Todos os guias de estudo

PFAA-001

Certified Expert - PingFederate

Parte de Certified Expert - PingFederate

Each voucher code is valid for a single exam attempt only; once redeemed it cannot be reused. Exam covers PingFederate version 11 or later.

Material de estudo independente, construído a partir do blueprint público da prova. Sem afiliação, autorização ou endosso da ping. Os objetivos são transcritos e mapeados à mão, então esta página pode conter erros ou estar defasada em relação ao material atual do fabricante: objetivos, versões, numeração e disponibilidade mudam sem aviso. Trate o portal de certificação da própria ping como a fonte da verdade e confira nele antes de agendar qualquer coisa.

Objective 1: Advanced Configuration

  • PFAA-001.1.01 Configure attribute mappings from an external data store with failover.

    O que saber:
    • Failover on a data store means a second source is queried when the first is unreachable.
    • Configure multiple LDAP hostnames on the data store so the connection itself fails over.
    • Failover addresses availability of the source, not missing data within a reachable source.
    • Test by taking the primary offline, not by reasoning about the configuration.
    • Connection timeouts must be short enough that failover happens faster than the user gives up.
  • PFAA-001.1.02 Configure attribute mappings from an external data store with failsafe mappings.

    O que saber:
    • Failsafe mapping supplies a default value when the lookup succeeds but returns nothing.
    • This is the distinction the exam draws: failover is about the source being down, failsafe is about the attribute being absent.
    • Without a failsafe, a missing attribute can fail the whole SSO transaction.
    • Choose defaults that are safe if they reach the service provider, since they will.
    • Log when the failsafe fires, or silent defaults hide a data-quality problem.
  • PFAA-001.1.03 Configure attribute mappings from an external data store using OGNL.

    O que saber:
    • OGNL expressions compute attribute values from data-store results and other context.
    • Expressions can concatenate, conditionally select, and transform values during mapping.
    • Enable expressions explicitly; they are disabled by default because they are powerful.
    • Keep expressions short and testable; complex logic belongs in the data source or an adapter.
    • An expression that throws will fail the transaction, so handle null cases deliberately.
  • PFAA-001.1.04 Create custom attributes using OGNL.

    O que saber:
    • Custom attributes are contract entries populated by expression rather than by direct lookup.
    • Extend the attribute contract first, then fulfil the new attribute with an OGNL expression.
    • Typical uses: deriving a role string, normalizing case, or combining given and family names.
    • The service provider only receives what the contract carries, so contract and expression must agree.
    • Test with representative data including the awkward cases, not just a clean sample user.
  • PFAA-001.1.05 Customize the HTML Form Adapter.

    O que saber:
    • The HTML Form Adapter renders the username and password login page.
    • Customization uses the template files in the server's template directory, with localization files for text.
    • Branding changes belong in templates and CSS; behavior changes belong in adapter configuration.
    • Template changes must be replicated to every node in a cluster.
    • Keep customizations minimal and documented, because they must be reapplied at upgrade.
  • PFAA-001.1.06 Self-service password reset.

    O que saber:
    • Self-service password reset lets users set a new password without an administrator.
    • Configuration is on the HTML Form Adapter and requires a password credential validator that supports writes.
    • Verification method matters: email one-time link, one-time password, or challenge questions each have different security properties.
    • Rate limiting and account-enumeration protection belong in the design; a reset flow that reveals whether an account exists is a finding.
    • The directory account used by PingFederate needs the rights to change passwords.
  • PFAA-001.1.07 Self-service account recovery.

    O que saber:
    • Account recovery unlocks or restores access after lockout, distinct from resetting a known-bad password.
    • It uses the same verification mechanisms as password reset and the same enumeration concerns.
    • Coordinate with the directory's lockout policy so recovery actually clears the lock.
    • Audit recovery events specifically; they are a common target for social engineering.
    • Decide whether recovery requires stronger verification than reset, and justify the choice.
  • PFAA-001.1.08 Self-service username recovery.

