Kubernetes Ingress Controller Fake Certificate served instead of yours

Quick fix

Read the commands before running them. Anything that restarts a service, deletes data or changes permissions should be tried on a non-production system first.

# What certificate is actually served for this hostname?
openssl s_client -connect app.example.com:443 -servername app.example.com </dev/null 2>/dev/null | openssl x509 -noout -subject -issuer -dates

# The secret must exist in the same namespace as the Ingress
kubectl get ingress app -n prod -o jsonpath="{.spec.tls}"; echo
kubectl get secret app-tls -n prod -o jsonpath="{.type}"; echo

# Hosts in spec.tls must match spec.rules exactly
kubectl get ingress app -n prod -o yaml | grep -A3 -E "tls:|host:"

# The controller logs why it fell back
kubectl logs -n ingress-nginx deploy/ingress-nginx-controller | grep -i -E "ssl|certificate" | tail

# cert-manager: is the certificate actually ready?
kubectl get certificate -n prod

How to diagnose Ingress errors

An Ingress failure is a routing failure, and there are only a few places it can break: the IngressClass is missing so no controller claims the resource, the path or host does not match, the backing Service has no endpoints, or the TLS secret is absent or in the wrong namespace. The "default backend - 404" page is the controller saying it received the request but found no rule for it.

If the quick fix above does not resolve it, work through these steps. They apply to this whole class of error, not just to this one message, which is usually what saves the time.

1. Work backwards from the pod: kubectl get endpoints <service>. If it is empty, the Service selector does not match any ready pod and no Ingress configuration will help.
2. Check that a controller has claimed the Ingress: kubectl describe ingress <name> should show events and an assigned address. No address means no controller is watching that IngressClass.
3. Read the controller's own logs (kubectl logs -n ingress-nginx deploy/ingress-nginx-controller). They log rejected configuration and certificate problems explicitly.
4. Remember TLS secrets must live in the same namespace as the Ingress. This is the single most common TLS mistake.
5. Verify path type semantics: Prefix, Exact and ImplementationSpecific match differently, and regex behaviour varies between controllers.

Tools worth reaching for

• kubectl describe ingress
• kubectl get endpoints
• controller logs
• curl -H 'Host: …'
• openssl s_client -servername

Authoritative references

Primary documentation for this error, worth reading before applying any fix in production.

📖 kubernetes.github.io 📖 kubernetes.io

Related Ingress errors

• AWS Load Balancer Controller: couldn't auto-discover subnetsThe controller builds an ALB by finding subnets tagged for load balancers, and none in the…
• cert-manager: Failed to determine a valid solver configurationNone of the solvers on the Issuer matches the domains being requested, so cert-manager has no…
• cert-manager: Order is pending / challenge failed to propagatecert-manager created an ACME order but the challenge never validated. For HTTP-01 the…
• Gateway API: HTTPRoute Accepted=False, reason NoMatchingParentThe HTTPRoute names a Gateway in parentRefs, but no listener on that Gateway will take it, so…
• ingress-nginx: 400 The plain HTTP request was sent to HTTPS portThe controller is speaking plain HTTP to a backend that only accepts HTTPS, and the backend's…
• ingress-nginx: failed calling webhook "validate.nginx.ingress.kubernetes.io"The API server tried to ask the ingress-nginx admission webhook whether your Ingress is valid…
• ingress-nginx: host "example.com" and path "/" is already defined in ingressThe admission webhook rejected the Ingress because another Ingress, possibly in a different…
• ingress-nginx: Service does not have any active EndpointThe controller resolved the Ingress to a Service but the Service selects no ready Pods, so it…

See all 20 Ingress errors →

Browse other categories

• HTTP 494xx client errors, 5xx server errors, redirects, headers and protocol problems.
• JavaScript 42npm resolution, async pitfalls, hydration, memory limits and runtime type…
• Database 41Connections, deadlocks, constraints, replication and memory limits.
• AI 35Rate limits, context windows, GPU memory and model-serving failures.
• Network 35Refused connections, timeouts, resets, MTU problems and port exhaustion.
• Python 35Imports, virtual environments, encoding, concurrency and dependency conflicts.
• Kubernetes 34CrashLoopBackOff, ImagePullBackOff, OOMKilled, RBAC, scheduling and storage.
• Docker 27Daemon connectivity, disk space, image pulls, ports and architecture mismatches.
• System 26Disk space, systemd units, file descriptors, OOM killer and scheduled jobs.
• Cloud 25IAM permissions, quotas, service limits and credential failures.
• Security 25JWT validation, CSRF, OAuth grants, SELinux, SSH host keys and CSP.
• TLS 24Untrusted authorities, expiry, hostname mismatch, chains and cipher negotiation.

Something missing or wrong?

This entry is maintained by hand. If the fix is out of date, incomplete, or you have a better one, email a correction and it will be reviewed.