Splunk Core Certified Advanced Power User (SPLK-1004) — Free Practice Questions
10 free sample questions from a bank of 107, with the correct answers and explanations. No signup required — start practising right now.
1Which statement about tsidx files is accurate?
Splunk updates tsidx files every 30 minutes.
Splunk removes outdated tsidx files every 5 minutes.
A tsidx file consists of a lexicon and a posting list.
Each bucket in each index may contain only one tsidx file.
Answer: C
The short version
C — a tsidx file holds a lexicon plus a posting list. These two structures power indexed search over bucketed events.
Key concepts in this question
tsidx files: time-series index files inside buckets mapping terms to event locations.
Lexicon: sorted dictionary of unique terms found in indexed events.
Posting list: per-term list of event pointers and offsets where the term occurs.
Bucket lifecycle: hot, warm, cold, and frozen stages managing many tsidx files per bucket over time.
Why C is correct
C is correct because Splunk builds a lexicon of tokens during indexing and pairs each entry with a posting list of matching event locations. Query execution consults the lexicon to find terms, then follows posting lists to retrieve only relevant slices. This lexicon-plus-postings design is the documented anatomy of tsidx storage.
Why the others are wrong
A. Tsidx files update continuously during indexing, not on a fixed 30-minute schedule.
B. Outdated tsidx removal follows bucket rolling and retention, not a 5-minute purge timer.
D. Each bucket normally contains many tsidx files across slices, not a single file.
SPLK-1004 exam tip
Split tsidx into two parts: lexicon for terms, posting list for locations. Bloom filters then narrow candidate buckets before lexicon lookup, keeping searches efficient at scale.
2What is an example of the simple XML syntax for a base search and its post-process search?
,
,
,
,
Answer: A
The short version
A — the correct simple XML pairs a base search with post-process panels. One saved base powers multiple downstream post-process searches.
Key concepts in this question
Base search: saved search defined once in Simple XML powering dependent dashboard panels.
Post-process search: secondary search clause filtering base results without re-hitting indexers.
Simple XML structure: dashboard source with search id plus postProcess elements referencing the base.
Performance benefit: single index execution shared by many visualizations reduces load.
Why A is correct
A is correct because Splunk documentation shows a base search element carrying the full index query with an id, plus panels using postProcess to refine those results. This pattern matches the Basic post-process search example in Saved Searches docs. The settled key identifies the syntax that correctly nests base id and post-process clauses.
Why the others are wrong
B. Incorrect nesting or missing base reference breaks the post-process linkage.
C. Malformed search versus post-process tags fail Simple XML validation.
D. Omitting the base id or postProcess wrapper prevents result sharing.
SPLK-1004 exam tip
Look for id plus postProcess: base search defines once, panels refine with postProcess.
3What qualifies a report for acceleration?
Fewer than 100k events in search results, with transforming commands used in the search string.
More than 100k events in search results, with only a search command in the search string.
More than 100k events in the search results, with a search and transforming command used in the search string.
Fewer than 100k events in search results, with only a search and transaction command used in the search string.
Answer: C
The short version
C — acceleration needs 100K plus events with search plus transforming command. Volume plus summarizable structure qualifies the report.
Key concepts in this question
Report acceleration: prebuilt summary index speeding long-range transforming searches.
100K event threshold: minimum summary range size proving acceleration is worthwhile.
Transforming commands: stats, timechart, top, and chart producing aggregations suitable for summaries.
Distributable streaming prefix: only streaming commands before the transform so work parallelizes on indexers.
Why C is correct
C is correct because Splunk requires more than 100K events in the summary range to justify summary maintenance. It also requires a search plus a transforming command so results are aggregations that summaries can serve. Together volume and summarizable shape meet the documented acceleration prerequisites.
Why the others are wrong
A. Fewer than 100K events falls below the volume threshold despite having a transform.
B. Search-only without a transforming command yields raw events unsuitable for summaries.
D. Fewer than 100K plus transaction command fails both volume and transform suitability.
SPLK-1004 exam tip
Check volume plus verb: over 100K plus stats slash timechart slash top means acceleratable.
4What happens to panels with post-processing searches when their base search is refreshed?
The panels are deleted.
The panels are only refreshed if they have also been configured.
The panels are refreshed automatically.