    O que saber:
    • Username recovery tells a user which username is associated with a verified identifier such as an email address.
    • This is the highest enumeration risk of the three self-service flows, so responses should be uniform regardless of whether the account exists.
    • Delivery should go to the registered channel, never displayed on screen.
    • Some organizations disable this entirely; that is a legitimate design decision.
    • Rate limit aggressively and log every attempt.
  • PFAA-001.1.09 Advanced use of authentication policies.

    O que saber:
    • Authentication policies are a decision tree over adapters and selectors, evaluated per request.
    • Selectors branch on context: network range, requested authentication context, client, or attribute values.
    • Policy contracts carry attributes out of the tree to the connection, decoupling the tree from the target.
    • Advanced use includes step-up authentication, where a branch requires a second factor before reaching success.
    • Draw the tree before building it; policies are hard to read in the console once they grow.
  • PFAA-001.1.10 Local identity profiles.

    O que saber:
    • Local identity profiles support registration and profile management for users PingFederate itself manages.
    • They define the fields collected, the data store where they are written, and the associated templates.
    • Registration, profile update, and email verification are the common flows.
    • This is for use cases without an existing enterprise directory of these users, typically customer identity.
    • Field validation and required-attribute handling belong in the profile definition.
  • PFAA-001.1.11 SP Target URL Mapping.

    O que saber:
    • SP target URL mapping decides where the user lands at the service provider after SSO.
    • It maps a requested target to an allowed destination, rather than trusting an arbitrary supplied URL.
    • This is an open-redirect control as much as a usability feature.
    • Wildcards must be scoped carefully so they cannot match an attacker-controlled host.
    • Coordinate with the service provider on the exact URLs their application uses.
  • PFAA-001.1.12 Configure Redirect Validation.

    O que saber:
    • Redirect validation restricts the URLs PingFederate will redirect a browser to after authentication or logout.
    • Configure allowed target resources for SSO and for SLO separately.
    • Without it, a crafted target parameter turns the identity provider into an open redirector, which is a phishing amplifier.
    • Validation should match on scheme and host, not merely on a substring.
    • Test with a deliberately hostile value to confirm the rejection path works.
  • PFAA-001.1.13 Using RelayState to request a specific resource.

    O que saber:
    • RelayState carries application state through the SAML exchange so the user returns to the resource they requested.
    • The identity provider echoes it back; it is opaque to the protocol.
    • Because it is attacker-influenceable, it must be validated against redirect-validation rules before use.
    • Size limits apply; large state should be stored server-side with a reference in RelayState.
    • Understand the difference between RelayState and the target parameter of an IdP-initiated URL.
  • PFAA-001.1.14 Configure certificate revocation checking.

    O que saber:
    • Revocation checking confirms a presented certificate has not been revoked, via CRL or OCSP.
    • OCSP gives fresher answers; CRLs are simpler but can be large and stale.
    • Decide the failure behavior when the revocation source is unreachable: fail open or fail closed, each with consequences.
    • Network access to the CRL distribution point or OCSP responder must exist from the PingFederate host.
    • Cache settings trade freshness against latency and load on the responder.
  • PFAA-001.1.15 Configure certificate rotation for self-signed certificates.

    O que saber:
    • Self-signed certificates in PingFederate can be rotated so a new certificate is published before the old one stops being used.
    • The rotation creates a successor certificate that appears in metadata, letting partners pick it up in advance.
    • Partners consuming metadata automatically will follow; partners with pinned certificates need manual coordination.
    • Plan rotation before expiry, not at it; a signing certificate that expires breaks every federation at once.
    • Verify each partner has accepted the new key before activating it.

