TLS Certificate Monitoring and Renewal Runbook for Websites, APIs, and Load Balancers
TLSSSL certificatesACMEcertificate monitoringwebsite security

TLS Certificate Monitoring and Renewal Runbook for Websites, APIs, and Load Balancers

SStorageTech Editorial Team
2026-11-16
7 min read

Use this TLS certificate monitoring and renewal runbook to prevent expired or misconfigured certificates from disrupting websites and APIs.

An expired TLS certificate can make a healthy website, API, CDN, reverse proxy, or load balancer appear unavailable. This runbook gives you a repeatable way to inventory certificates, assign ownership, monitor expiry, renew safely, and verify that the new certificate is deployed across every public endpoint.

Overview

TLS certificates establish encrypted connections and help clients verify that they are communicating with the intended hostname. When a certificate expires, has the wrong names, is missing an intermediate certificate, or is not deployed consistently, browsers and API clients may reject the connection. The result can range from browser warnings to failed mobile applications, broken webhooks, and unavailable administrative systems.

Preventing an incident requires more than setting a calendar reminder. A useful certificate-management process combines four controls:

  • An inventory: every certificate, hostname, endpoint, issuer, renewal method, and owner is recorded.
  • Monitoring: expiry and endpoint behavior are checked from outside the infrastructure, not only from the server that performs renewal.
  • Ownership: a named team or person is accountable for renewal, deployment, and escalation.
  • Verification: renewal is followed by checks on every relevant listener, protocol, certificate chain, and client path.

This approach applies whether you use ACME certificate renewal on a web server, certificates supplied by a cloud hosting platform, a managed CDN, a Kubernetes ingress controller, or a manually operated load balancer. The exact interface and renewal timing vary by platform, so treat provider documentation as the final authority for implementation details.

What to track

Build a certificate inventory

Start with a single inventory that covers production, staging, internal services, and less frequently used endpoints. Include certificates installed on origin servers even when a CDN terminates public TLS, because a separate certificate may still be required between the CDN and origin.

Use fields such as these:

FieldWhat to record
Service and environmentWebsite, API, admin portal, staging, production, or internal service
HostnamesCommon name and all relevant subject alternative names
Public endpointHostname, IP, CDN, reverse proxy, or load balancer listener
Certificate locationWeb server, ingress secret, CDN, control panel, or load balancer
Issuer and typeCertificate authority, ACME-managed certificate, or manually supplied certificate
Not-before and expiry datesDates observed from the deployed endpoint, not only from an internal file
Renewal methodAutomatic, scheduled job, platform workflow, or manual process
Validation methodHTTP-01, DNS-01, provider validation, or another approved method
Owner and backupPrimary operator, secondary operator, and escalation channel
Last verifiedDate and result of the most recent external and internal checks

Do not rely on filenames such as cert-new.pem or final-2025.pem as your source of truth. Files can be copied, renamed, or left unused. The certificate presented by the endpoint is what customers and API clients receive.

Monitor more than the expiry date

Certificate monitoring should report the remaining validity period, hostname coverage, chain status, and connection result. A certificate can be unexpired but still fail because it does not include the requested hostname, the private key does not match, an intermediate certificate is missing, or a load balancer is serving an older certificate on one listener.

For each monitored endpoint, check:

  • The certificate's expiration date and remaining days.
  • Whether the requested hostname appears in the certificate's subject alternative names.
  • Whether the certificate chain is accepted by the clients you support.
  • Whether TLS negotiation succeeds on the intended port and hostname.
  • Whether IPv4, IPv6, CDN, origin, and alternate load-balancer paths behave consistently.
  • Whether the endpoint redirects or routes as expected after the TLS handshake.

A practical alert policy uses multiple thresholds rather than one last-minute notification. Choose thresholds that match your renewal method, approval process, and incident response time. Automatic renewal still needs an alert for failed validation, while manual renewal needs enough lead time for procurement, domain validation, deployment, and rollback.

Cadence and checkpoints

Monthly certificate review

Once a month, review the inventory and monitoring results. Confirm that every production hostname is represented, owners are still correct, and no endpoint has been added without a certificate record. Sort the list by expiration date and investigate any certificate inside your warning window.

During the same review, check renewal job health. For ACME-based automation, inspect recent logs, challenge validation results, account permissions, DNS credentials, and the location where the renewed certificate is installed. A successful certificate issuance does not necessarily prove that a web server, ingress controller, CDN, or load balancer reloaded the new files.

Quarterly operational check

At least quarterly, test the full path for important services. Confirm the certificate presented by the public hostname, then check the relevant origin or internal listener separately. Review firewall rules, DNS records, certificate secrets, deployment permissions, and scheduled jobs for changes.

