How to read DMARC aggregate reports and move safely to p=reject

Set up the rua tag, read the XML reports mailbox providers send, spot unauthorized senders, and use the data to tighten DMARC from none to quarantine to reject.

· 7 min read · By the Spot Downtime team

A DMARC policy tells receiving mail servers what to do with messages that fail authentication for your domain. The reports are how you find out what those messages are. Without them, moving to p=quarantine or p=reject is guesswork, and a wrong guess means your own invoices, password resets or newsletters land in spam. If you need the basics of SPF, DKIM and DMARC first, read SPF, DKIM and DMARC explained.

Turn reports on

Reports are requested with the rua tag in your DMARC TXT record at _dmarc.yourdomain:

dns
_dmarc.example.com.  TXT  "v=DMARC1; p=none; rua=mailto:[email protected]"

p=none means "monitor only": nothing is blocked, and mailbox providers that support reporting start sending you daily summaries. Use a dedicated mailbox, because the volume is high and the content is machine-readable.

Reports to another domain need permission

If the rua address is on a different domain than the one being protected, the receiving domain must publish an authorization record, or providers will not send the reports. For reports about example.com sent to reports.example.net, publish a TXT record at example.com._report._dmarc.reports.example.net with the value v=DMARC1. Hosted DMARC services do this for you.

What a report is

A report is an XML file, usually compressed with zip or gzip and attached to an email. Each provider (Google, Microsoft, Yahoo and others) sends one per reporting period, normally a day, covering the mail they received that claims to be from your domain. It contains no message content, only counts grouped by source.

xml
<record>
  <row>
    <source_ip>203.0.113.25</source_ip>
    <count>1842</count>
    <policy_evaluated>
      <disposition>none</disposition>
      <dkim>pass</dkim>
      <spf>fail</spf>
    </policy_evaluated>
  </row>
  <identifiers>
    <header_from>example.com</header_from>
  </identifiers>
  <auth_results>
    <dkim><domain>example.com</domain><result>pass</result></dkim>
    <spf><domain>bounce.mailer.net</domain><result>pass</result></spf>
  </auth_results>
</record>

The fields that matter:

  • source_ip and count: which server sent how many messages.
  • policy_evaluated: the DMARC verdict. DMARC passes if either SPF or DKIM passes and aligns with the From domain.
  • disposition: what the receiver did, none, quarantine or reject.
  • auth_results: the raw SPF and DKIM results and which domains they were checked against.

Why alignment explains most failures

In the example above, SPF passed for bounce.mailer.net but is shown as a fail under policy_evaluated. SPF checks the envelope domain, and DMARC requires that domain to match the visible From domain. It doesn't, so SPF doesn't count. DKIM passed and aligned, so DMARC passes overall. This is why setting up DKIM signing with your own domain for every service is the reliable path.

Read them with a tool, not by eye

One busy domain can receive dozens of XML files a day. Parsing them by hand does not scale, so use a DMARC reporting service or an open-source parser that aggregates them by source. What you want is a table of sending sources with message counts and pass or fail per source.

Classify every source

Go through the sources for a couple of weeks and put each into one of three groups:

  1. Yours and passing: your mail server, transactional email provider, newsletter tool. Nothing to do.
  2. Yours and failing: a CRM, helpdesk, invoicing tool or old server that sends as you but isn't signing with your domain or isn't in SPF. Fix these before enforcing.
  3. Not yours: unknown IPs sending as your domain. These are spoofing attempts, which is what DMARC enforcement stops.

Identify an unknown IP with a reverse lookup (dig -x 203.0.113.25 +short) or a WHOIS query. When a legitimate service is failing, look at its setup guide for the DKIM CNAME records to add, and keep the number of SPF lookups under the limit, as described in SPF PermError: too many DNS lookups.

Tighten the policy in steps

  1. p=none while you collect reports and fix legitimate senders.
  2. p=quarantine once failures from your own services are gone. Receivers send failing mail to spam.
  3. p=reject when quarantine has run for a while without surprises. Failing mail is refused.
dns
_dmarc.example.com.  TXT  "v=DMARC1; p=quarantine; rua=mailto:[email protected]"

Keep reading the reports after each change. A new tool someone signs up for next quarter will show up as a failing source, and under p=reject its mail bounces. The error you will see on those bounces is often 550 5.7.26 unauthenticated email.

Subdomains inherit the main policy unless you set sp=. Domains that never send mail can go straight to p=reject, which stops them being used for spoofing.

Check what you publish

Email authentication checkerLook up the SPF, DKIM and DMARC records for a domain and see whether they are valid.

Watch the records themselves

DMARC, SPF and DKIM are DNS records, and a DNS edit by someone else can remove or break them without any visible error until mail starts failing. A DNS monitor in Spot Downtime watches records for changes, so an unexpected edit to your TXT records is flagged instead of discovered through bounced customer email.

Get new guides by email

Practical posts on monitoring, outages, error fixes and on-call, when they're published. No spam.

Unsubscribe from any email with one click. See our privacy policy.

Keep reading