Objective 2: SSO IdP/SP Configuration

  • PFAA-001.2.01 Create custom URLs using single sign-on (SSO) endpoint variables

    O que saber:
    • SSO endpoint variables let a single startSSO URL serve multiple targets by parameter.
    • Common parameters: PartnerSpId, TargetResource, and IdP adapter selection.
    • Custom URLs simplify bookmarking and application integration without a connection per target.
    • Every parameter that influences redirection must be covered by redirect validation.
    • Document the URL format for application teams, since malformed URLs are a frequent support issue.
  • PFAA-001.2.02 Configure identity provider to service provider (IdP-to-SP) bridging

    O que saber:
    • IdP-to-SP bridging makes PingFederate act as a service provider to an upstream identity provider and as an identity provider to a downstream service provider.
    • This is the federation hub pattern: many upstream identity sources, many downstream applications, one translation point.
    • Attribute mapping happens twice, and the contract must carry attributes through both hops.
    • Authentication policies decide which upstream identity provider handles a given request.
    • Bridging concentrates trust, so the hub's own security and availability become critical.
  • PFAA-001.2.03 Configure session cookies

    O que saber:
    • Session cookies represent the authenticated session at PingFederate itself.
    • Configure timeouts, whether the session is tracked, and the scope of the cookie.
    • Authentication sessions can be stored so re-authentication is not required across applications.
    • Cookie attributes (Secure, HttpOnly, SameSite) must be set appropriately for the deployment.
    • Session behavior interacts with single logout; understand what a logout actually terminates.

Objective 3: OAuth/OIDC Configuration

  • PFAA-001.3.01 Configure dynamic client registration

    O que saber:
    • Dynamic client registration lets OAuth clients register themselves at runtime rather than being created by an administrator.
    • Access to the registration endpoint must be controlled, typically by an initial access token, or anyone can create clients.
    • Client registration policies constrain what a self-registering client may request.
    • Registered clients receive credentials and a registration access token for later management.
    • This is essential at scale and dangerous when unrestricted; the policy configuration is the control.
  • PFAA-001.3.02 Configure an external DB for persistent grants

    O que saber:
    • Persistent grants record a user's consent and refresh-token state, and must survive restarts.
    • An external database gives durability and shared visibility across a cluster.
    • Schema is supplied by PingFederate; the database account needs the documented rights.
    • Grant lifetime and cleanup of expired grants are operational settings with storage consequences.
    • Without external storage, grants are limited in ways that break refresh-token behavior at scale.

Objective 4: Clustering

  • PFAA-001.4.01 Configure an adaptive cluster

    O que saber:
    • An adaptive cluster distributes runtime state dynamically among engine nodes without dedicated state servers.
    • Nodes discover each other and share state, which simplifies deployment.
    • It suits deployments where all engines are equivalent and network conditions between them are good.
    • Understand the rebalancing behavior when a node joins or leaves.
    • Compare against the directed model before choosing; the trade is simplicity versus explicit control.
  • PFAA-001.4.02 Configure dedicated state servers in a directed cluster

    O que saber:
    • In a directed cluster, specific nodes are designated as state servers holding runtime state for the others.
    • Engines are configured with the state servers they use, making the topology explicit.
    • This gives predictable placement, which matters across latency boundaries.
    • State server failure has a defined impact that must be understood and tested.
    • Sizing the state servers is a separate capacity exercise from sizing the engines.
  • PFAA-001.4.03 Configure sub-clusters in a directed cluster

    O que saber:
    • Sub-clusters group nodes, typically by data centre or region, so state is shared within a group rather than globally.
    • This limits cross-site chatter and keeps latency-sensitive state local.
    • Requests must be steered so a user's session stays within the sub-cluster that holds its state.
    • Cross-sub-cluster failover behavior is a design decision with a user-visible consequence: re-authentication.
    • Configuration must be consistent across every node or the grouping silently misbehaves.
  • PFAA-001.4.04 Configure cluster runtime state-management services

    O que saber:
    • Runtime state management covers sessions, grants, and other data that must be consistent across nodes.
    • Services are configured per state type, so different data can use different strategies.
    • Some state can be regenerated (re-authenticate) and some cannot (consent); treat them differently.
    • Monitor replication of state, not just node liveness.
    • Test node loss deliberately and observe what users experience.