Nothing happens to the panels.
Answer: C
The short version
C — post-process panels refresh automatically with the base. Dependent visualizations re-derive whenever base results update.
Key concepts in this question
Base search refresh: scheduled or manual rerun updating the shared result set.
Post-process dependency: panel queries derived from base output rather than independent index searches.
Automatic propagation: platform re-executing post-process logic on every base refresh.
Dashboard consistency: all panels staying aligned to the same base time range and data.
Why C is correct
C is correct because post-process searches operate on the in-memory base results, so refreshing the base invalidates and recomputes each dependent panel. Splunk documentation states post-process panels refresh whenever their base search refreshes. No extra panel configuration is needed to trigger this cascade.
Why the others are wrong
A. Panels are not deleted on base refresh, they are recomputed from new results.
B. No extra per-panel refresh flag is required beyond the base dependency.
D. Claiming nothing happens ignores the dependent recomputation mechanism.
SPLK-1004 exam tip
Dependence means auto-refresh: base rerun always recomputes its post-process panels. Time-range changes on the base therefore propagate instantly, keeping every dependent visualization synchronized.
5How is a cascading input used?
As part of a dashboard, but not in a form.
Without token notation in the underlying XML.
As a way to filter other input selections.
As a default way to delete a user role.
Answer: C
The short version
C — cascading inputs filter other input selections. Country choice narrows state, state narrows city for guided dashboards.
Key concepts in this question
Dashboard inputs: dropdowns and pickers feeding tokens into panel searches.
Cascading behavior: upstream selection constraining downstream choices through tokenized searches.
Token notation: dollar-wrapped variables passing selected values between inputs.
Form versus dashboard: inputs usable in both, with cascading as a filtering pattern not a deletion tool.
Why C is correct
C is correct because a cascading input defines its choices via a search filtered by the prior input token, such as states where country matches selection. Selecting a country therefore limits state options, and selecting a state limits city options. This chained filtering is the documented purpose of cascading inputs.
Why the others are wrong
A. Cascading inputs work in both dashboards and forms, not dashboards excluding forms.
B. Token notation is required to pass values, so tokenless operation is incorrect.
D. Deleting user roles is unrelated to input filtering behavior.
SPLK-1004 exam tip
Think country state city: upstream pick filters downstream picks via tokens. Search-driven dropdowns using token constraints keep selections valid and prevent impossible combinations.
6Which commands can run on both search heads and indexers?
Transforming commands
Centralized streaming commands
Dataset processing commands
Distributable streaming commands
Answer: D
The short version
D — distributable streaming commands run on search heads and indexers. They process each event independently wherever data lives.
Key concepts in this question
Distributable streaming: per-event operations like eval, fields, and where executable in parallel on indexers.
Centralized streaming: operations needing grouped context that must run on the search head.
Transforming commands: aggregations like stats and timechart that finalize on the search head.
Dataset processing: newer SPL2 dataset verbs with distinct execution placement.
Why D is correct
D is correct because distributable streaming commands touch one event at a time without global state, so Splunk can push them to indexers for parallel reduction. They also run on search heads when refining merged results, making them dual-capable. This portability reduces data movement and speeds distributed search.
Why the others are wrong
A. Transforming commands aggregate and finalize primarily on the search head.
B. Centralized streaming commands require search-head context and cannot distribute fully.
C. Dataset processing commands follow dataset semantics, not dual streaming placement.
SPLK-1004 exam tip
Per-event means portable: distributable streaming runs anywhere, centralized stays central.
7If a nested macro expands to a search string that begins with a generating command, what additional syntax is needed?
Double tick marks around the nested macro.
A comma before the nested macro.
Square brackets around the nested macro.
A pipe character before the nested macro.
Answer: D
The short version
D — prefix the nested macro with a pipe when it starts with a generating command. The pipe connects the macro output into the pipeline.
Key concepts in this question
Nested macros: macros calling other macros, expanded during parsing before execution.
Generating commands: search starters like inputlookup, makeresults, and tstats producing events without input.
Pipeline syntax: pipe character separating commands in SPL flow.
Macro expansion edge: nested string beginning with a command needing explicit piping at call site.
