Splunk IT Service Intelligence Certified Admin (SPLK-3002) — Free Practice Questions
10 free sample questions from a bank of 102, with the correct answers and explanations. No signup required — start practising right now.
1After a notable event has been closed, how long will the meta data for that event remain in the KV Store by default?
6 months.
9 months.
1 year.
3 months.
Answer: A
The short version
A — Closed notable metadata stays 6 months by default. ITSI keeps the KV Store record for a closed notable for half a year, so analysts can still review, reopen, report and audit it before automatic cleanup removes the metadata.
Key concepts in this question
Notable event lifecycle: ITSI notables move through open, acknowledged, assigned and closed states, with closure marking operational completion but not immediate deletion.
KV Store metadata: ITSI stores notable state, owner, severity, comments and grouping info in KV Store collections separate from the raw index data that triggered the alert.
Retention default: The platform applies a time-based housekeeping rule that purges closed-notable metadata after a fixed window to bound KV growth while preserving recent history.
Operational review need: Teams often revisit recently closed notables for post-incident review, SLA proof, custom reporting and re-escalation, which requires months of retained context.
Why A is correct
A is correct because ITSI default retention for closed notable metadata in the KV Store is 6 months. That value balances audit and rework needs against KV Store size and performance. After an analyst closes a notable, its metadata remains searchable and reportable for that window, then housekeeping removes it automatically unless the retention setting was customized.
Why the others are wrong
B. 9 months overstates the default window and would retain closed records longer than ITSI configures out of the box.
C. 1 year overstates retention by double and confuses long-term index retention of raw events with short-term KV retention of notable state.
D. 3 months understates the default and would discard review, reopen and reporting context too quickly for normal operations.
SPLK-3002 exam tip
Remember closed notables equal 6 months in KV, half a year of housekeeping before auto-cleanup.
2What are valid ITSI Glass Table editor capabilities? (Choose all that apply.)
Creating glass tables.
Correlation search creation.
Service swapping configuration.
Adding KPI metric lanes to glass tables.
Answer: A, C
The short version
AC — Glass Table editor creates tables and configures service swapping. In ITSI the Glass Table editor builds visual service topologies and lets an operator swap which underlying service a widget points to, while search authoring and KPI lane design belong to other views.
Key concepts in this question
Glass Table purpose: Glass Tables are free-form operational canvases that combine service health, KPIs, text, shapes and external content for NOC and executive views.
Editor capabilities: The editor focuses on layout, data source binding, service mapping and visual thresholds rather than on SPL or detection logic.
Service swapping: Swapping lets one template table be reused across services by repointing widgets, which speeds rollout of standard views for many similar services.
Deep Dive lanes contrast: KPI metric lanes and swim-lane time analysis live in Deep Dives, while correlation search creation lives in the correlation search management interface.
Why AC is correct
AC is correct because creating glass tables and configuring service swapping are native Glass Table editor functions. Creation covers new canvases, cloning, widget placement and service health binding. Service swapping lets editors change the backing service without rebuilding the layout, supporting templated operations views. Both align with the visual mapping role of Glass Tables in ITSI.
Why the others are wrong
B. Correlation search creation is wrong because detection logic is authored in the correlation search manager, not in the visual Glass Table editor.
D. Adding KPI metric lanes is wrong because metric lanes and swim-lane investigation are Deep Dive constructs for time-based KPI comparison, not Glass Table widgets.
SPLK-3002 exam tip
Link Glass with seeing and swapping, Deep Dive with lanes over time, correlation searches with detection authoring.
3Which of the following is the best use case for configuring a Multi-KPI Alert?
Comparing content between two notable events.
Using machine learning to evaluate when data falls outside of an expected pattern.
Comparing anomaly detection between two KPIs.
Raising an alert when one or more KPIs indicate an outage is occurring.
Answer: D
The short version
D — Multi-KPI Alert fires when combined KPIs signal an outage. It watches several KPIs together so a single alert can represent service-level failure, reducing noise compared with one alert per KPI when the business impact is what matters.
Key concepts in this question
Multi-KPI Alert definition: A single ITSI alert definition that evaluates logical conditions across two or more KPIs in a service, such as critical on any one or majority breach.
Outage detection pattern: Outages usually depress or spike multiple related indicators at once, so combined evaluation is more reliable than isolated thresholds.
Alert noise reduction: Grouping related KPI breaches into one service-level alert prevents a flood of per-KPI notables during the same incident.
Contrast with analytics: Anomaly comparison or ML pattern checks study behavior shape, while Multi-KPI Alerts make an operational fire-or-not decision for response.