Objective 5: Troubleshooting

  • PFAA-001.5.01 Interpret the contents of individual log files

    O que saber:
    • Key logs: server.log for runtime and errors, transaction and audit logs for per-request detail, admin logs for configuration changes.
    • The audit log records the SSO transaction with identity, connection, and result, which is the first place to look for a failed login.
    • Correlate across logs using the tracking or transaction identifier.
    • Timestamps must be synchronized across cluster nodes or correlation is impossible.
    • Know which log answers which question before an incident, not during one.
  • PFAA-001.5.02 Change logging levels

    O que saber:
    • Logging levels are set in log4j2.xml, per logger, and can be changed for a specific component rather than globally.
    • DEBUG on the right package is far more useful than DEBUG everywhere, which buries the signal and hurts performance.
    • Some changes require a restart while others are picked up on the configured monitor interval.
    • Raise the level, reproduce, capture, and lower it again as a disciplined loop.
    • Never leave verbose logging on in production; it can log sensitive data.
  • PFAA-001.5.03 Configure log file rotation using log4j2.xml

    O que saber:
    • Rotation is configured with log4j2 appenders using time-based or size-based triggering policies.
    • A rollover strategy sets how many archived files are kept.
    • Rotation prevents a single unbounded file and makes archival practical.
    • Align rotation with your log-shipping cadence so files are complete when collected.
    • Confirm disk capacity accounts for the retained set, not just the active file.
  • PFAA-001.5.04 Configure custom log listeners using log4j2.xml

    O que saber:
    • Custom appenders in log4j2.xml can direct specific loggers to their own files or to remote destinations.
    • Syslog and socket appenders push logs to central collection without a separate agent.
    • Filters let one appender receive only the events that matter, keeping targeted logs small.
    • Pattern layouts control the format, which downstream parsers depend on; changing it breaks them.
    • Validate the configuration on one node before rolling it across the cluster.
  • PFAA-001.5.05 Configure database logging

    O que saber:
    • Database logging writes audit records to a relational database instead of, or alongside, files.
    • It requires a JDBC data store and the schema PingFederate supplies.
    • Querying audit data becomes practical, which is why compliance teams ask for it.
    • Database availability now affects logging; understand the behavior when the database is unreachable.
    • Retention and archival of the audit table is an operational responsibility that files handled implicitly.

Objective 6: Basic Configuration

  • PFAA-001.6.01 Configure an HTML Form Adapter

    O que saber:
    • The HTML Form Adapter authenticates users with a username and password against a password credential validator.
    • Configuration covers the validator, session behavior, and which self-service features are exposed.
    • The adapter contract defines which attributes it makes available to policies and connections.
    • Challenge and lockout behavior come from the validator and the underlying directory, not the adapter alone.
    • This is the default interactive adapter and the base for most authentication policies.
  • PFAA-001.6.02 Customize the HTML Form Adapter

    O que saber:
    • Customization means editing the velocity templates and the localization property files.
    • Files live in the server template directory and must be identical on every cluster node.
    • Separate branding (CSS, images) from logic; logic in templates is hard to maintain across upgrades.
    • Keep a record of every customized file, because upgrades ship new defaults that must be merged.
    • Test the customized page for accessibility and for correct behavior on error, not just the happy path.
  • PFAA-001.6.03 Configure an HTTP Basic Adapter

    O que saber:
    • The HTTP Basic Adapter authenticates using the browser's basic authentication prompt.
    • It is simple and unbranded, which makes it suitable for service or test scenarios rather than user-facing login.
    • Credentials are sent on every request, so TLS is mandatory.
    • There is no logout mechanism for basic authentication, which is a real limitation.
    • It still uses a password credential validator, so the backing directory configuration is the same.
  • PFAA-001.6.04 Configure a Kerberos Adapter

    O que saber:
    • The Kerberos Adapter provides desktop single sign-on for domain-joined machines.
    • It requires a service account and keytab in Active Directory, plus a correct service principal name.
    • Browsers must be configured to trust the PingFederate hostname for integrated authentication.
    • Clock skew between the domain controller and PingFederate breaks Kerberos; time synchronization is mandatory.
    • Provide a fallback adapter for machines that cannot use Kerberos, or remote users are locked out.
  • PFAA-001.6.05 Configure an OpenToken Adapter

