms sell you a dashboard and ask your engineers to do the labour. In high-stakes environments, manual oversight means delayed enforcement, configuration drift and unmonitored shadow domains.

EmailAuth is the agentic email provenance platform. Our agents enforce p=reject, keep third-party senders aligned and move your domains to full RFC 9989 compliance, the new DMARC standard,, with human-in-the-loop auditability.

Governments Large enterprises Tier-1 financial institutions

SaaS · Dedicated tenancy · Air-gapped on-prem · BYOK encryption

Knowledge Base