What is SSL certificate monitoring?
Check the TLS certificate an endpoint actually serves, catch missed renewals, and understand why hostname, port, SNI, client trust, and CDN paths change what a certificate check can prove.
Renewal can succeed while clients still receive the old certificate. The web server may not have reloaded it, or the public hostname may terminate TLS at a CDN rather than on that server. SSL certificate monitoring checks the certificate a service actually presents and alerts before it expires. “SSL” is the familiar name; modern connections should use TLS. Expiry can cause browsers and API clients that validate certificates to reject new connections. Monitoring reduces the risk of a missed renewal, but cannot guarantee that every certificate incident will be prevented.
How certificate monitoring works
A monitor connects to a hostname and port, performs a TLS handshake, and reads the certificate. Checks might run hourly or daily, depending on certificate lifetime and the time needed to respond. The target also includes SNI and the access path. Multiple domains on one IP address can present different certificates; a connection without the right Server Name Indication (SNI) may receive the default certificate. See the OpenSSL connection and SNI options.
The checks available depend on the tool:
- Time to expiry. Compare the certificate's
notAfterwith the current time and alert before the remaining lifetime becomes too short. Thresholds of 30, 14, and 7 days are examples for longer-lived certificates, not a suitable default for every lifetime. - Certificate chain and trust. Check whether the server supplies the required intermediate certificates and whether the client can build a valid path to a trusted root. The client's trust store matters; counting the certificates sent by the server is not sufficient.
- Hostname matching. Match the requested domain against DNS identifiers in the certificate's Subject Alternative Name (SAN) extension. Checking the Common Name (CN) alone is insufficient. RFC 9525 defines TLS service identity verification.
- Protocol and cipher configuration. Detecting deprecated TLS 1.0/1.1 or weak cipher suites requires appropriate configuration review or dedicated tests. One successful modern TLS handshake does not prove that all older options are disabled. See the TLS 1.0/1.1 deprecation standard.
Why certificates expire—and why automatic renewal still fails
Limited lifetimes reduce the period of exposure after key compromise or mistaken issuance and encourage regular replacement. Renewal must still satisfy the CA's validation requirements; it does not necessarily mean a person revalidates the domain each time.
As of November 5, 2026, Let's Encrypt's classic profile issues 90-day certificates, tlsserver issues 45-day certificates, and shortlived uses 160 hours. A renewal schedule based on an unconditional 90-day lifetime will not fit all three. Check the selected profile, the profiles advertised by the ACME server, and the certificate actually issued. Let's Encrypt profiles
Configuring Certbot renewal leaves several possible failure points:
- A scheduled renewal fails, and nobody receives or acts on its error logs or notifications.
- A server migration leaves the renewal configuration behind.
- A DNS validation plugin loses access because its API credential has expired.
- A certificate renews successfully, but the web server continues serving the old one because it was not reloaded.
Monitor the deployed result as well as the renewal process. Issuing a certificate, updating files, and loading them into a running service are separate steps. Certbot's deploy hooks can run deployment commands after successful renewal; their outcome still needs checking. Certbot renewal and hooks
What happens when a certificate expires?
- Browsers: A validating browser may show a security error or block access. The exact behavior depends on the browser and site policy.
- API clients: Clients that validate the certificate may fail to establish new connections through the affected endpoint. That does not imply every integration is down.
- Search: HTTPS problems can affect crawling and the choice of canonical version. Inspect the actual Search Console results rather than assuming a fixed ranking penalty. Google's HTTPS report
- User trust: Failed access can disrupt business and damage confidence. The impact depends on which users and services rely on the affected endpoint.
Set alerts around the renewal window
Allow time for retries, deployment, and escalation. The following table is an example for a 90-day certificate, not a CA renewal policy. Adjust thresholds and channels to the normal renewal point, certificate lifetime, and team's response capacity.
| Time remaining | Severity | Example channel |
|---|---|---|
| 30 days | Reminder | Email or ticket |
| 14 days | Warning | Team messaging |
| 7 days | Urgent | Phone or an on-call group mention |
Inventory the endpoints using certificates: the main website, API domains, CDN custom domains, mail services such as SMTPS/IMAPS, and internal gateways. A forgotten subdomain can leave a gap. A CDN edge and its origin are separate TLS endpoints. Checking the public hostname will usually show the edge certificate. Different regions, IP addresses, or load-balancing paths may also serve different certificates.
Match the port and handshake method to the service. Direct TLS used by SMTPS/IMAPS differs from STARTTLS, which upgrades an application-protocol connection to TLS. An ordinary TLS handshake task cannot be assumed to test STARTTLS. An internal endpoint also needs a probe with network access; one public probe does not establish coverage of every entry point. OpenSSL STARTTLS options
In Guance SSL tests, configure the actual Host and Port, set availability conditions for certificate days remaining and response time, and choose nodes that represent the access paths you need. Overseas nodes require the Commercial Plan or above; self-built nodes are another option. After renewal, inspect the fingerprint and expiration time in the test details to check whether that node sees the replacement certificate. A changed fingerprint alone does not establish correct identity, trust, or application behavior. If server certificate errors are ignored, a successful result is not proof of client trust. These tasks do not replace a full TLS configuration security review.
Common questions
Which thresholds should I use for Let's Encrypt certificates?
Start with the actual lifetime and renewal plan. Let's Encrypt recommends clients that support ACME Renewal Information (ARI), which supplies a renewal window. Without ARI, automation can renew approximately two-thirds of the way through the certificate's lifetime. Alert when expected renewal has not completed, with enough time left for retries, deployment, and a person to intervene. Do not copy 30/14/7-day thresholds onto 45-day or 160-hour certificates. Let's Encrypt renewal guidance
Do internal certificates need monitoring?
Yes. Expiry can break connections from internal clients that validate the affected certificate. The dependency path determines the scope; it is not necessarily an organization-wide outage. Internal PKI can also automate issuance and rotation. Monitor what is actually deployed, using a node that can reach the endpoint and trusts the appropriate CA.
Who owns a CDN certificate's renewal?
Check the certificate type and service agreement. Cloudflare manages issuance and renewal for offerings such as Universal SSL, while customers must update uploaded custom certificates. Assign ownership and monitoring for the origin certificate separately. A healthy edge certificate does not establish that the origin certificate is healthy. Cloudflare custom certificate renewal · Strict origin validation
Contact us
Join the community
to join the community
Try Guance
Start online and pay only for what you use.
Get startedChoose a Guance plan
