Engineering & Industry 4 min readORG-F16

OPC UA Certificates and Trust Lists: A Commissioning Troubleshooting Guide

Share

An OPC UA secure-channel failure is usually diagnosable when the team separates network reachability, endpoint and security-policy selection, application identity, certificate validity, URI and hostname checks, trust chains, revocation, time, permissions and user authentication. Automatically accepting an unknown certificate may hide the symptom while destroying the trust decision.

The short answer

Record the exact status code and logs at both client and server. Prove basic reachability and the intended endpoint, then inspect each application instance certificate: subject and application URI, DNS/IP names, issuer, validity period, key use, signature algorithm and thumbprint. Confirm clocks and time zones, place the correct peer or issuing CA in the correct trust store on both sides, install current CRLs where used, protect private keys and restart or reload only as the product requires. Test again with the intended security policy and user credentials—application trust and user authorization are separate controls.

  • Never solve commissioning by enabling permanent automatic trust or disabling security.
  • Treat every installed application instance as a distinct identity with its own certificate and private key.
  • Verify the exact certificate thumbprint before moving it from rejected to trusted storage.
  • Plan renewal, revocation, ownership and rollback before certificates expire in production.
Separate transport, application trust and user authorization so each failed control can be diagnosed safely.
Text-free OPC UA trust troubleshooting path from endpoint and time through application certificate, TrustList, revocation, secure channel and user authorization

Separate transport, application trust and user authorization so each failed control can be diagnosed safely.

Decision guide

CheckTypical symptomEvidence
Endpoint and policyNo matching endpoint, policy or mode; connection fails before sessionEndpoint URL, discovery output, supported policy and product logs
Certificate identityURI, host, purpose, date or signature validation failsDecoded certificate, time source and approved application identity
Trust and revocationPeer or chain is untrusted or revokedTrustList, issuer store, CRL, thumbprint and admin action
AuthorizationSecure channel succeeds but session or node access failsUser method, role mapping, permissions and audit event

A practical workflow

  1. 01

    Freeze the symptom: capture status code, UTC timestamps, endpoint, policy, client/server versions and both logs.

  2. 02

    Verify routing, port, DNS and discovery without weakening certificate validation.

  3. 03

    Inspect both application certificates, URI/hostname, validity, key protection, chain and time synchronization.

  4. 04

    Compare rejected, trusted, issuer and revocation stores on both peers; approve only the verified identity.

  5. 05

    Retest secure channel, session and least-privilege data access; document renewal, expiry monitoring and recovery.

Evidence and acceptance

Retain architecture, asset and application identity, endpoint, product versions, certificate metadata and thumbprints—not private keys—trust and issuer store state, CRL date, clock source, status codes, redacted logs, approval to trust, configuration change, test cases, least-privilege access result, renewal owner, expiry date and rollback record.

Common mistakes

  • Clicking “trust all” or disabling security to make commissioning pass.
  • Adding a certificate to only one side when mutual application authentication is required.
  • Ignoring application URI or SubjectAltName mismatch because the dates look valid.
  • Copying private keys between hosts or attaching them to a ticket.
  • Confusing a trusted application certificate with successful user authentication and authorization.

Responsibility boundary

Primary sources

Technical claims and product boundaries were checked against these primary sources. Fact-checked

  1. OPC Unified Architecture Part 2: Security Model — Application AuthenticationOPC Foundation
  2. OPC Unified Architecture Part 12: Certificate ManagementOPC Foundation

Frequently asked questions

Can I automatically accept every rejected certificate?

No. Verify the peer identity and thumbprint through an approved channel before adding it to trust; permanent auto-trust removes the decision the certificate is meant to support.

Why is a valid certificate still rejected?

Validity dates are only one check. Application URI, hostname, purpose, key use, chain, trust, revocation, algorithm and clock can also fail.

Does trusting the server authenticate the user?

No. Application authentication establishes peer trust; user credentials and role permissions are evaluated separately.

Can AutoNexit manage OPC UA private keys?

No. Keep private keys in protected certificate stores or dedicated key infrastructure; record only safe metadata and evidence.

Related articles

ORG-F15

OPC UA vs MQTT: Which Belongs Where in an Industrial Data Architecture?

OPC UA and MQTT solve different parts of an industrial information problem. OPC UA defines interoperable services, information models, security, and client-server or PubSub behavior; MQTT is a lightweight publish-subscribe transport through a broker. Many architectures use both, but only when semantics, topic governance, security, quality, and ownership are explicit.

ORG-F17

Modbus TCP vs PROFINET vs EtherNet/IP: A Protocol Selection Matrix

Modbus TCP, PROFINET, and EtherNet/IP all use Ethernet, but their device models, real-time behavior, engineering tools, diagnostics, profiles, conformance ecosystems, and security options differ. Selection should follow application requirements and verified device support—not a generic speed ranking.