Skip to content
AdminKit

How to read a TLS/SSL certificate check result

A TLS/SSL certificate check returns specific facts: who issued the certificate, which hostnames it covers, when it expires, and whether the chain to a trusted root is complete. Connection safety depends on reading those fields, not on a green padlock. AdminKit's SSL/TLS check shows the issuer, names and expiry, plus the negotiated protocol and cipher.

By AdminKit team, Netbase JSC · Updated

Key takeaways

  • A certificate check summarises a handful of RFC 5280 fields: subject names, issuer, validity dates, chain and algorithms.
  • Clients trust a connection when three things hold: the hostname is in the SAN list, today falls inside the validity window, and the chain reaches a trusted root.
  • Publicly trusted certificates issued since 15 March 2026 can be valid for at most 200 days. That drops to 100 days in March 2027 and 47 days in March 2029.
  • AdminKit’s SSL/TLS check reports names, issuer, expiry, self-signed status and the negotiated protocol and cipher. It does not validate the chain. Use openssl s_client for that.
  • “Valid” means the certificate is fine today. It does not mean the whole TLS setup is sound.

What’s actually in a certificate check result?

A TLS/SSL certificate, as defined by RFC 5280 (canonical record on the IETF Datatracker), carries a fixed set of fields regardless of which tool reads it, and a check result is really just a human-readable summary of those fields:

  • Subject — the hostname(s) the certificate is issued for, listed in its Subject Alternative Name (SAN) extension.
  • Issuer — the Certificate Authority (CA) that signed the certificate.
  • Validity period — a notBefore and notAfter date, outside of which the certificate is not valid regardless of anything else being correct.
  • Chain — the sequence of intermediate certificates linking the server’s certificate back to a root CA that the connecting client already trusts.
  • Key and signature algorithm — what cryptography protects the certificate itself, separate from the TLS protocol version used for the connection. That protocol version is negotiated over a regular TCP connection; QUIC (the transport HTTP/3 runs on) carries its own TLS 1.3 handshake over UDP instead, which this check does not test.

Different tools label these fields differently. The table maps the usual labels to the RFC 5280 field and shows which ones AdminKit’s check displays:

What a tool showsRFC 5280 fieldShown by AdminKit’s SSL/TLS check
“Names”, “SANs”, “Covers”subjectAltName extension (dNSName entries)Yes, as a list
“Subject”, “Common Name”subject (CN attribute)Yes, the CN
“Issuer”, “Issued by”issuerYes, the issuer CN
“Valid from / until”, “Expires”validity (notBefore, notAfter)Yes, with days remaining
“Chain”, “Trust path”Path built from the server’s certificate listNo
“Signature algorithm”signatureAlgorithmNo

Which fields decide whether a connection is trusted?

Not every field carries equal weight. Three checks, done in combination, are what actually determine whether a client accepts the connection:

  1. Does the hostname being connected to appear in the certificate’s SAN list? RFC 9525, which obsoleted RFC 6125 in 2023, says the server’s identity can only be expressed in the SAN extension; the Common Name is no longer valid for this. A certificate issued for example.com does not automatically cover api.example.com — that subdomain needs its own entry or a matching wildcard (*.example.com).
  2. Is today’s date inside the validity window? An expired certificate fails regardless of how well everything else checks out, and the CA/Browser Forum’s Baseline Requirements cap how long a publicly trusted certificate can remain valid in the first place, specifically so that key compromises and CA trust changes propagate on a bounded schedule rather than lingering for years.
  3. Does the chain resolve to a trusted root? A server needs to present not just its own certificate but every intermediate certificate between it and a root the client already trusts. A result showing “chain incomplete” or “unable to get local issuer certificate” almost always means the server is missing an intermediate in its configuration, not that the certificate itself is fraudulent.

The validity cap has been shrinking. Under Ballot SC081v3, now part of the Baseline Requirements, the maximum is:

Certificate issuedMaximum validity
Before 15 March 2026398 days
15 March 2026 to 14 March 2027200 days
15 March 2027 to 14 March 2029100 days
From 15 March 202947 days

The practical effect: a certificate you install today expires in well under a year. Renewal has to be automated, usually over the ACME protocol, or it will be missed.

What does AdminKit’s SSL/TLS check report?

AdminKit’s SSL/TLS check completes one TLS handshake with the host you enter and reads the certificate the server presents. It shows the subject and issuer common names, the SAN list, the validity dates with days left, whether the certificate is self-signed, and the protocol version and cipher the handshake agreed on. Then it raises findings:

FindingSeverityWhen it appears
Certificate expiredcriticalnotAfter is in the past
Certificate expires in N day(s)high14 days or fewer left
Certificate expires in N day(s)medium15 to 30 days left
Certificate does not cover the hostnamehighThe name you entered matches no SAN entry
Deprecated TLS/SSL versionhighThe handshake used SSLv2, SSLv3, TLS 1.0 or TLS 1.1
Certificate is self-signedmediumSubject and issuer are identical

The overall status is a fail if any finding is high or critical, a warning if all findings are lower, and OK if there are none. TLS 1.0 and 1.1 are flagged because RFC 8996 formally deprecates them.

Three limits matter when you read the result. The check reports one handshake, not every protocol version the server would accept. It reads only the server’s own certificate and does not validate the chain to a root store. If a certificate has no SAN entries at all, it compares the hostname against the Common Name, which clients following RFC 9525 no longer do. Treat a certificate without SANs as broken whatever the hostname result says.

What do common results mean?

