CBOM Documentation

CBOM Secure

  • Product Overview
    • Overview
    • Dashboard Interface
    • Dashboard Operations
    • Data Visualization
    • Overview
    • Host
    • Template
      • Networked Deployments
      • Cloud Deployments
      • Off-Grid Deployments
    • Discovery History
      • Tags
      • Tag Scope
      • Integrations
      • Cloud Servers
      • CycloneDX Export
    • Sensors Reference
    • Overview
    • Summary Report
    • PQC Report
    • AI Services
    • Source Code Visualizer
    • Key
    • Certificate
    • CipherSuites
    • Analysis
    • Policy Management
    • User Policy
    • Standard Policy
    • Overview
    • Cases
    • Criteria
    • Notification Format
    • Rules
      • Connections
      • Health
    • User Management
    • Roles & Permissions
    • Audit Trail
    • System Logs
    • Top Bar and User Menu
    • System Configuration
  • Login and Authentication
    • Introduction
      • Components
      • Data Flows
      • Execution Models
      • Incremental Scanning
      • Security Model
      • Topologies
      • Sizing Guide
      • HA and Scaling
      • Ports and Firewall
      • Prerequisites
      • Docker Compose
      • Native on Linux
      • Deployment Server
      • Install Troubleshooting
      • Validation Checklist
      • Managing Services
      • Logging
      • Certificates
      • Configuration
      • Backup and Recovery
      • Ops Troubleshooting
      • Methods
      • SAML 2.0 SSO
      • OIDC and Entra ID
      • Token Authentication
      • Password Policy
      • Sensor vs Connector
      • Sensor Guide Structure
      • Capability Matrix
      • GitHub Sensor
        • AWS
        • Azure
        • Google Cloud
        • Active Directory
        • LDAP Directories
        • AD CS
        • Thales Luna HSM
        • Entrust nShield HSM
        • HashiCorp Vault KMIP
        • Windows Cert Store
        • Linux Trust Store
        • MSSQL TDE
      • Roadmap
      • Integration Model
      • CrowdStrike
      • ServiceNow
      • ServiceNow and Jira Setup
      • Notification Engine
  • Administrative Guide
    • Glossary
    • Configuration Reference
    • Data Model Reference
    • API Reference
    • API Authentication
    • CycloneDX Export Specification
    • Risk and PQC Scoring
    • Built-in Standards
    • Widget Catalogue
    • AI Services Data Handling

4.1 Authentication Methods

This part is the configuration side of authentication: how an administrator federates the platform to a corporate identity provider, and how non-interactive callers authenticate. Login and Authentication, User Management and Roles & Permissions cover the same ground from the end-user side.

CBOM Secure delegates authentication to Keycloak. Three methods are available for interactive users, plus token-based access for non-interactive callers. Choose between them before configuring anything.

  • Local username and password: accounts held in the platform realm. Simplest to set up. Appropriate for evaluation, small deployments, and break-glass administrative accounts.

  • SAML 2.0 federation: the platform realm brokers to a corporate SAML identity provider. Users authenticate against corporate identity; group or role claims map onto platform roles. Appropriate for most enterprise deployments.

  • OIDC federation: the same pattern over OpenID Connect. Appropriate where the corporate provider is Microsoft Entra ID or another OIDC provider, and generally simpler to configure than SAML.

  • Token-based access: for non-interactive callers. API clients and sensors do not use any of the above; see 4.4.

Keep at least one local administrative account even after federating, so that a misconfigured or unavailable identity provider cannot lock every administrator out of the platform. Treat it as a deliberate break-glass account, with a strong password held in a credential store.

Choosing between them

Local accountsSAML 2.0OIDCToken
Who it is forPeoplePeoplePeopleMachines
Credential held byThe platform realmCorporate directoryCorporate directoryThe platform
Setup effortNoneModerateLowNone
Joiner and leaver handlingManual, per accountFollows the directoryFollows the directoryManual
Corporate password policy appliesNoYesYesn/a
Multi-factorConfigured in the realmInherited from the providerInherited from the providern/a
Single logoutn/aSupportedSupportedn/a
Suitable for productionBreak-glass onlyYesYesYes

If your corporate provider offers both, choose OIDC. It is materially simpler to configure, the failure messages are clearer, and there is nothing SAML does here that OIDC does not. Choose SAML when the corporate provider only speaks SAML, or when a security team requires it.

Local accounts are not an enterprise option. They are correct for an evaluation, for a very small deployment, and for the break-glass administrator every federated deployment should keep. Running an estate on them means maintaining a second user directory with its own joiners and leavers, and that directory will drift.

What federation does and does not change

ChangesDoes not change
Who authenticates the userWhat the user can do once in — roles and permissions stay in the platform
Where the password policy livesThe audit trail, which attributes actions the same way
How accounts are created — on first sign-in, from the group claimService accounts, which stay local
How leavers are handled — remove from the groupThe need for a break-glass administrator

Federation moves authentication. It does not move authorisation: the group claim maps onto a platform role, and that role's permissions are still set under Roles and Permissions.

Planning a federation

Decide these before touching any configuration. Each one is painful to change afterwards.

1. The external address. It goes into the redirect URI, which must match exactly. Changing it later means changing it at the identity provider too, and a mismatch fails authentication with a clear error — which is at least a good failure.

2. The group-to-role mapping. Which directory groups map onto which platform roles. Keep it narrow, and especially keep the group that maps to administrator small enough to name the members of.

3. What happens to existing local accounts. A person with both a local account and a federated identity has two platform records. Decide whether to migrate the roles onto the federated record and delete the local one, and do it deliberately rather than discovering the duplicate later.

4. Where the client secret lives, and when it expires. This is the one that bites. The secret is shown once at creation and has an expiry date. When it expires, sign-in stops for everyone at once, with no warning and no symptom beforehand. Put the expiry date in a calendar that someone reads, and the secret in a credential store.

5. The break-glass account. A local administrator, strong password, in a credential store, tested after every change to the federation. The scenario it exists for is the one where the directory is unavailable or misconfigured — which is exactly when you cannot ask the directory team for help.

Rolling out safely

  1. Configure the federation while local sign-in still works.

  2. Test with one account that holds a non-administrative role.

  3. Confirm the group claim arrives and maps to the expected role — a user who signs in successfully and sees nothing has authenticated but has no role.

  4. Test single logout if you intend to use it.

  5. Widen to a pilot group.

  6. Keep local sign-in available throughout.

  7. Once the pilot is clean, migrate the remaining users and retire their local accounts — except the break-glass one.

Token access for machines

Non-interactive callers do not use any of the above. An integration authenticates with its own service account and receives a short-lived session it renews; an agent authenticates with the host certificate issued when the host was registered. Neither path involves the corporate identity provider, which is deliberate: an automated job should not stop working because a directory is briefly unavailable.

Details are on Token Authentication and API Authentication.

When federation breaks

SymptomCause
Everyone's sign-in fails at once, with no change madeThe client secret expired
Redirect URI errorThe configured address and the registered URI differ — scheme, host or trailing path
Signs in, sees nothingThe group claim did not arrive, or maps to no role
Some users work, others do notThose users are not in a mapped group
Sign-in loops back to the sign-in pageThe public address does not match what the proxy forwards, or the forwarded-proto header is missing
Works internally, fails externallyTwo addresses, one registered

In every one of these, the break-glass local administrator is how you get in to fix it.

On this Page:
Support Privacy Policy Terms of Service

Copyright 2026 Encryption Consulting LLC. All rights reserved.