The NetApp Storage Installation Engineer - ONTAP Professional Exam (NS0-185)

Broadcom BES-53248 cluster interconnect switches require a Reference Configuration File (RCF) to define supported port roles, speeds, breakout behavior, ISLs, and node-facing ports. By default, only a subset of ports is enabled for cluster connectivity. Ports 0/17 through 0/20 are not active for node connections unless explicitly configured by the correct RCF.

When new controllers are cabled to ports that are not enabled or correctly configured for cluster interconnect traffic, the ONTAP nodes cannot establish cluster LIF communication and therefore cannot join the cluster. Installing the correct RCF configures these ports with the required VLANs, MTU, flow control, and speed settings so that cluster traffic can pass.

A license file is not required for basic cluster interconnect functionality. The Cluster Switch Health Monitor (CSHM) configuration file enables monitoring, not port functionality. Rebooting controllers does not resolve switch-side configuration issues.

Therefore, installing the reference configuration file is the required corrective action.

Default failover groups are defined by broadcast domains . Therefore, the LIF fails over to any operational port in the broadcast domain.

In ONTAP, LIF failover target selection is controlled by failover groups and failover policies . A key installation and validation concept is that ONTAP uses broadcast domains to simplify and automate failover-group membership. Specifically, ONTAP documents that when you create a broadcast domain, ONTAP automatically creates a failover group with the same name and manages it so that as ports are added or removed from the broadcast domain, they are also added or removed from the associated failover group.

Because of this coupling, if a LIF “moves to a port that is not expected,” the most direct and common cause is that the broadcast domain contains ports that do not belong there (or ports that should be in a different broadcast domain/IPspace). When the original port goes down (for example, because you disabled the connected switch), ONTAP looks for eligible failover targets in the LIF’s assigned failover group. If the broadcast domain was defined too broadly or incorrectly, the failover group will also be too broad, and ONTAP may pick a port that appears valid per configuration but is “unexpected” from the design intent.

Option C (“There are no online ports within the same failover group”) would more commonly result in no valid failover target (and therefore an outage) rather than a clean migration to an unexpected port; ONTAP even warns that without a valid target, an outage occurs during failover verification. Option A is too generic and does not specifically explain why a different port becomes a valid target. Option D is not a typical ONTAP networking root cause for LIF migration behavior.

Therefore, the behavior is best explained by an incorrectly configured broadcast domain.

The AutoSupport history output shows multiple transport attempts using SMTP, HTTP, and HTTPS , with SMTP being re-queued, HTTP ignored, and HTTPS failed. This indicates that AutoSupport is enabled and functioning at the configuration level, but the delivery mechanism is being blocked or restricted by the customer’s network infrastructure.

SMTP failures commonly occur when mail servers block large attachments or require explicit whitelisting of sending systems. Since full AutoSupport messages include large log bundles, mail servers often reject or delay these messages unless the sending nodes are explicitly allowed. Therefore, whitelisting the nodes on the mail server directly addresses the SMTP delivery issue.

Additionally, NetApp installation best practices recommend HTTPS as the preferred AutoSupport transport protocol because it bypasses mail server limitations and is more reliable for large payloads. Switching the transport protocol to HTTPS avoids SMTP attachment-size restrictions entirely and ensures successful delivery when outbound HTTPS access is allowed.

Enabling AutoSupport on both nodes is not relevant because the output confirms AutoSupport is already active. Excluding nodes from attachment-size limits is not a supported ONTAP action.

Thus, the two correct actions are to change the transport protocol to HTTPS and whitelist the nodes on the mail server .

After completing the initial configuration of a multi-node ONTAP cluster and its cluster interconnect switches, NetApp installation best practices require enabling proactive monitoring of the cluster network. The Cluster Switch Health Monitor (CSHM) is the ONTAP feature designed specifically for this purpose.

CSHM continuously monitors the health of supported cluster switches by collecting information such as switch configuration, port status, firmware versions, cabling, and error conditions. It validates this information against NetApp-supported configurations and best practices. When deviations or risks are detected, CSHM generates alerts and EMS messages that allow administrators to take corrective action before cluster communication is affected.

Cluster switch logging alone does not provide health analysis or proactive validation; it only records switch events. IPv6 is not required or recommended for cluster interconnect networks in standard ONTAP installations. Data Center Bridging (DCB) is used for specific Ethernet storage protocols and is not applicable to ONTAP cluster interconnect networks.

