Reports show sources we do not recognize
An unfamiliar source in a DMARC report isn't automatically an attacker. It can be a forgotten service, a device, a forwarding path, or someone forging your From domain. Check which domains passed SPF and DKIM, whether those passes aligned, and who owns the sending path before changing DNS. Don't add the IP to SPF merely because the report shows a failure.
Aggregate reports give you observations from receiving systems, not a verified identity for the person behind an IP. An unfamiliar address is a lead to investigate, not a verdict.
Start with the authentication pattern
An aggregate report includes the source IP, message counts, identifiers, authentication results, and policy results. If those fields are unfamiliar, use the main-site guide to reading DMARC reports, or the field-by-field example in A field-by-field walk-through of a DMARC aggregate report. You don't need to interpret every XML field to start triage.
| What you see | What it could mean | Next check |
|---|---|---|
| DKIM passes with alignment, but the IP is unfamiliar | An authorized platform, a forwarded message retaining your signature, or unauthorized use of an authorized sending system | Match the DKIM selector to your systems and ask the service owner. Don't block solely because the IP is new. |
| SPF doesn't pass or align; DKIM passes with alignment | An indirect delivery path, such as forwarding | Confirm the route. DMARC already passes; don't add the forwarder to SPF to repair something that isn't broken. See Forwarded mail fails DMARC. |
| SPF or DKIM passes only for unrelated domains | A vendor using its own bounce and signing domains, or another source with no aligned identity | Confirm that you actually use the service before following the vendor repair steps in Third-party mail fails alignment. |
| Neither method supplies an aligned pass, and a report mentions a mailing list or trusted forwarder | An indirect flow the receiver recognized; modifications may have broken authentication | Investigate the path and a sample message. The override reason is evidence, not a promise about every delivery. |
| Neither method supplies an aligned pass, and no owner recognizes the source | Spoofing or another unauthorized path is possible; a missed device or application is also possible | Check internal owners and network ranges before classifying. Don't authorize the source to clear the warning. See Legitimate mail vs spoofing in reports. |
The distinction is passing with alignment, not just passing. DMARC needs an SPF or DKIM pass tied to the visible From domain. A valid signature from an unrelated domain doesn't establish authority to use yours.
Check an IP without assuming its owner sent the message
These are practical investigation methods, not guaranteed identification:
- Look up the reverse DNS name and IP allocation with PTR and RDAP or WHOIS tools.
- Compare the DKIM domain and selector with your configured services (DKIM selectors and DNS records).
- Ask marketing, finance, and IT whether they introduced a new platform or changed a sending region.
- Check printers, scanners, website forms, and applications as well as mailbox providers.
- Ask the suspected service owner to compare a real message and its delivery path with the report row.
A shared cloud address can belong to infrastructure used by many customers. Its network owner doesn't prove which customer sent the mail. Likewise, aligned DKIM isn't a reason to ignore an unknown source: it warrants checking how your signature or authorized account was used.
Non-Microsoft services sending on behalf of your domain are a common source of unexpected authentication failures. That supports checking for forgotten vendors; it doesn't establish how frequently unknown rows are legitimate or spoofed.
Record the conclusion before you change policy
Give each unresolved source an owner and a next check. Once identified, record the service, sending domain, selector, and authentication setup in your inventory (How to inventory sending sources). If it is legitimate but unaligned, fix the sender. If it is forwarding, assess that path. If the evidence supports spoofing, adding it to SPF is the opposite of the intended response.
You don't need an empty list of unknown rows before considering enforcement. You need enough evidence to distinguish legitimate paths from unauthorized use and understand the impact of the policy change (A safe rollout: p=none to quarantine to reject). Receivers can continue reporting spoofing attempts even when they honor a reject request; enforcement doesn't stop an attacker attempting to send.
Don't treat reports as exhaustive proof. Some receivers don't send them, mail rejected before DMARC evaluation can be absent, and report data may be forged. Use reports alongside message samples and service-owner confirmation, not as a substitute for them.
Free DMARC account
Not sure which sender is failing?
Connect your domain's DMARC reports and see every service that sends as your domain, and whether each one passes SPF, DKIM and DMARC.
- Every sending source in one dashboard
- SPF, DKIM and alignment results per sender
- No XML files to read
FAQ
Is an unknown IP in my DMARC report an attacker?
Not necessarily. Check whether DKIM passed for your domain, whether the IP belongs to a forwarder or a service you use, and ask the people who buy email tools.
Should I add the unknown IP to SPF?
Only after you've confirmed you own or contracted the service, and only if its Return-Path is your domain. Otherwise the change authorizes a stranger or does nothing for DMARC.
Why do mailbox-provider IPs show SPF failures for my domain?
Forwarding is one possible explanation: the forwarder's server isn't in your SPF record. If DKIM passed with alignment on those rows, DMARC passed (Forwarded mail fails DMARC).
Will these rows stop once we're at p=reject?
No. Receivers can keep reporting forged attempts. What changes is that receivers honoring your policy are asked to reject them.