[ Reports & investigation ]

Interpret ASN, reverse DNS and geolocation clues

Network registration, route announcements, reverse names and location estimates answer different questions and cannot independently identify an approved mail service.

By the EmailAuth research team · 6 min read

[ Key points ]

  • Registered resource holders and observed announcing ASNs are different evidence layers.
  • A reverse name is a delegated DNS statement, not proof of From-domain permission.
  • Geolocation estimates do not identify a person or precise physical address.
  • No address lookup or representative customer message was used in the fictional case.
On this page

Ask three different questions of three different clues

An ASN, a reverse-DNS name and an IP geolocation estimate describe different aspects of infrastructure. Combining them can help an investigator choose the next evidence request, but agreement among them does not prove which application sent a message or who approved it. They may all describe the same shared relay while leaving the customer account and business owner entirely unknown.

This guide reviews public network-authority and provider documentation as of 8 November 2026. No address, domain or customer message was queried. The worked case uses invented labels rather than a real network attribution. A supplementary RIPE database-policy page timed out and is not used to support the article; the accessible RIPEstat documentation supplies the routing scope described below.

Distinguish registration from observed routing

ARIN's Whois documentation explains registration records for networks, autonomous systems and related organizations. Registration can identify an allocation or reassignment relationship. It does not identify the software that generated an email. An organization's contact address is also not necessarily the physical location of the resource. Keep the registry source and last-change information rather than turning a company name into a sender verdict.

Observed routing is another layer. RIPEstat Network Info documents a containing prefix and announcing ASN data derived from RIPE RIS. The retained page labels endpoint version 1.1 and describes eight-hour data dumps. Its output can contain multiple ASNs. This is a routing-data snapshot with collection limits, not a live audit of application ownership or a guarantee that one organization controlled every message at a historical time.

Read reverse DNS as a delegated naming statement

A PTR record associates an address with a domain name in the reverse namespace. ARIN's reverse-DNS guide describes the separate IPv4 and IPv6 reverse zones and delegation management. The name may be a provider's generic relay label or a customer-controlled hostname. It is a naming claim from the relevant DNS administration, not proof that the host is approved to send for the visible From domain.

A forward-confirmed reverse relationship can show that a returned name resolves back to the address under the checks performed. That improves consistency evidence but does not establish application purpose or account approval. Keep DNS absence, lookup failure and an unexpected name as different conditions. No lookup was executed here, and neither an attractive hostname nor a missing PTR alone establishes a malicious sender.

Avoid using PTR as a substitute for SPF

RFC 7208 defines SPF around the connecting IP and the relevant SMTP identity. Its ptr mechanism is discouraged; reverse-DNS branding is not a reason to build an SPF policy around it. The authorizing domain and evaluated identity matter more than whether a hostname contains a recognizable provider name. A provider-like substring can be a poor substitute for the full, verified domain relationship.

Similarly, finding a name under a familiar domain does not show that DKIM or DMARC passed. Those mechanisms have their own evidence and identity rules. Use infrastructure clues to decide what to inspect next, not to manufacture an authentication result that the receiver did not report. Any future DNS inspection should retain the exact owner, record type, time and response status separately from the interpretation.

Treat geolocation as an estimate with a scope

MaxMind's accuracy guidance explains that accuracy varies and that city-level data can include an accuracy radius. It explicitly does not locate a specific person, household or street address. A hosting or proxy location can differ from the end user's location. A reported city should therefore not become an allegation that an employee or customer sent mail from that city.

Record the dataset, edition or observation date and any available uncertainty information. An omitted city can mean the source did not provide that precision; it is not evidence that the address has no location. Different databases can disagree because their methods and inputs differ. This guide does not import a location database, call a geolocation service or assign coordinates to anyone.

Work through a fictional infrastructure bundle

The pre-draft paper case contains a network label, North Network; a PTR name, relay.north.example; and a broad regional estimate. All are invented. There is no real ASN number, actual address lookup or representative message behind them. Taken together, these clues suggest where a future investigator might seek network and service documentation. They do not establish a particular mail application, customer tenant or authorized business purpose.

The next evidence layer would connect a representative message's identities and trusted receiving trace to a customer-specific service record. RFC 9990 defines a report's source IP as the connecting address, so even the infrastructure bundle may describe an intermediary rather than the original composing system. Preserve that possibility instead of translating “network associated with the observed connection” into “application that authored the mail.”