    O que saber:
    • The OpenToken Adapter exchanges identity with an application using a shared-secret encrypted token.
    • It is an agentless integration where the application uses an OpenToken agent library.
    • Shared secret, transport (cookie or query parameter), and token lifetime are the key settings.
    • Both sides must agree on the cipher suite and the attribute names carried in the token.
    • It is a legacy integration pattern; prefer standards-based options where the application supports them.
  • PFAA-001.6.06 Configure a One-Time Passcode Adapter

    O que saber:
    • The One-Time Passcode Adapter delivers a code out of band, typically by email or SMS.
    • It is used as a second factor in an authentication policy, after a primary adapter.
    • Configure the delivery channel, code length, expiry, and retry limits.
    • The user's contact attribute must be retrievable, so the data store mapping is part of the setup.
    • Delivery failures are the operational risk; monitor them and provide an alternative factor.
  • PFAA-001.6.07 Configure a RefID Adapter

    O que saber:
    • The RefID Adapter passes a reference identifier rather than the identity itself, which the partner resolves separately.
    • It suits integrations where attributes must not travel in the token.
    • The reference must be resolvable by the receiving side within its validity window.
    • Understand the lifetime and single-use characteristics of the reference.
    • This is a privacy-preserving pattern: the token carries a pointer, not the data.
  • PFAA-001.6.08 Configure an X509 Adapter

    O que saber:
    • The X509 Adapter authenticates using a client certificate presented during the TLS handshake.
    • The connector must request or require client certificates, and the issuing CA must be trusted.
    • Attributes are extracted from the certificate subject or SAN and mapped into the contract.
    • Revocation checking should be enabled, or a revoked certificate still authenticates.
    • This is strong authentication tied to a device or a smartcard rather than to knowledge.
  • PFAA-001.6.09 Configure a Composite Adapter

    O que saber:
    • The Composite Adapter chains several adapters so multiple factors are collected in one adapter instance.
    • Each chained adapter contributes attributes; the composite contract exposes the combination.
    • Ordering matters, and the policy of which adapters are required versus optional is part of the configuration.
    • Authentication policies can express similar logic; the composite is simpler for a fixed chain.
    • Test the failure of each member adapter to confirm the overall behavior is what you intended.
  • PFAA-001.6.10 Configure attribute mappings from an external data store

    O que saber:
    • Data store mapping enriches the identity with attributes fetched at authentication time.
    • Configure the data store, the search filter or query, and the mapping into contract attributes.
    • The search filter is the critical piece: it must uniquely identify the user, and it must be indexed at the directory.
    • Failed lookups and empty results are different failure modes; handle both.
    • Every mapped attribute costs a lookup, so retrieve what is needed and no more.
  • PFAA-001.6.11 Create custom attributes using OGNL

    O que saber:
    • OGNL expressions in fulfilment produce values that are not directly present in the source.
    • They must be explicitly enabled in configuration before they can be used.
    • Common patterns: conditional defaults, string manipulation, and joining values from multiple sources.
    • Expressions execute per transaction, so keep them cheap and null-safe.
    • Validate against real data, because a null in production will throw where the test data did not.
  • PFAA-001.6.12 Configure an SP/IdP SAML connection using the POST binding

    O que saber:
    • A SAML connection using POST binding sends the assertion in a form auto-submitted by the browser.
    • Configuration requires the partner entity ID, ACS URL, signing certificate, and the attribute contract.
    • POST binding suits large assertions; redirect binding is size-limited by URL length.
    • Assertion signing and optional encryption are configured per connection and must match partner expectations.
    • Metadata exchange is the reliable way to get these values right rather than transcribing them.
  • PFAA-001.6.13 Configure an SP/IdP connection using OpenID Connect (OIDC)

    O que saber:
    • An OIDC connection uses OAuth flows with an ID token instead of a SAML assertion.
    • The relying party is registered as an OAuth client with redirect URIs and allowed grant types.
    • OIDC policies determine what claims land in the ID token.
    • Scopes control which claims the client may request, and the policy fulfils them.
    • Discovery metadata lets clients configure themselves from the well-known endpoint.
  • PFAA-001.6.14 Configure an SP/IdP connection utilizing multiple ACS endpoints