Why D is correct
D is correct because a macro expanding to stats count by something or another leading command must attach to the outer pipeline with a pipe. Writing pipe plus backtick macro backtick ensures the parser treats the expansion as the next pipeline stage. Without the pipe the SPL becomes syntactically disconnected and fails.
Why the others are wrong
A. Double ticks quote text and do not join pipeline stages.
B. Leading comma is invalid SPL between pipeline stages.
C. Square brackets denote subsearches, not macro pipeline insertion.
SPLK-1004 exam tip
See generating macro then add pipe: pipe plus backtick macro connects the nested starter. This convention keeps outer pipelines valid when inner content begins with inputlookup, makeresults, or stats generation.
8When using a nested search macro, how can an argument value be passed to the inner macro?
The argument value may be passed to the outer macro.
An argument cannot be used with an inner nested macro.
An argument cannot be used with an outer nested macro.
The argument value must be specified in the outer macro.
Answer: A
The short version
A — pass inner-macro arguments through the outer macro. Outer parameters forward values during layered expansion.
Key concepts in this question
Macro arguments: numbered placeholders filled at invocation for reusable search fragments.
Outer pass-through: defining outer signature to accept and forward values to the inner call.
Parsing-stage resolution: argument substitution happening before search execution begins.
Why A is correct
A is correct in this settled key because the outer macro signature can declare arguments that it substitutes into the inner macro invocation during expansion. Supplying the value once to the outer call propagates it inward without separate inner syntax. This pass-through model supports layered reuse while keeping invocation simple.
Why the others are wrong
B. Arguments can be used with nested macros through outer forwarding, so impossibility is incorrect.
C. Outer macros do accept arguments for forwarding, contrary to the prohibition claim.
D. Requiring values only inside the outer definition ignores the documented pass-through invocation path.
SPLK-1004 exam tip
Feed the outer to reach the inner: outer macro arguments forward into nested macro calls.
9Which is a regex best practice?
Use complex expressions rather than simple ones.
Avoid backtracking.
Use greedy operators (.) instead of non-greedy operators (.?).
Use * rather than +.
Answer: B
The short version
B — avoid backtracking as the core regex best practice. Simple non-greedy patterns with plus quantifiers run faster and safer.
Key concepts in this question
Backtracking: engine re-exploring pattern paths on failure, causing exponential slowdowns on large logs.
Greedy versus non-greedy: dot-star consuming maximally versus dot-question-star stopping early.
Plus versus star: requiring one-or-more versus allowing zero matches for tighter validation.
Simplicity principle: clear specific expressions outperforming complex catch-all patterns.
Why B is correct
B is correct because catastrophic backtracking stalls Splunk searches and risks timeouts on high-volume event data. Favoring anchored, specific, non-greedy expressions with appropriate quantifiers keeps extraction efficient. Avoiding backtracking is therefore the prescriptive best practice among the choices.
Why the others are wrong
A. Complex expressions increase maintenance and backtracking versus simple precise ones.
C. Greedy dot-star over-consumes and backtracks more than non-greedy dot-question-star.
D. Star allows empty matches where plus enforces meaningful content, weakening validation.
SPLK-1004 exam tip
Keep regex tight: simple plus non-greedy plus plus-quantifier avoids costly backtracking.
10What does the query | makeresults generate?
A timestamp
A results field
An error message
The results of the previously run search
Answer: A
The short version
A — pipe makeresults yields a timestamp field. It fabricates _time rows for testing without indexing data.
Key concepts in this question
makeresults command: generating command creating temporary in-memory results for demos and tests.
Default _time field: timestamp column produced even when no other fields are specified.
Count control: optional count and annotate options shaping how many rows and fields appear.
Pipeline seeding: starter for eval, stats, and formatting practice without real index data.
Why A is correct
A is correct because bare pipe makeresults emits one temporary event carrying _time set to now. Documentation notes it generates results in memory with a single _time field by default. This makes it ideal for dashboard scaffolding and SPL experimentation.
Why the others are wrong
B. No generic results field is created, only the timestamp unless annotate adds more.
C. Successful execution returns rows, not an error message.
D. It does not return prior search output, it fabricates fresh temporary events.
SPLK-1004 exam tip
Need a quick _time row: bare makeresults fabricates one timestamped event in memory. Add count or annotate options to shape volume and fields for dashboard scaffolding and SPL drills.