Therefore, enabling the cluster switch health monitor is the correct and required post-installation step.

This is a post-install validation and troubleshooting scenario for SAS shelf cabling. ONTAP hardware installation guidance distinguishes between multipath/tri-path and quad-path configurations based on how shelves are cabled. For shelves with IOM12/IOM12B modules, quad-path configurations require double-wide shelf-to-shelf connectivity: first using the standard connections (IOM ports 3 and 1) and then adding double-wide connections (IOM ports 4 and 2). If quad-path HA is intended but Config Advisor reports mixed-path HA , then ONTAP’s view of the end-to-end SAS topology does not match the expected double-wide design.

The question states LEDs look normal and rebooting the SAS expander did not change the status. That reduces the likelihood of a transient expander condition and shifts focus to physical seating/connection integrity . In installation practice, a common next corrective step is to re-seat the SAS I/O modules (IOMs on the shelf side or SAS modules/adapters on the controller side depending on platform architecture). Reseating is a controlled, practical action to address partial lane connectivity, marginal contacts, or module insertion issues that can still present as “connected” at an LED level but result in ONTAP detecting fewer active paths than expected.

Option D is incorrect because multipath HA is not “the only” supported method on such platforms; ONTAP documentation explicitly describes quad-path (“double-wide”) cabling rules and steps. Option B (disconnect and rerun Config Advisor) is destructive and does not correct the underlying condition; it is more of a diagnostic reset than a validated next fix step, and it risks creating additional variables. Option A (open a case and power off shelves) is premature given there is a clear, standard physical remediation action remaining.

Therefore, the correct next step is to re-seat the SAS IO modules of the affected controller.

This question tests installation cabling constraints and supported cabling media for SAS connectivity. The key issue is distance: the customer wants to use 20-meter copper SAS cables between a controller and a shelf that are installed in separate rows.

NetApp hardware cabling guidance indicates that when long distance SAS connectivity is required, the supported approach is to use SAS optical cabling (for applicable platforms and shelf/IOM combinations). The ONTAP hardware systems documentation describes “mini-SAS HD SAS optical cable rules,” noting that optical AOC SAS cables are available in lengths up to 50 meters for controller-to-stack and shelf-to-shelf connections in supported configurations. This highlights a critical installation principle: long SAS runs are addressed with optical , not copper.

The documentation also establishes that SAS cabling within a stack must be consistent (all copper or all optical) and that mixed approaches have specific constraints. While the excerpt emphasizes optical rules (because optical is the mechanism for long distance), it implicitly reinforces that copper has practical distance limitations and is not the long-distance solution.

Therefore, if the customer insists on copper SAS, the correct guidance is that the controller and disk shelf must be closer together to remain within supported copper SAS distance limits and to ensure signal integrity and supported operation. Option B is incorrect because simply “obtaining” 20-meter copper SAS cables does not make the configuration supported. Option D is incorrect because Ethernet adapters do not replace SAS for shelf connectivity. Option A is nonsensical in this context because SAS shelf connectivity cannot be replaced by a “wireless adapter” as a supported installation practice.

Thus, the correct answer is C : if copper cables are to be used, the devices must be closer together; otherwise the supported solution is to use a supported SAS optical approach.

Flow control configuration is an important validation step in ONTAP SAN and NAS environments, especially in switchless cluster designs. NetApp documentation clearly differentiates flow control requirements between data networks and cluster networks.

For data network ports, full flow control must be enabled when hosts are configured with full flow control. This ensures proper congestion management and prevents packet loss during high-throughput NAS operations such as NFSv4.1 and SMB traffic. Matching flow control settings between hosts and storage data ports is a best practice explicitly stated in NetApp installation guidance.

In contrast, cluster network ports must have flow control disabled. The cluster network is highly sensitive to latency and congestion, and enabling flow control on cluster interconnect ports can introduce delays that negatively affect cluster quorum, node communication, and overall stability. NetApp explicitly requires flow control to be disabled on cluster network interfaces.

Enabling flow control on cluster ports or disabling it on data ports would violate NetApp-supported configurations and can lead to performance degradation or cluster communication issues.

Therefore, the two correct settings to verify are:

Data network ports have full flow control enabled

Cluster network ports have full flow control disabled

This makes C and D the correct answers.