Correct: C — the recovery URL. Cisco provides a recovery URL that bypasses SAML single sign-on so an administrator can log in to Unified Communications Manager Administration and the IM and Presence interfaces for troubleshooting. Only application users with administrative privileges can access it, and it is also used when changing a server domain or hostname.

Why not the others: Rebuilding the cluster (A) is a catastrophic response to a recoverable condition. Stopping Tomcat (B) removes the web interfaces entirely rather than restoring access. SAML assertion expiry (D) does not re-enable local authentication; SSO remains enabled until an administrator changes it.

Correct: A — restart Cisco Tomcat on all affected nodes. Cisco's deployment guide states that before and after enabling or disabling SAML single sign-on, the Cisco Tomcat service must be restarted on all Unified CM and IM and Presence Service cluster nodes where single sign-on is running.

Why not the others: User accounts are not recreated for an SSO change (B) — they are mapped, not rebuilt. Reinstalling the identity provider (C) is unnecessary and unrelated. Changing cluster security mode (D) governs encrypted signalling and media, not SSO enablement.

Correct: B — intercluster peers. Intercluster peer relationships interconnect standalone IM and Presence Service clusters, allowing users in one cluster to communicate with and subscribe to the availability of users in a remote cluster within the same domain.

Why not the others: A PSTN SIP trunk (A) carries external calls and has no bearing on presence subscription. A shared voicemail pilot (C) relates to message retrieval, not availability. Matching device pool names (D) is a local configuration convention that creates no intercluster relationship.

Correct: B — high availability is at the system level, with some components still single points of failure. Cisco notes this restriction explicitly for the WAN-with-remote-failover model, naming the Cisco Sync Agent, Cisco Intercluster Sync Agent, and the IM and Presence administration interface as components that may still have a single point of failure.

Why not the others: Remote clients are supported (A) — the Jabber client can be local or remote to the node. Messaging is not disabled during failover (C); clients fail over to the remote peer node. Clients reconnect automatically when the failed node comes back online (D), rather than requiring manual repointing.

Correct: D — the multiple-SCCP-port model is replaced. Unified Communications Manager interacts with Unity Connection over either SCCP or a SIP trunk. With traditional SCCP integration you create multiple voicemail ports and place them in a line group and route or hunt list; SIP trunk integration avoids that per-port arrangement.

Why not the others: Message waiting indicators still function (A) — users receive message-waiting lights regardless of integration type. Users must still be provisioned (B), whether manually, through AXL, or via LDAP. Calls still traverse Unified Communications Manager (C); the trunk is between the two systems.

Correct: B — identified user messaging. Unity Connection automatically identifies a user who leaves a message during a forwarded internal call, based on the extension the call originated from. The feature is enabled by default and can be disabled system-wide in the advanced conversation settings.

Why not the others: Text-to-speech (A) reads email messages aloud and is unrelated to caller identification. Personal call transfer rules (C) govern how a user's incoming calls are routed. Survivable remote site voicemail (D) provides branch voicemail during WAN outages.

Correct: C — they can be modified but not deleted. Unity Connection creates default users at installation time. They can be modified but cannot be deleted, and they handle message delivery and retrieval when callers are routed to default call management objects.

Why not the others: They cannot be deleted at all (A), let alone as a prerequisite for production. They exist regardless of LDAP integration (B). They do have mailboxes (D) — the Operator user's mailbox stores messages left for the Operator call handler, which is why Cisco advises assigning someone to monitor it or reconfiguring the handler.

Correct: D — one-key dialing. One-key dialing lets you designate a single digit to represent a user extension, an alternate contact number, a call handler, an interview handler, or a directory handler, so callers reach a destination with one keypress during a greeting.

Why not the others: Message waiting indication (A) signals a waiting message on a user's phone. Identified user messaging (B) recognises who left a message. An interview handler (C) collects recorded answers to a series of questions and forwards them as one message, which is a different function entirely.

Correct: A — the client can be local or remote to the IM and Presence node. Cisco states that in this deployment model the IM and Presence node need not be co-located with the Unified Communications Manager database publisher, and the Jabber client can be either local or remote to the IM and Presence Service node.

Why not the others: There is no same-subnet requirement (B). Presence continues to function (C) — that is the point of the model. Identity providers are configured per cluster in a SAML SSO deployment, not per client (D).

Correct: A — a termination warning prompt plays. By default Unity Connection plays a termination warning prompt before the maximum allowable message length is reached, sounding 15 seconds before the end of the recording provided the recording is not restricted to under 30 seconds.

Why not the others: Silent truncation (B) is precisely what the warning exists to prevent. Transferring to the operator mid-message (C) is not default behaviour. Discarding and restarting the recording (D) would lose the caller's message and is not how the system behaves.