The /etc/multipath.conf file is the standard configuration file for the Device Mapper Multipath (DM-Multipath) service on modern Linux distributions. This service is responsible for managing multiple I/O paths to storage devices, including those connected via Fibre Channel (FC). Administrators edit this file to define multipath device settings, path grouping policies, and device-specific configurations, such as those recommended by NetApp for optimal performance and failover with ONTAP LUNs. The multipathd daemon reads this file upon startup to apply the specified configuration.

A. /etc/kernel/mp.conf: This is not a standard path or filename for multipath configuration. Kernel-level parameters are typically managed through sysctl or module files in /etc/modprobe.d.

B. /opt/multipath/multipath.cnf: The /opt directory is generally used for optional, third-party software packages, not for core system services like DM-Multipath.

D. /etc/dm-multipath.conf: Although the technology is called Device Mapper Multipath (DM-Multipath), the conventional name for its configuration file is multipath.conf, not dm-multipath.conf.

1. NetApp

Inc. (2023). Linux Unified Host Utilities 7.3 Installation and Setup Guide. NetApp Documentation. In the section "Setting up the multipath.conf file

" the guide explicitly states

"You must have a properly configured /etc/multipath.conf file on your host." (See Chapter 3: SAN Administration).

2. Red Hat

Inc. (2023). Configuring device mapper multipath on RHEL 8. Red Hat Customer Portal. Chapter 3

"The multipath configuration file

" states

"The multipath configuration file is /etc/multipath.conf."

3. NetApp

Inc. (2022). TR-4080: Linux Host Configuration for ONTAP. NetApp Technical Report. Section 4.1

"Device Mapper Multipath

" details the configuration process which centers on modifying the /etc/multipath.conf file with NetApp-specific stanzas.

Changing the Maximum Transmission Unit (MTU) from the default of 1500 bytes to a larger value, such as 9000 bytes (known as Jumbo Frames), significantly increases iSCSI throughput. A larger MTU allows more data to be sent in a single network packet. This reduces the total number of packets and headers that need to be processed for a given amount of data, which lowers the CPU overhead on both the host and the E-Series storage system. The result is more efficient data transfer and higher overall throughput. For this change to be effective, the MTU must be increased consistently across the entire data path, including the host, network switches, and the storage controller ports.

B. flow control: Flow control is a mechanism to prevent packet loss due to buffer overruns, not a setting to increase throughput. Improper configuration can actually hinder performance.

C. Challenge-Handshake Authentication Protocol (CHAP): CHAP is a security feature for authenticating iSCSI sessions. It adds negligible overhead during login and does not impact data transfer throughput.

D. Time to Live (TTL): TTL is an IP network parameter that prevents packets from looping indefinitely. It has no direct effect on the throughput of an established iSCSI connection.

1. NetApp Technical Report TR-4394

"NetApp E-Series with VMware vSphere 6.5 Best Practices

" Page 18

Section: iSCSI Network Configuration.

The document states

"For optimal performance

NetApp recommends enabling jumbo frames on all components in the iSCSI data path... Using jumbo frames reduces the processing overhead on the hosts and on the storage controllers

resulting in improved throughput."

2. NetApp E-Series SANtricam System Manager

"Configure iSCSI ports

" Online Documentation.

The documentation for configuring iSCSI ports explains the MTU setting: "The maximum transmission unit (MTU) is the size (in bytes) of the largest protocol data unit that a communication protocol can pass across the network. The default is 1500. You can change the value up to 9000 for jumbo frames

which can improve performance."

3. NetApp Technical Report TR-4569

"NetApp E-Series and EF-Series Best Practices for Microsoft SQL Server

" Page 20

Section: iSCSI Configuration.

This guide recommends

"Use jumbo frames (MTU 9000) for all iSCSI traffic. Jumbo frames improve performance by reducing the number of frames that must be sent and processed."

The connector labeled '1' in the exhibit is a Quad Small Form-factor Pluggable (QSFP) connector. This type of connector is used for high-speed interconnects, including 4X InfiniBand (such as QDR and FDR) and 40GbE. NetApp systems utilize these cables and connectors for the private cluster interconnect network, which requires high bandwidth and low latency. The "4X" designation in InfiniBand signifies that the link uses four parallel data lanes, a configuration supported by the QSFP physical interface.

A. 3: This is an RJ-45 connector, which is used for standard Ethernet networking (e.g., 1000BASE-T, 10GBASE-T), not InfiniBand.