    O que saber:
    • Multiple ACS endpoints let one connection serve several service-provider environments or indexes.
    • The assertion consumer service index or explicit URL in the request selects the endpoint.
    • A default endpoint applies when none is specified.
    • Each endpoint must be validated to prevent assertions being posted to an unintended location.
    • This is how a single connection supports development, test, and production URLs where the partner requires it.
  • PFAA-001.6.15 Create custom URLs using SSO endpoint variables

    O que saber:
    • Endpoint variables in a startSSO URL select the partner, the target resource, and the adapter to use.
    • This enables deep linking from a portal into a specific application through the identity provider.
    • All target values must satisfy redirect validation.
    • Encode parameter values correctly, since unencoded characters are a common cause of failures.
    • Publish the canonical URL format to application owners rather than letting each team invent one.
  • PFAA-001.6.16 Create OAuth clients

    O que saber:
    • An OAuth client is defined by client ID, authentication method, allowed grant types, redirect URIs, and scopes.
    • Client authentication options include shared secret, mutual TLS, and private key JWT; prefer the stronger options for confidential clients.
    • Public clients (SPAs, native apps) must use authorization code with PKCE and cannot hold a secret.
    • Redirect URIs must be exact-matched; wildcards here are an open-redirect and token-theft risk.
    • Scope restrictions on the client limit what it can ever request, independent of user consent.
  • PFAA-001.6.17 Configure persistent grant settings

    O que saber:
    • Persistent grants store the user's authorization so it need not be re-consented on every request.
    • Settings cover grant lifetime, idle timeout, and whether grants are reusable across clients.
    • Persistent grants underpin refresh tokens; without them refresh cannot survive a restart.
    • Grant storage may be local or in an external database for cluster durability.
    • Provide a revocation path, because a persistent grant is standing authorization until removed.
  • PFAA-001.6.18 Configure common scopes

    O que saber:
    • Common scopes are available to any client that requests them, subject to the client's own restrictions.
    • They form the shared vocabulary of the authorization server.
    • Each scope should map to a meaningful capability, not to an implementation detail.
    • Descriptions matter because they appear on the consent screen the user reads.
    • Keep the common set small; proliferation makes consent meaningless.
  • PFAA-001.6.19 Configure exclusive scopes

    O que saber:
    • Exclusive scopes are restricted so only explicitly permitted clients may request them.
    • Use them for privileged capabilities that must never be granted to a general client.
    • The client configuration must list the exclusive scope for the request to succeed.
    • This is the mechanism for separating administrative APIs from ordinary ones at the authorization server.
    • Audit which clients hold exclusive scopes as a periodic review.
  • PFAA-001.6.20 Configure internally referenced access tokens

    O que saber:
    • Internally referenced access tokens are opaque strings; the resource server calls introspection to learn their meaning.
    • Revocation is immediate because every validation consults the authorization server.
    • The cost is a network round trip per validation, mitigated by caching at the resource server.
    • Token contents never leave the authorization server, which is a privacy advantage.
    • Choose this when revocation immediacy outweighs latency.
  • PFAA-001.6.21 Configure JWT access tokens

    O que saber:
    • JWT access tokens are self-contained and validated locally against the signing key.
    • The resource server must verify signature, issuer, audience, and expiry, and reject anything failing.
    • Revocation is the weakness: a valid JWT remains valid until it expires, so lifetimes should be short.
    • Access token management instances define the claims, signing algorithm, and lifetime.
    • Publish keys via JWKS so resource servers can rotate without manual coordination.
  • PFAA-001.6.22 Configure attribute mapping for ROPC grant type

    O que saber:
    • The resource owner password credentials grant sends the user's password to the client, which is why it is deprecated.
    • Attribute mapping for ROPC has no browser interaction, so attributes come from the validator and data stores directly.
    • Where legacy systems still require it, restrict it to specific trusted clients.
    • The RFC 9700 security best practice is explicit that ROPC should not be used; document the migration plan.
    • Treat any ROPC client as a high-value credential-handling component.
  • PFAA-001.6.23 Configure attribute mapping for all other grant types