For systems managed as code, compare the actual configuration with the intended configuration. Certificate references, listener bindings, secret names, and reload hooks can drift during infrastructure changes. The Infrastructure as Code Drift Detection Guide provides a useful framework for finding that type of mismatch.

Before a change

Run a certificate checkpoint before changing domains, DNS, CDN configuration, ingress rules, reverse proxies, hosting providers, or load balancers. Record the current certificate fingerprint or serial number, confirm the renewal method at the destination, and identify the rollback path.

For a migration, lower-risk preparation usually includes issuing or staging the replacement certificate before the traffic switch, validating domain ownership through the planned method, and testing the destination with a temporary hostname or direct endpoint where appropriate. The Cloud Hosting Migration Checklist can be used alongside this certificate-specific review.

After every renewal

Do not close a renewal task when the certificate authority reports success. Verify deployment from an external network and, where possible, from the service's internal path. Confirm the new expiry date, hostname coverage, chain, listener, and application response. Record the result in the inventory and retain relevant logs without storing private keys in the ticket or monitoring system.

How to interpret changes

When a monitor reports a change, first determine whether it is an expected certificate rotation or an unplanned deployment. Compare the endpoint's certificate details with the inventory and recent release records. A changed serial number with the expected issuer and hostnames may be normal; a changed issuer, missing hostname, or unexpected validity period requires investigation.

Use a command-line check to inspect what a host is actually presenting:

openssl s_client -connect example.com:443 -servername example.com -showcerts < /dev/null 2>/dev/null | openssl x509 -noout -subject -issuer -dates -serial -ext subjectAltName

The -servername option matters when several domains share an IP address. Without the correct Server Name Indication value, you may inspect a default certificate rather than the certificate used by the requested hostname. For an API, repeat the check against the API hostname and port used by clients. If a CDN or load balancer is involved, test the public address and the origin path separately.

Common findings have different remedies:

  • Expired certificate: renew or replace it, deploy the replacement, reload the relevant service, and confirm all listeners.
  • Wrong hostname: request a certificate containing the exact DNS name used by clients, including required alternate names.
  • Incomplete chain: install the certificate and the intermediate chain in the format required by the serving platform.
  • Only one endpoint is updated: inspect DNS, IPv6, CDN regions or configurations, load-balancer listeners, and origin pools.
  • Renewal succeeds but deployment does not: check file permissions, secret synchronization, reload hooks, and whether the service reads certificates only at startup.
  • Validation fails: review DNS records, HTTP routing, firewall access, authorization credentials, and recent domain changes.

Keep TLS protocol and cipher configuration changes separate from certificate renewal unless there is a defined compatibility test. A certificate replacement should not silently introduce an unrelated protocol change. If you are also reviewing CDN behavior, pair the certificate check with the CDN Cache-Control Guide so that security and content-delivery changes are evaluated together without confusing their symptoms.

When to revisit

Use this runbook on a monthly or quarterly schedule, and revisit it immediately after an infrastructure or domain change. Update the procedure when you change certificate authorities, ACME clients, DNS providers, hosting platforms, CDNs, ingress controllers, load balancers, or deployment tools. Platform interfaces and automation behavior can change, so confirm current provider instructions rather than assuming that an older command or renewal workflow still applies.

Set a calendar checkpoint at least twice a year to review the document itself. Confirm that supported client requirements, certificate validation methods, monitoring thresholds, escalation contacts, and rollback steps still match your environment. Also review certificates that are easy to overlook: wildcard domains, regional endpoints, mail or webhook services, staging systems exposed to partners, and administrative portals.

For the next review, take these actions:

  1. Export or open the current certificate inventory.
  2. Compare it with DNS, CDN, load-balancer, and hosting records.
  3. Sort certificates by expiration date and resolve missing owners.
  4. Test one public endpoint from outside the hosting network.
  5. Inspect recent renewal logs and confirm deployment or reload success.
  6. Document any certificate, hostname, listener, or chain discrepancy.
  7. Update the runbook after the issue is resolved, not months later.

Certificate monitoring is most effective when it is treated as an operational control rather than a one-time security task. A maintained inventory, clear ownership, early alerts, and post-renewal endpoint checks substantially reduce the chance that an otherwise healthy website or API is interrupted by an avoidable TLS failure.

Related Topics

#TLS#SSL certificates#ACME#certificate monitoring#website security
S

StorageTech Editorial Team

Cloud Infrastructure Editors

Senior editor and content strategist. Writing about technology, design, and the future of digital media. Follow along for deep dives into the industry's moving parts.