C. 2: This is a Small Form-factor Pluggable Plus (SFP+) connector, typically used for 10Gb Ethernet or 8/16Gb Fibre Channel, not 4X InfiniBand.

D. 4: This is an external Mini-SAS connector (SFF-8088 or SFF-8644), used for connecting controllers to disk shelves via the SAS protocol.

1. NetApp FAS8000 Series Installation and Setup Instructions: In the section "Step 3 Cable the controllers

" the guide provides diagrams of the controller's rear panel. It explicitly identifies the InfiniBand cluster interconnect ports as using QSFP connectors

which visually match connector '1' in the exhibit. (See "Rear view of the FAS8020/FAS8040/FAS8060 controller module

" page 28).

2. NetApp CN1610 Cluster-Interconnect Switch Installation and Configuration Guide: This document shows the different ports available on a NetApp cluster switch. It clearly illustrates SFP+ ports (matching connector '2') and RJ-45 management ports (matching connector '3')

distinguishing them from the QSFP ports used for the main interconnect. (See "CN1610 switch front panel

" page 10).

3. NetApp Universal SAS and ACP Cabling Guide (TR-4089): This technical report details the cabling for disk shelves. It includes images and descriptions of Mini-SAS connectors (SFF-8088) and Mini-SAS HD connectors (SFF-8644)

which match the appearance of connector '4'. (See "SAS Connectors

" page 6).

The ONTAP pre-upgrade health check (cluster image validate) is designed to prevent upgrades on clusters with underlying issues that could cause disruption or data loss.

A hot spare in use indicates that a disk has failed and an aggregate is in a degraded state while it reconstructs. Upgrading a system with a degraded aggregate is a high-risk operation and is therefore blocked by the pre-check.

A volume (LUN) on a non-preferred path indicates a SAN multipathing or configuration issue. During a non-disruptive upgrade, controllers perform takeovers and givebacks. If paths are not configured correctly, hosts can lose connectivity to their LUNs during this process. This is flagged as an error to ensure host connectivity is maintained.

C. An excess number of event log events is not a direct failure condition. The specific content of critical events, not the quantity, would cause a failure.

D. ONTAP upgrades are designed to be non-disruptive (NDU). The system is expected to be processing I/O during the entire upgrade process.

1. NetApp ONTAP 9 Documentation

"ONTAP upgrade and revert/downgrade guide": In the section "Preparing for the upgrade

" it is stated that you must "verify that the cluster and all nodes are healthy." A system with a spare in use (degraded aggregate) or multipathing errors (non-preferred path) is not considered healthy. The cluster image validate command enforces these health requirements.

2. NetApp ONTAP 9 Commands Manual Page Reference

cluster image validate: The documentation for this command lists the specific validation checks performed. These checks include spare-low (which would be triggered by a spare in use) and lun-maps-not-on-preferred-path. Both are listed as checks that result in an "Error" severity

which fails the pre-upgrade validation.

3. NetApp Knowledge Base

Article ID 1030339

"Top 10 issues that cause ONTAP upgrades to fail": This article explicitly lists "Degraded Aggregates" and "SAN/LIF/Multipathing issues" as common and critical failures identified by pre-upgrade health checks. A spare in use leads to a degraded aggregate

and a non-preferred path is a primary multipathing issue.

The SANtricity System Manager initial setup wizard guides an administrator through the essential first-time configuration steps for an E-Series storage array. One of the dedicated steps in this wizard is "Configure Alerts," which specifically prompts for the configuration of email notifications. This involves entering SMTP server details and recipient email addresses. While other alerting methods like SNMP and Syslog are available, they are configured separately through the main System Manager interface (Settings > Alerts) after the initial setup wizard has been completed.

A. Syslog: Syslog alerting is a valid feature but is configured under the main settings menu after the initial setup wizard is finished, not within the wizard itself.

B. SNMP: SNMP trap configuration is performed in the Settings > Alerts section of the System Manager GUI, subsequent to completing the initial setup wizard.

C. Web Services Proxy: The Web Services Proxy is an API endpoint for programmatic management and monitoring; it is not a direct alerting method configured through the alerts setup wizard.

1. NetApp E-Series SANtricity System Manager 11.70 Online Help

Section: "Tasks > Complete the setup wizard".

The documentation explicitly states: "The Configure Alerts page lets you set up email alerts so you can be notified of storage array problems... You can configure other types of alerts

such as SNMP traps and syslog

after you complete the setup wizard." This confirms that only email is part of the wizard.