Why D is correct
D is correct because the documented best use for Multi-KPI Alerts is raising an alert when one or more KPIs indicate an outage is occurring. The feature expresses combined breach logic, then opens one high-urgency notable. That matches service-centric monitoring where the outage condition spans several health signals at once.
Why the others are wrong
A. Comparing content between two notable events is wrong because that is episode review and notable comparison work, not a threshold alert use case.
B. Using machine learning to evaluate expected patterns is wrong because that describes anomaly detection or adaptive thresholds, not multi-KPI Boolean alerting.
C. Comparing anomaly detection between two KPIs is wrong because it describes analysis of detection output, not firing an outage alert from live KPI states.
SPLK-3002 exam tip
Hear multi-KPI alert and think outage logic across signals, one alert for service down not per-metric science.
4In distributed search, which components need to be installed on instances other than the search head?
SA-IndexCreation and SA-ITSI-Licensechecker on indexers.
SA-IndexCreation and SA-ITOA on indexers; SA-ITSI-Licensechecker and SA-UserAccess on the license master.
SA-IndexCreation on idexers; SA-ITSI-Licensechecker and SA-UserAccess on the license master.
SA-ITSI-Licensechecker on indexers.
Answer: C
The short version
Banked answer A is wrong and the correct choice is C, because ITSI requires SA-IndexCreation on indexers and SA-ITSI-Licensechecker plus SA-UserAccess on the license manager, not on indexers. This split matches the official distributed install matrix and the search head cluster setup steps.
Key concepts in this question
Key ideas are where ITSI add-ons live in a distributed deployment, what SA-IndexCreation does for index-time setup, what SA-ITSI-Licensechecker and SA-UserAccess do for licensing and access, and the difference between an indexer role and a license manager role. Search heads run ITSI itself while remote tiers get only the pieces they need.
Why C is correct
C is correct because Splunk docs state SA-IndexCreation is required on all indexers, with copy steps for standalone indexers and bundle replication for clusters, while SA-ITSI-Licensechecker and SA-UserAccess belong on any license manager. That is exactly the C wording, including license master placement. Vendor prose pointing to slide 117 and unanimous community vote for C align with this documented split.
Why the others are wrong
A is wrong because it places SA-ITSI-Licensechecker on indexers, where docs never place it. B is wrong because it adds SA-ITOA on indexers, which is not part of the documented indexer requirement, and it misstates the license tier contents. D is wrong because it omits SA-IndexCreation on indexers and omits SA-UserAccess on the license master, leaving both required tiers incomplete for a working distributed install.
SPLK-3002 exam tip
For SPLK-3002, memorize the three-row install table: search heads get full ITSI, indexers get only SA-IndexCreation, license manager gets SA-ITSI-Licensechecker plus SA-UserAccess. If an answer moves the licensechecker to indexers or drops UserAccess, eliminate it fast and look for the license master choice.
5When deploying ITSI on a distributed Splunk installation, which component must be installed on the search head(s)?
SA-ITOA
ITSI app
All ITSI components
SA-ITSI-Licensechecker
Answer: C
The short version
C — Distributed ITSI puts all components on the search heads. ITSI search-time knowledge, data model acceleration, KPI calculation and UI all execute at search time, so each search head needs the full stack for correct results.
Key concepts in this question
Distributed Splunk roles: Indexers store and search raw data while search heads parse, correlate, calculate and render, with add-ons needed where their knowledge is used.
ITSI component set: The ITSI app plus supporting add-ons such as SA-ITOA, SA-ITSI-Licensechecker and content modules together provide KPIs, services, notables and visualizations.
Search-head execution: KPI base searches, service health calculation, notable generation and Glass Table rendering all run on the search head tier.
Consistency requirement: In search head clusters every member must present identical ITSI behavior, which demands identical full installation across members.
Why C is correct
C is correct because Splunk documentation requires all ITSI components on the search heads in a distributed deployment. Installing only a single add-on leaves KPI searches, service definitions, license checks or UI elements missing, causing failed searches or inconsistent health scores. Full installation ensures every search head can independently compute services, generate notables and serve ITSI views.
Why the others are wrong
A. SA-ITOA alone is wrong because that support add-on supplies only part of the data and search-time logic without the main ITSI app and UI.
B. ITSI app alone is wrong because the app depends on its supporting add-ons for collection, license checks and technology content.
D. SA-ITSI-Licensechecker alone is wrong because license validation is only one prerequisite and provides no services, KPIs or views by itself.
SPLK-3002 exam tip
For distributed ITSI think search heads carry everything, indexers hold data, full stack travels together.
