SPLK-2002 Splunk Enterprise Certified Architect Free Practice Exam Questions (2026 Updated)

According to the Splunk documentation1, the “Unknown sid” error message means that the search job associated with the link has expired or been deleted. The sid (search ID) is a unique identifier for each search job, and it is used to retrieve the results of the search. If the sid is not found, the search cannot be displayed. The other options are false because:

The users having insufficient permissions would result in a different error message, such as “You do not have permission to view this page” or "You do not have permission to run this search"1.

An add-on needing to be updated would not affect the validity of the sid, unless the add-on changes the search syntax or the data source in a way that makes the search invalid or inaccessible1.

One or more indexers being down would not cause the “Unknown sid” error, as the sid is stored on the search head, not the indexers. However, it could cause other errors, such as “Unable to distribute to peer” or "Search peer has the following message: not enough disk space"1.

 The logical first step when starting a deployment plan is to collect the initial requirements for the deployment from all stakeholders. This includes identifying the business objectives, the data sources, the use cases, the security and compliance needs, the scalability and availability expectations, and the budget and timeline constraints. Collecting the initial requirements helps to define the scope and the goals of the deployment, and to align the expectations of all the parties involved.

Inventorying the currently deployed logging infrastructure, determining what apps and use cases will be implemented, and gathering statistics on the expected adoption of Splunk for sizing are all important steps in the deployment planning process, but they are not the logical first step. These steps can be done after collecting the initial requirements, as they depend on the information gathered from the stakeholders.

A lookup table is a file that contains a list of values that can be used to enrich or modify the data during search time1. Lookup tables can be stored in CSV files or in the KV Store1. Troubleshooting lookup tables involves identifying and resolving issues that prevent the lookup tables from being accessed, updated, or applied correctly by the Splunk searches. Some of the tools and methods that can help with troubleshooting lookup tables are:

web_access.log: This is a file that contains information about the HTTP requests and responses that occur between the Splunk web server and the clients2. This file can help troubleshoot issues related to lookup table permissions, availability, and errors, such as 404 Not Found, 403 Forbidden, or 500 Internal Server Error34.

btool output: This is a command-line tool that displays the effective configuration settings for a given Splunk component, such as inputs, outputs, indexes, props, and so on5. This tool can help troubleshoot issues related to lookup table definitions, locations, and precedence, as well as identify the source of a configuration setting6.

search.log: This is a file that contains detailed information about the execution of a search, such as the search pipeline, the search commands, the search results, the search errors, and the search performance. This file can help troubleshoot issues related to lookup table commands, arguments, fields, and outputs, such as lookup, inputlookup, outputlookup, lookup_editor, and so on .

Option B is the correct answer because web_access.log is the best place to start troubleshooting lookup table issues, as it can provide the most relevant and immediate information about the lookup table access and status. Option A is incorrect because btool output is not a log file, but a command-line tool. Option C is incorrect because health.log is a file that contains information about the health of the Splunk components, such as the indexer cluster, the search head cluster, the license master, and the deployment server. This file can help troubleshoot issues related to Splunk deployment health, but not necessarily related to lookup tables. Option D is incorrect because configuration_change.log is a file that contains information about the changes made to the Splunk configuration files, such as the user, the time, the file, and the action. This file can help troubleshoot issues related to Splunk configuration changes, but not necessarily related to lookup tables.

Index files (. tsidx files) are the main components of an index that store the raw data and the inverted index of terms. They take the most space in an index, especially if the raw data has many unique terms that increase the size of the inverted index. Bloom filters, source metadata, and sourcetype metadata are much smaller in comparison and do not depend on the number of unique terms in the raw data.

The Splunk Managing Indexes and Storage Documentation explains that when multiple data sources with significantly different ingestion rates share a single index, index bucket management is governed by volume-based rotation, not by source or time. This means that high-volume data causes buckets to fill and roll more quickly, which in turn causes low-volume data to age out prematurely, even if it is relatively recent — hence Option C is correct.