  • “Self-signed certificate” — the subject and issuer are the same entity, meaning no third-party CA vouched for it. Expected and fine for internal tooling you control; a red flag on anything meant to be reached by the public, since no outside party verified the operator controls that hostname.
  • “Certificate expired” — the notAfter date has passed. This is purely a date comparison and the single most common reason a previously-working site suddenly shows a browser warning.
  • “Hostname mismatch” — the name you connected to isn’t in the certificate’s SAN list, per RFC 9525’s matching rules. Common after a domain migration or a wildcard certificate that doesn’t cover a newly added subdomain.
  • “Weak signature algorithm” — reported by chain-validating tools, not by AdminKit’s check. The certificate (or one in its chain) uses a signature algorithm, like SHA-1, that current Baseline Requirements and browser policy no longer treat as acceptable, regardless of key length. The Baseline Requirements set 15 September 2026 as the date to sunset all remaining use of SHA-1 in certificates and CRLs.
  • “Deprecated protocol” — the server agreed to TLS 1.0 or 1.1. The certificate may be fine; the fix is in the server’s TLS settings.

How do you check the chain and the algorithm yourself?

For the fields AdminKit’s check does not cover, OpenSSL’s s_client gives you the raw material. The manual is careful about one detail: -showcerts prints the certificate list “as sent by the server”, and “it is not a verified chain”. The verification result appears separately, near the end of the output. The output below is an illustrative, shortened example for the documentation domain example.com, not a capture from a real server:

$ openssl s_client -connect example.com:443 -servername example.com -showcerts </dev/null
Certificate chain
 0 s:CN = example.com
   i:C = US, O = Example CA, CN = Example Issuing CA 1
 1 s:C = US, O = Example CA, CN = Example Issuing CA 1
   i:C = US, O = Example CA, CN = Example Root CA
...
    Verify return code: 0 (ok)

Read it top to bottom. Entry 0 is the server’s certificate, with its subject (s:) and issuer (i:). Each next entry should be the issuer of the one above it. Verify return code: 0 (ok) means OpenSSL built a path to a root in its trust store. A code such as 20 (unable to get local issuer certificate) usually points at a missing intermediate; see checking TLS certificate chain completeness. To see the signature algorithm, save entry 0 to a file and run openssl x509 -noout -text on it.

How do you check your own certificate?

Start with AdminKit’s SSL/TLS check: it reads the certificate’s subject, issuer, names and validity window, and the negotiated protocol and cipher, and it flags the ones that would make a real client reject the connection, not just anything unusual. Then run the openssl command above for the chain. If the hostname itself is also in question (for example, confirming which server a name currently resolves to before you check its certificate), start with AdminKit’s DNS lookup tool. See AdminKit’s SSL/TLS check vs Qualys SSL Labs if you’re deciding which one to run first, or whether you need both, or best free SSL/TLS certificate checker tools for how AdminKit’s check compares to the wider category of free checkers. See our developers and SaaS teams page for the fuller picture of checking a certificate alongside headers and MCP exposure.

Why does “valid” not mean “secure”?

A certificate check that returns “valid” answers a narrower question than it sounds like: it confirms the cryptographic chain and hostname match hold up today, not that the server configuration is otherwise sound, that the private key hasn’t been separately compromised, or that the TLS versions and cipher suites offered are still considered strong. Treat a passing certificate check as one necessary condition among several, not a complete transport-security audit — see our guide on mail server STARTTLS enforcement for an example of a transport-layer gap that a valid certificate alone does nothing to close. Who may issue for your domain is a separate control again, covered in what CAA records are. For the rest of the checks that make up a fuller picture of a site’s exposed surface, see AdminKit’s security basics cluster.

This guide is general information, not a security audit of your specific network — see our disclaimer for what AdminKit's guidance does and does not commit to.

Frequently asked questions

What's the difference between the certificate's subject and its issuer?

The subject is who the certificate was issued to — the hostname(s) it's meant to protect. The issuer is the Certificate Authority (CA) that signed it, vouching that it verified the subject's control over those hostnames before issuing it. A self-signed certificate has the same value in both fields, which is a signal to look closer, not necessarily a sign of malice.

Why does a certificate's validity period matter beyond just 'not expired yet'?

The CA/Browser Forum's Baseline Requirements, which every publicly trusted CA must follow, cap how long a certificate can remain valid (200 days for certificates issued since 15 March 2026) specifically so that compromised keys and revoked trust get replaced on a predictable cycle. A certificate that's technically still valid but close to expiring is a maintenance signal worth acting on before it becomes an outage.

What does it mean if the hostname I'm checking isn't in the certificate's SAN list?

It means the connection will fail hostname verification even if every other part of the certificate is valid — RFC 9525, which replaced RFC 6125 in 2023, specifies that a client must match the hostname it's connecting to against the certificate's Subject Alternative Names; the Common Name no longer counts. A certificate valid for 'example.com' doesn't automatically cover 'api.example.com' unless that name, or a matching wildcard, is explicitly listed.

What does an 'incomplete chain' result actually mean?

It means the server isn't sending one or more intermediate certificates needed to link its certificate back to a root that browsers trust directly. Some browsers tolerate this by fetching the missing intermediate themselves, but many clients — especially non-browser ones like mobile apps, API clients, and mail servers — will not, and will simply reject the connection.

Is a certificate from a free CA like Let's Encrypt less trustworthy than a paid one?

No — trust comes from whether the CA is in a browser's or OS's trusted root program and follows the CA/Browser Forum's Baseline Requirements, not from price. Free, automated CAs generally only validate domain control (not organization identity), which is a narrower validation level than some paid certificates offer, but that's a difference in what's validated, not in cryptographic trustworthiness.

Want AdminKit to watch this for you?

Free checks tell you the state today. Protective DNS and monitoring tell you when it changes.

Huy Nguyen (David), Founder and CEO of Netbase JSC

Talk to an expert

Huy Nguyen (David), Founder & CEO, Netbase JSC

Ask whether this fix applies to your setup, or get a second pair of eyes on your DNS and mail records.