6Which of the following describes entities? (Choose all that apply.)
Entities must be IT devices, such as routers and switches, and must be identified by either IP value, host name, or mac address.
An abstract (pseudo/logical) entity can be used to split by for a KPI, although no entity rules or filtering can be used to limit data to a specific service.
Multiple entities can share the same alias value, but must have different role values.
To automatically restrict the KPI to only the entities in a particular service, select “Filter to Entities in Service”.
Answer: D
The short version
D — Filter to Entities in Service scopes a KPI to service members. That checkbox makes the KPI search only over entities currently assigned to the service, keeping service health tied to its true inventory without manual SPL filtering.
Key concepts in this question
Entity definition: In ITSI an entity is any managed object that can affect service health, including hosts, services, databases, containers, users or abstract logical items, not only network gear.
Service membership: Entities join services through import rules, manual assignment, lookup or search-based discovery, and membership can change as infrastructure evolves.
KPI entity filtering: KPIs can split by entity and can either scan broadly or be constrained to in-service entities, which controls which rows influence health.
Alias and role model: Aliases group equivalent entities and roles tag function, allowing flexible matching without requiring identical host or address values.
Why D is correct
D is correct because selecting Filter to Entities in Service automatically restricts KPI calculation to only the entities in that particular service. That keeps shared searches from pulling unrelated hosts. It is the supported declarative way to bind KPI scope to dynamic service membership instead of hard-coding lists in SPL.
Why the others are wrong
A. Must be IT devices identified by IP, host or MAC is wrong because entities can be non-device, abstract or business objects with varied identifiers.
B. Abstract entity with no rules or filtering possible is wrong because service-level entity filtering and split-by remain available design controls.
C. Same alias requires different roles is wrong because alias sharing and role assignment are flexible modeling choices, not a strict uniqueness rule.
SPLK-3002 exam tip
Remember Filter to Entities in Service equals KPI follows service inventory automatically.
7Which of the following describes a realistic troubleshooting workflow in ITSI?
Correlation Search –> Deep Dive –> Notable Event
Service Analyzer –> Notable Event Review –> Deep Dive
Service Analyzer –> Aggregation Policy –> Deep Dive
Correlation search –> KPI –> Aggregation Policy
Answer: B
The short version
B — Real triage goes Service Analyzer to Notables to Deep Dive. Operators first see degraded service health, then pick the related notable for context and ownership, then drill into KPI timelines to find root cause.
Key concepts in this question
Service Analyzer role: The high-level operations dashboard that surfaces service health scores and KPI status with color and priority for rapid impact recognition.
Notable Event Review role: The queue where correlation searches and alerts create actionable notables with severity, owner, status and episode grouping for workflow.
Deep Dive role: The time-based investigative canvas with swim lanes, KPI values and threshold bands that reveals when and in what order signals degraded.
Workflow direction: ITSI is designed to flow from business impact down to technical evidence, not from raw search syntax up to impact.
Why B is correct
B is correct because it follows the intended impact-to-evidence path. An operator notices red or critical health in Service Analyzer, pivots to Notable Event Review to select the relevant notable or episode and claim ownership, then opens Deep Dive to compare KPI trends, threshold breaches and event timing. Each step narrows scope while adding diagnostic detail, which matches training and product design.
Why the others are wrong
A. Correlation Search to Deep Dive to Notable is wrong because it starts with detection config and ends with the queue, reversing triage order.
C. Service Analyzer to Aggregation Policy to Deep Dive is wrong because aggregation policies group notables behind the scenes and are not an analyst triage stop.
D. Correlation search to KPI to Aggregation Policy is wrong because it lists configuration objects rather than the operational see-triage-investigate sequence.
SPLK-3002 exam tip
Memorize see it, queue it, dive it, Analyzer to Notables to Deep Dive is the ITSI mantra.
8Which of the following accurately describes base searches used for KPIs in a service?
Base searches can be used for multiple services.
A base search can only be used by its service and all dependent services.
All the metrics in a base search are used by one service.
All the KPIs in a service use the same base search.
Answer: A
The short version
A — One base search can feed multiple services. ITSI base searches separate expensive data gathering from per-KPI thresholding, so the same shared search output can supply KPIs in different services and cut search load.
Key concepts in this question
Base search purpose: A saved SPL fragment that retrieves and pre-aggregates raw data once, with individual KPIs applying additional stats, splits and threshold evaluation on its results.
Reuse model: Because the base output is generic rows with entity and metric fields, many KPIs across many services can consume it without rerunning the raw search.