2. NetApp E-Series SANtricity System Manager 11.70 Online Help

Section: "Settings > Configure alerts".

This section details the configuration for all alert types (Email

SNMP

Syslog)

showing that SNMP and Syslog are configured from this general settings page

which is accessed after the initial setup is done.

In NetApp ONTAP, an asynchronous SnapMirror relationship is established between a source and a destination volume, forming a "mirror pair." Each relationship is managed independently and is associated with a specific SnapMirror policy and a schedule. The schedule dictates the frequency of updates, which in turn defines the Recovery Point Objective (RPO). To support multiple relationships with different RPOs, an administrator must create a separate mirror pair for each data set and assign a unique schedule to each pair corresponding to its required RPO. For example, one pair could have a schedule to update every hour (1-hour RPO), while another has a schedule to update daily (24-hour RPO).

A. "Mirror group" is not a standard ONTAP term for managing SnapMirror relationships with distinct RPOs. Schedules are applied to individual relationships or consistency groups.

C. A single consistency group is governed by one schedule and one policy. Therefore, all volume relationships within that group share the same RPO.

D. While multiple consistency groups can be used to achieve different RPOs for different groups of volumes, the fundamental and most direct method is creating individual mirror pairs, which is always required.

1. NetApp ONTAP 9 Data Protection Administration Guide

Document Version 9.11

Page 31

"Overview of the SnapMirror relationship configuration workflow". The guide states

"You create a SnapMirror relationship to protect a volume... A SnapMirror policy and a schedule are assigned to each relationship." This confirms that schedules (which determine RPO) are assigned on a per-relationship basis.

2. NetApp Technical Report TR-4015: SnapMirror Configuration and Best Practices Guide for ONTAP 9

Page 11

Section 2.2 "SnapMirror Policies". The document explains

"A SnapMirror policy is a set of rules that defines the behavior of a SnapMirror relationship... The schedule defines the RPO for the relationship." This directly links the schedule of an individual relationship to its RPO.

3. NetApp ONTAP 9 Command Reference

snapmirror create command documentation. The syntax snapmirror create -source-path... -destination-path... -schedule demonstrates that a schedule is applied when creating a specific source-destination pair

allowing for unique RPOs for each created relationship.

Based on a strict calculation using the provided figures, none of the answers are correct. The available weight capacity is the floor limit (700 lb) minus the cabinet weight (90 lb), which equals 610 lb. With each fully populated E2824 shelf weighing 120 lb, the maximum number of shelves is 610 / 120 = 5.08. As only whole shelves can be installed, the true maximum is 5.

However, since 5 is not an option, the question is flawed. The intended answer is likely 6, which would be correct if the floor weight limit was 810 lb instead of 700 lb, as (810 lb - 90 lb) / 120 lb = 6. This suggests a typographical error in the question's premise.

A. 22: The total weight of (22 120 lb) + 90 lb = 2730 lb would grossly exceed the specified 700 lb floor limit.

B. 10: The total weight of (10 120 lb) + 90 lb = 1290 lb would significantly exceed the specified 700 lb floor limit.

D. 14: This result is obtained by incorrectly using the empty shelf weight (43 lb). Capacity planning must always use the fully populated weight (120 lb) for safety and operational standards.

1. NetApp E2800 and E5700 Hardware Installation and Maintenance Guide: The physical specifications for the E2824 shelf are listed

confirming the weight for a fully populated shelf is 120 lb (54.4 kg). This document validates the weight value used in the calculation. (Refer to Chapter 1: "System Overview"

Section: "Physical specifications").

2. NetApp E-Series Site Requirements and Cabling Guide: This guide outlines the environmental and physical planning requirements for E-Series systems. It emphasizes that the customer is responsible for ensuring the floor can support the total weight of the cabinet and all installed components

confirming the correct calculation methodology. (Refer to Chapter 2: "Site requirements"

Section: "Weight and floor loading specifications").

Throughput is the performance metric that measures the rate of data transfer over a specific period. It is typically expressed in units like megabytes per second (MB/s) or gigabytes per second (GB/s). The scenario describes measuring the "amount of data" over a "span of an hour," which is the exact definition of throughput. This metric is crucial for diagnosing performance issues related to bandwidth-intensive workloads where the volume of data transferred is the primary concern.

A. I/O alignment: This refers to the proper positioning of I/O requests with the underlying storage blocks. Misalignment can cause performance degradation but is not a measurement of data volume over time.