Use disagreements to guide a bounded investigation

A mismatch among registry organization, PTR name and region is a question to investigate, not an automatic spoofing classification. Changes of address assignment, outsourced relays or different dataset times can produce apparently inconsistent labels. Those are possible explanations only. Ask which specific evidence would distinguish them, and mark the attribution unresolved until that evidence is available.

The exact intake gate remains open: Validate current primary specifications and retain a worked example before drafting. This documentary work retains current source passages and a fictional case, but it does not meet the original-evidence requirement: Combine infrastructure clues with a representative message and customer-specific service evidence. No representative message or customer record was supplied. Keep approval of a sender, external disclosure and any DNS or credential action separate from reading infrastructure documentation.

[ Takeaway ]

Use infrastructure clues to choose the next evidence request, not to manufacture sender identity.

[ By role ]

What this means for you

Choose your role for practical next steps.

Network investigator

Label each source by what it measures

Keep registration, routing, naming and location datasets separate so apparent agreement does not become an unsupported account attribution.

Next steps

  1. Retain source version, collection time and coverage.
  2. Distinguish a lookup error from absent or imprecise data.
  3. Document conflicting labels without forcing one provider verdict.

Email-authentication analyst

Connect clues to message identities

Use infrastructure context only as a lead toward the real SMTP and signing evidence.

Next steps

  1. Preserve the connecting-address boundary in aggregate reports.
  2. Do not infer SPF or DKIM results from a PTR name.
  3. Request an approved representative message and service mapping before classification.

Privacy and service reviewer

Prevent location overreach

Ensure regional network estimates are not used to locate a person or imply a business approval that no one supplied.

Next steps

  1. Keep precise-person claims out of the decision record.
  2. Record uncertainty and omit unnecessary location detail.
  3. Authorize any external evidence sharing separately from public documentation research.

[ FAQs ]

Frequently asked questions

How does an announcing ASN differ from a registry organization?

An announcing ASN describes observed routing of a prefix in a particular data source. Registry records describe resource registration and associated entities. Neither alone identifies the application that submitted a message.

Is the documented RIPEstat result an instantaneous routing truth?

No. The retained Network Info page labels version 1.1 and says results use RISwhois data based on eight-hour dumps. Preserve the source and timing; no query was run for this article.

What does forward-confirmed reverse DNS add to an investigation?

It can establish consistency between a reverse name and a forward address under the performed checks. It still does not establish application ownership, customer approval or successful message authentication.

Should a familiar PTR hostname be used as an SPF shortcut?

No. SPF has its own evaluated identity and authorization rules, and RFC 7208 discourages the ptr mechanism. A recognizable hostname is not evidence that the visible From domain authorized a service.

Can an IP geolocation city identify the person sending email?

No. MaxMind says its geolocation data does not locate a specific individual, household or street address. A hosting or proxy location can differ from the end user’s location.

Does missing city data imply that an address is suspicious?

No. The data source may not provide that precision or coverage. Preserve missingness and uncertainty rather than converting them into a maliciousness verdict.

What should conflicting ASN, PTR and region labels trigger?

A bounded investigation of source timing, delegation and service relationships. They do not automatically prove spoofing. Record competing explanations and the specific evidence needed to distinguish them.

What real evidence is missing from the North Network example?

The network, PTR and region labels are invented. A representative message and customer-specific service evidence are still required to connect actual infrastructure clues to an approved application.

[ Sources ]

Sources

  1. ARIN: Reverse DNS
  2. ARIN: Using Whois and interpreting registration data
  3. MaxMind: Geolocation accuracy
  4. RFC 1035: Domain Names, Implementation and Specification
  5. RFC 7208: Sender Policy Framework
  6. RFC 9990: DMARC Aggregate Reporting
  7. RIPE NCC: RIPEstat Network Info

[ Glossary ]

Glossary

Resource-registration record
Registry information about an Internet-number resource and associated entities, not a message-level application ownership record.
Announcing ASN
An autonomous-system identifier associated with a prefix announcement in a routing dataset; its evidentiary scope depends on that source and collection time.
PTR naming statement
Reverse-DNS data associating an address with a domain name through the relevant delegated namespace.
Accuracy radius
A geolocation uncertainty measure supplied by some city-level products, not a precise address or a guarantee about a person’s location.
Attribution evidence layer
A separately sourced observation, such as routing or application configuration, whose scope must be preserved when combining clues.

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