Identify the certificate actually presented by the affected endpoint and validate its chain before adding anything to the clients’ trusted-root store.
1. Capture the affected endpoint and presented certificate
On one affected domain-joined client:
- Record the exact DNS name and port being accessed, such as
server.contoso.com:443. - Open the certificate from the client application or browser and record:
- Subject
- Issuer
- Thumbprint
- Valid from/to
- Subject Alternative Name (SAN)
- Certification Path status
- Confirm that the DNS name used by the client appears in the SAN.
A trusted chain must lead to a trusted root, with every certificate correctly signed, unexpired, unrevoked, and compliant with the issuing CA’s policies.^1^
The path status distinguishes common cases:
- Untrusted root/CA: the root CA is absent or deliberately untrusted.
- Incomplete chain: an intermediate CA certificate is missing.
- Expired certificate: replace the expired server, intermediate, or root certificate.
- Name mismatch: reissue the server certificate with the accessed DNS name in its SAN.
- Revocation failure: verify access to the CRL or OCSP locations encoded in the certificate.
Do not install the server certificate itself as a trusted root; an incorrect trusted-root change weakens security for the entire computer.^2^
2. Determine which service presents it
If the connection uses HTTPS or another HTTP.sys-hosted service, run this elevated on the server:
netsh http show sslcert
Get-ChildItem Cert:\LocalMachine\My |
Select-Object Subject, Issuer, Thumbprint, NotAfter,
HasPrivateKey
Match the binding’s Certificate Hash to the certificate thumbprint. netsh http show sslcert displays the SSL certificate associated with each IP address and port.^3^
For IIS:
- Open IIS Manager.
- Select Sites → the relevant site → Bindings.
- Edit the
httpsbinding. - Compare its selected certificate with the certificate observed on the client.
IIS 7 and later also store SSL binding information in HTTP.sys, so both the IIS binding and HTTP.sys certificate hash/store must be correct.^4^
If no matching HTTP.sys binding exists, identify the process listening on the affected port and inspect that product’s TLS configuration. The available information does not establish whether this is IIS, LDAPS, Remote Desktop, SQL Server, or another service.
3. Validate the server certificate and chain
On the server, run certlm.msc, then inspect:
- Personal → Certificates for the server certificate.
- Intermediate Certification Authorities → Certificates for intermediates.
- Trusted Root Certification Authorities → Certificates for the root.
Open the server certificate and check Certification Path, SAN, validity dates, Server Authentication usage, and whether the certificate has its private key. Windows documents certlm.msc as the Local Computer store and provides read-only PowerShell inspection for these properties.^2^
If the presented certificate has been exported as server.cer, validate it from an approved troubleshooting environment:
certutil -urlfetch -verify .\server.cer
This checks the certificate while retrieving its embedded validation URLs; those URLs can generate outbound traffic.^2^
Also review Event Viewer → Windows Logs → System for Schannel errors at the time of a failed connection.
4. Distribute an internal CA only when the chain proves it is missing
If the certificate was issued by your organization’s internal CA and the affected clients lack that approved root or intermediate certificate:
- Verify the CA certificate’s issuer and thumbprint with the PKI administrator.
- Open Group Policy Management on a domain controller or RSAT management device.
- Create or edit a GPO linked to the affected clients’ domain, site, or OU.
- Navigate to:
Computer Configuration → Policies → Windows Settings → Security Settings → Public Key Policies - Import:
- The root CA into Trusted Root Certification Authorities.
- Any intermediate CA into its appropriate intermediate store.
- On a test client, run:
gpupdate /force
- Reopen
certlm.mscand confirm that the expected CA thumbprints appear, then retest the connection.
Group Policy is the supported domain mechanism for distributing certificates to Windows devices.^5^ Domain or Enterprise administrative permissions, or equivalent delegated permissions, are required.^6^
If this is a publicly issued certificate instead, do not deploy an internal trust override. Check whether the clients can receive Microsoft certificate trust list updates; Windows normally updates trusted and untrusted CTLs automatically, and disabling that mechanism is not recommended.^7^
References