B. parity: This is a data protection technique used in RAID configurations to provide redundancy. It is a feature of the storage layout, not a performance metric measuring data transfer.

D. IOPS: IOPS (Input/Output Operations Per Second) measures the number of read and write operations per second, not the amount of data transferred. A high IOPS can have low throughput if the I/O sizes are small.

1. NetApp Technical Report TR-4393

"E-Series and EF-Series Best Practices for Performance

" Page 6

Section 2.1 Performance Fundamentals:

"Throughput is the amount of data that can be transferred in a given amount of time. It is typically measured in megabytes per second (MBps)." This directly supports the correct answer by defining throughput as the amount of data over time.

"IOPS is the number of I/O operations that can be performed in a given amount of time. It is typically measured in I/Os per second." This clarifies why option D is incorrect.

2. NetApp SANtricity 11.70 Documentation

"Monitor performance

" Online Help Center:

Under the "Performance graphs" section

the key metrics are defined: "Throughput – The rate at which data is transferred to or from the storage array

measured in megabytes per second (MB/s)." This definition aligns perfectly with the scenario in the question.

The NetApp E2812 is an E-Series storage system. When ordered as a "Duplex" system, it includes two controllers for high availability. Each E2800 series controller module is equipped with two onboard host interface ports. The question specifies the system is a "10 GbE Base-T" model, which defines the type of these onboard ports. Therefore, with two controllers, each having two 10GbE Base-T ports, the total maximum number of available ports is four (2 controllers × 2 ports/controller). The exhibit visually confirms the layout of a duplex controller shelf with two host ports per controller.

A. 8: This would incorrectly assume four onboard 10GbE Base-T ports per controller, which is not a standard E2800 configuration.

C. 6: This number does not align with the symmetric design of E2800 duplex controllers, which have two ports each.

D. 12: This number is incorrect and might be confused with the number of drive slots in the E2812 chassis, not the host port count.

1. NetApp E-Series E2800 Installation and Maintenance Guide

Document Rev. 5.

Section: "E2800 controller features"

Content: This section states

"Each E2800 controller includes two host ports." The accompanying diagrams of the controller rear panel clearly show two host ports

which can be configured for various protocols

including 10Gb iSCSI (Base-T).

2. NetApp Hardware Universe (HWU)

E2800 Controller Specifications.

Section: "Controller Specifications" for E2800.

Content: The official specifications list "Onboard host ports" as "2 per controller." It details the available protocols for these ports

including "10-Gbps iSCSI (SFP+ or Base-T)." For a duplex system

this confirms a total of four ports.

When the AutoSupport feature on a NetApp E-Series or EF-Series storage array is not functioning, the administrator can manually generate the same diagnostic package. This is accomplished using SANtricity System Manager, the web-based management interface for a single array. The "Collect Support Data" function gathers all relevant logs, configuration details, and performance data into a single compressed file. This file, often called a support bundle, contains the exact information that would have been sent by AutoSupport and can be manually uploaded to NetApp for analysis by ActiveIQ or support personnel.

A. Use SANtricity Unified Manager to collect the support data remotely

SANtricity Unified Manager is for managing multiple arrays, but the primary tool for in-depth, single-array diagnostic collection is SANtricity System Manager.

B. Access the event log by using SANtricity System Manager

The event log is only one component of the full support bundle. Collecting only the event log would omit critical configuration and performance data.

D. Ask NetApp Support to help you to access the data by using special support tools

This is a procedural step for escalation. The exam tests the administrator's ability to perform the task directly, which involves using standard management tools first.

1. NetApp E-Series SANtricity 11.7 System Manager Online Help

Section: "Collect support data".

Quote/Paraphrase: The documentation states

"You can collect extensive information about the storage array and save it to a file. This information is used for troubleshooting problems... The support data is collected in a single file with a .7z extension." This procedure is initiated from the Support > Support Center > Diagnostics tab in SANtricity System Manager

confirming it as the correct tool for manual data collection.

2. NetApp E-Series and EF-Series SANtricity AutoSupport On Demand and Proactive Monitoring

NetApp Technical Report TR-4555

Page 5

Section 2.1 "AutoSupport Feature".

Quote/Paraphrase: This report details the contents of an AutoSupport dispatch

which includes "event logs

the state of all storage array components

and performance data." This confirms that simply viewing the event log (Option B) is insufficient as it does not represent the complete data set. The manual collection process in System Manager is designed to gather this comprehensive bundle.