    O que saber:
    • Other grant types (authorization code, client credentials, device, token exchange) each have their own attribute-mapping configuration.
    • Client credentials has no user, so the mapping draws from the client rather than an authenticated person.
    • Authorization code maps from the authentication session and any data stores.
    • The mapping determines what the access token carries, which is what the resource server will authorize on.
    • Map only what the resource server needs; extra claims are unnecessary exposure and token bloat.
  • PFAA-001.6.24 Configure refresh tokens

    O que saber:
    • Refresh tokens let a client obtain a new access token without user interaction.
    • Configuration covers whether refresh tokens are issued, their lifetime, and whether they rotate on use.
    • Rotation with reuse detection is the recommended posture for public clients: a replayed token invalidates the chain.
    • Refresh tokens depend on persistent grants, so grant storage and lifetime govern them.
    • A long-lived refresh token is a long-lived credential and should be protected accordingly.
  • PFAA-001.6.25 Configure OIDC policies

    O que saber:
    • OIDC policies define which attributes become claims in the ID token and at the userinfo endpoint.
    • A policy is bound to clients and can differ per client, so different applications receive different claims.
    • Attribute sources and OGNL fulfilment work the same way as in SAML connections.
    • Include only necessary claims; the ID token is visible to the client and often logged by it.
    • The policy is where the identity contract with the application is actually defined.
  • PFAA-001.6.26 Configure OIDC attribute scopes

    O que saber:
    • Attribute scopes bind claims to requested scopes, so a client receives extra claims only when it asks and is permitted.
    • This implements data minimization: the profile scope yields profile claims, and nothing more by default.
    • The mapping between scope and claim set is configured in the OIDC policy.
    • Consent screens should reflect what each scope actually reveals.
    • Review the scope-to-claim mapping when adding claims, or data leaks through an existing scope.
  • PFAA-001.6.27 Configure an external LDAP database

    O que saber:
    • An external LDAP data store is configured with connection URLs, bind credentials, and connection pooling settings.
    • Use LDAPS or StartTLS; the bind credential travels on every connection.
    • The service account should have only the read rights it needs, plus write rights only if self-service requires them.
    • Connection pool sizing and timeouts materially affect authentication latency under load.
    • Multiple hostnames give failover; test it rather than assuming.
  • PFAA-001.6.28 Manage certificates

    O que saber:
    • Certificate management covers signing certificates, SSL server certificates, and trusted partner certificates.
    • Signing certificates appear in metadata and must be rotated with partner coordination.
    • SSL server certificates protect the runtime and admin ports and are separate from signing keys.
    • Import partner certificates into the correct trust context; a certificate in the wrong store does nothing.
    • Track expiry for every certificate in the deployment; expiry is the most common federation outage.
  • PFAA-001.6.29 Manage certificate authorities (CAs)

    O que saber:
    • Certificate authorities in the trust store determine which partner and backend certificates are accepted.
    • Importing a CA trusts everything it issues, so the trust store should be curated rather than filled with defaults.
    • Separate trust anchors for different purposes where the product allows it.
    • Intermediate certificates must be present for chain validation to succeed.
    • Review the trust store periodically; an unnecessary CA is unnecessary risk.

Fontes públicas, usadas de boa-fé

Estes guias são materiais de estudo independentes, montados a partir de conteúdo publicamente disponível: blueprints de exame publicados, documentação oficial de produto e catálogos de treinamento dos fabricantes. Nomes de produtos, códigos de exame e marcas pertencem a seus titulares e são usados apenas para identificar o assunto ensinado. Este site não é afiliado a nenhum fabricante aqui citado nem endossado por ele. Se você detém direitos sobre material publicado nesta página e entende que ele deve ser removido ou corrigido, envie a URL exata e uma breve nota sobre o problema pela página de contato; os pedidos são analisados com rapidez e boa-fé. Ler o aviso legal completo