Performance benefit: Sharing reduces scheduler load, index scans and summary latency compared with duplicating near-identical searches per service.
Service independence: KPI-to-service binding happens at the KPI layer, so base search sharing does not force services to share thresholds, importance or health calculation.
Why A is correct
A is correct because ITSI explicitly allows base searches to be used for multiple services. Administrators build one efficient gathering search, then map different KPIs in different services to it with their own calculations and time policies. This reuse is a core scaling practice and is visible in the base search library where ownership is not locked to a single service.
Why the others are wrong
B. Only its service plus dependents is wrong because it invents a hierarchy restriction, while sharing is intentionally broad and cross-service.
C. All metrics used by one service is wrong because a base search can return many fields while each KPI selects only the metrics it needs.
D. All KPIs in a service use the same base is wrong because a service often mixes availability, performance and capacity KPIs that need different source searches.
SPLK-3002 exam tip
Think one base search many KPIs many services, gather once and threshold many times.
9Which scenario would benefit most by implementing ITSI?
Monitoring of business services functionality.
Monitoring of system hardware.
Monitoring of system process statuses.
Monitoring of retail sales metrics.
Answer: A
The short version
A — ITSI shines for business service monitoring. Its services, KPIs, health scores and episodes translate technical telemetry into whether checkout, payments, care or core apps actually work for users.
Key concepts in this question
ITSI value proposition: IT Service Intelligence correlates infrastructure and application metrics into service health that reflects user and revenue impact.
Service modeling: Services group entities and KPIs by business function, with importance and dependency weighting so executive views show what matters.
Versus raw monitoring: Hardware, process or single-metric checks are well served by core Splunk or infrastructure tools without service abstraction.
Business alignment: ITSI adds Glass Tables, Service Analyzer, Deep Dives and episode management to prioritize response by service consequence.
Why A is correct
A is correct because monitoring business service functionality is the primary ITSI use case. When leadership asks whether ordering, streaming or banking works, ITSI aggregates availability, latency, errors and capacity into color-coded health and actionable notables. That service lens, not isolated device status, justifies ITSI licensing and modeling effort.
Why the others are wrong
B. Monitoring system hardware is wrong because host CPU, disk and memory alerting needs no service tree and is handled by simpler infrastructure monitoring.
C. Monitoring process statuses is wrong because per-process up or down checks are low-level signals that become ITSI inputs, not the service outcome itself.
D. Monitoring retail sales metrics is wrong because pure business sales analytics belongs in BI reporting, while ITSI monitors whether the service enabling sales is healthy.
SPLK-3002 exam tip
If the question asks best fit for ITSI, choose business service health over device or single metric watching.
10ITSI Saved Search Scheduling is configured to use realtime_schedule = 0. Which statement is accurate about this configuration?
If this value is set to 0, the scheduler bases its determination of the next scheduled search execution time on the current time.
If this value is set to 0, the scheduler bases its determination of the next scheduled search on the last search execution time.
If this value is set to 0, the scheduler may skip scheduled execution periods.
If this value is set to 0, the scheduler might skip some execution periods to make sure that the scheduler is executing the searches running over the most recent time range.
Answer: B
The short version
B — With realtime_schedule set to 0, next run is based on last run time. A false or zero value tells the scheduler to anchor the next execution to when the search last ran, preserving continuity rather than snapping to wall-clock boundaries. Current-time anchoring only happens when the setting is enabled.
Key concepts in this question
Saved search scheduling: Splunk mechanism that fires KPI and alert searches on a cron or interval cadence to keep ITSI health and notables current.
realtime_schedule setting: savedsearches.conf flag controlling whether the scheduler tolerates drift, with false meaning preserve the original cadence and true meaning re-align to current time.
Last-run anchoring: behavior where a delayed search still schedules its successor relative to its actual execution, preventing gaps or pileups after scheduler lag.
Why B is correct
The savedsearches.conf reference states that when the value is set to false, the scheduler determines the next scheduled search run time based on the last run time for the search. A zero in this boolean context equals false, so the described behavior applies directly. That makes B the accurate restatement of documented scheduler behavior.
Why the others are wrong
A. Basing determination on current time describes the enabled or true behavior, which is the opposite of what zero selects.
C. Merely skipping execution periods without qualification is vague and actually aligns with the enabled mode that drops missed windows, not with the continuity-preserving false mode.
D. Skipping periods to run over the most recent time range describes the real-time-oriented true behavior, so it cannot also describe the zero setting.
SPLK-3002 exam tip
Zero means follow the last run: false preserves cadence from last execution, true snaps to now and may skip to the freshest window.