Advanced topics
What is TLS-RPT? SMTP TLS reporting explained
By DMARCHub Team6 min read
On this page
TLS-RPT (SMTP TLS Reporting, defined in RFC 8460) is a standard that lets domain owners receive daily reports about the success or failure of encrypted connections to their mail servers. You enable it with a single DNS TXT record, and sending providers then send you JSON summaries of any TLS problems they hit when delivering to your domain. It is the reporting companion to MTA-STS.
Why does TLS-RPT exist?
Encryption between mail servers uses an SMTP extension called STARTTLS, and it is opportunistic by default: if the receiving server does not offer TLS, or the negotiation fails, the sending server usually delivers the message in plain text anyway. That keeps mail flowing, but it also means an attacker who can interfere with the connection can strip the encryption without anyone noticing.
MTA-STS (RFC 8461) fixes the enforcement side. It lets a domain publish a policy saying, in effect, "always require TLS with a valid certificate when delivering to me". But enforcement without visibility is risky. If your certificate expires, or your MTA-STS policy file becomes unreachable, mail from enforcing senders starts failing silently, and the senders are the only ones who can see it. TLS-RPT closes that gap: it gives those senders a standard way to tell you what they observed.
The two standards are designed as a pair, but TLS-RPT works on its own too. Publishing it before MTA-STS is a sensible first step: it shows you the current state of TLS delivery with zero risk to mail flow, mirroring the way DMARC starts at p=none before moving to enforcement.
How do you set up the TLS-RPT DNS record?
TLS-RPT is enabled with one TXT record at the subdomain _smtp._tls of your domain:
_smtp._tls.example.com. IN TXT "v=TLSRPTv1; rua=mailto:tls-reports@example.com"
The record has two parts:
v=TLSRPTv1- the version tag, which must come first.rua=- where reports should be delivered. This can be amailto:address or anhttps:endpoint that accepts a POST. You can list more than one destination, separated by commas.
A few practical notes. The report address can be on a different domain from the one publishing the record. Keep exactly one TLS-RPT record at the name; multiple records make the result undefined, the same failure mode as multiple SPF records. And if you point rua at a mailbox on the same domain, think through the failure case: a TLS problem severe enough to block delivery to your domain can also block delivery of the report telling you about it. An address on a separate domain, or an HTTPS endpoint, avoids that circularity.
What do TLS-RPT reports contain?
Reports are JSON documents, normally gzip-compressed, sent once per day per reporting organisation. Each report covers a 24 hour window and summarises every delivery attempt the sender made to your domain in that window. They contain connection statistics only, not message content.
The useful fields inside a report are:
- organisation and date range - who is reporting (for example
google.com) and the window covered. - policy details - which policy the sender applied: your MTA-STS policy, a DANE TLSA record, or no policy at all.
- success and failure counts -
total-successful-session-countandtotal-failure-session-countfor each policy. - failure details - for each failure type, the sending and receiving IPs involved and how many sessions failed.
The failure result types are standardised, and they tell you where to look:
| Result type | What it means |
|---|---|
starttls-not-supported |
Your MX did not offer STARTTLS at all. |
certificate-expired |
The MX presented an expired certificate. |
certificate-host-mismatch |
The certificate does not cover the MX hostname. |
validation-failure |
Other certificate validation problems, such as an untrusted issuer. |
sts-policy-fetch-error |
The sender could not fetch your MTA-STS policy file over HTTPS. |
sts-policy-invalid |
Your policy file was fetched but could not be parsed or applied. |
sts-webpki-invalid |
The certificate on your MTA-STS policy host failed validation. |
A healthy domain shows large successful session counts and zero or near-zero failures. In practice most domains only receive reports from a handful of large senders, so absence of reports from a given provider does not by itself mean absence of problems.
How should you act on TLS failure reports?
Treat each failure type as a pointer to a specific piece of infrastructure:
- Certificate failures (
certificate-expired,certificate-host-mismatch,validation-failure) point at your mail server. Check that the certificate on each MX host is current, covers the exact MX hostname, and chains to a trusted authority. Automated renewal that quietly stopped working is a common cause. - Policy fetch failures (
sts-policy-fetch-error,sts-webpki-invalid) point at the web host servinghttps://mta-sts.yourdomain/.well-known/mta-sts.txt. That host needs a valid certificate of its own and must stay reachable; an expired certificate there breaks policy discovery even when your mail servers are fine. - Policy content failures (
sts-policy-invalid) usually mean the policy file's MX list no longer matches your real MX records, often after a mail provider migration. Update the policy file and bump theidin your_mta-stsTXT record so senders refetch it. starttls-not-supportedmeans an MX host is not offering encryption at all. That needs fixing before any move to MTA-STS enforce mode, because enforcing senders will refuse to deliver to it.
A steady trickle of failures from one sending IP range, while everything else succeeds, can also indicate interference on a specific network path rather than a fault at your end. That visibility is exactly what TLS-RPT was designed to provide.
The reports themselves are machine-readable JSON, which is accurate but not pleasant to read in a mailbox, and the same is true of DMARC's XML aggregate reports. Many teams point both at a monitoring service; DMARCHub, for example, ingests TLS-RPT alongside DMARC reports and surfaces the failures as readable dashboards (see /features).
Where does TLS-RPT fit alongside DMARC?
TLS-RPT and DMARC solve different problems. DMARC (with SPF and DKIM) proves who sent a message and protects your domain from spoofing; its rua reports describe authentication results. TLS-RPT says nothing about who sent a message; it reports on whether the connection carrying mail to your domain was properly encrypted. A complete posture typically includes SPF, DKIM and DMARC for authentication, plus MTA-STS and TLS-RPT for transport security.
Frequently asked questions
Can I use TLS-RPT without MTA-STS?
Yes. TLS-RPT reports on TLS negotiation regardless of whether an MTA-STS or DANE policy exists, and reporters will tell you which policy, if any, they applied. Publishing TLS-RPT first is the recommended order: it lets you confirm your mail servers negotiate TLS cleanly before you publish an enforcing policy that depends on it.
How often do TLS-RPT reports arrive?
RFC 8460 specifies one report per reporting organisation per day, covering a 24 hour UTC window. A domain with normal traffic typically sees a few daily reports from the large mailbox providers; low-traffic domains may only see reports on days when those providers actually attempted delivery.
Do TLS-RPT reports contain message content or personal data?
No. Reports contain aggregate session counts, policy details, failure result types and the server IP addresses involved. There are no subject lines, correspondent addresses or message bodies, which makes TLS-RPT far less sensitive than DMARC failure (ruf) reports.
Does TLS-RPT affect email delivery?
No. The record is purely an invitation to send you reports; it changes nothing about how senders deliver mail to you. Delivery behaviour only changes when you publish an enforcing MTA-STS policy or DANE records. That makes TLS-RPT one of the few email security records you can deploy with no rollout risk at all.