Additionally, because Splunk organizes data within index buckets based on event time and storage characteristics, low-volume data mixed with high-volume data results in inefficient searches for smaller datasets. Queries that target the low-volume source will have to scan through the same large number of buckets containing the high-volume data, leading to slower-than-necessary search performance — Option B.

Compression efficiency (Option A) and performance optimization through data mixing (Option D) are not influenced by mixing volume patterns; these are determined by the event structure and compression algorithm, not source diversity. Splunk best practices recommend separating data sources into different indexes based on usage, volume, and retention requirements to optimize both performance and lifecycle management.

References (Splunk Enterprise Documentation):

• Managing Indexes and Storage – How Splunk Manages Buckets and Data Aging

• Splunk Indexing Performance and Data Organization Best Practices

• Splunk Enterprise Architecture and Data Lifecycle Management

• Best Practices for Data Volume Segregation and Retention Policies

The repFactor (replication factor) attribute in the indexes.conf file determines whether an index participates in indexer clustering and how many copies of its data are replicated across peer nodes.

When repFactor is set to 0, it explicitly instructs Splunk to exclude that index from participating in the cluster replication and management process. This means:

The index is not replicated across peer nodes.

It will not be managed by the Cluster Manager.

It exists only locally on the indexer where it was created.

Such indexes are typically used for local-only storage, such as _internal, _audit, or other custom indexes that store diagnostic or node-specific data.

By contrast:

repFactor=auto allows the index to inherit the cluster-wide replication policy from the Cluster Manager.

available_sites and site_mappings relate to multisite configurations, controlling where copies of the data are stored, but they do not remove the index from clustering.

Setting repFactor=0 is the only officially supported way to create a non-clustered index within a clustered environment.

References (Splunk Enterprise Documentation):

• indexes.conf Reference – repFactor Attribute Explanation

• Managing Non-Clustered Indexes in Clustered Deployments

• Indexer Clustering: Index Participation and Replication Policies

• Splunk Enterprise Admin Manual – Local-Only and Clustered Index Configurations

 The splunk diag --exclude command is a way to exclude search artifacts when creating a diag. A diag is a diagnostic snapshot of a Splunk instance that contains various logs, configurations, and other information. Search artifacts are temporary files that are generated by search jobs and stored in the dispatch directory. Search artifacts can be excluded from the diag by using the --exclude option and specifying the dispatch directory. The splunk diag --debug --refresh command is a way to create a diag with debug logging enabled and refresh the diag if it already exists. The splunk diag --disable=dispatch command is not a valid command, because the --disable option does not exist. The splunk diag --filter-searchstrings command is a way to filter out sensitive information from the search strings in the diag

This is the fall-back method for Splunk if .delta replication fails during knowledge bundle replication. Knowledge bundle replication is the process of distributing the knowledge objects, such as lookups, macros, and field extractions, from the search head cluster to the indexer cluster1. Splunk uses two methods of knowledge bundle replication: .delta replication and .bundle replication1. .Delta replication is the default and preferred method, as it only replicates the changes or updates to the knowledge objects, which reduces the network traffic and disk space usage1. However, if .delta replication fails for some reason, such as corrupted files or network errors, Splunk automatically switches to .bundle replication, which replicates the entire knowledge bundle, regardless of the changes or updates1. This ensures that the knowledge objects are always synchronized between the search head cluster and the indexer cluster, but it also consumes more network bandwidth and disk space1. The other options are not valid fall-back methods for Splunk. Option A, restarting splunkd, is not a method of knowledge bundle replication, but a way to restart the Splunk daemon on a node2. This may or may not fix the .delta replication failure, but it does not guarantee the synchronization of the knowledge objects. Option B, .delta replication, is not a fall-back method, but the primary method of knowledge bundle replication, which is assumed to have failed in the question1. Option D, restarting mongod, is not a method of knowledge bundle replication, but a way to restart the MongoDB daemon on a node3. This is not related to the knowledge bundle replication, but to the KV store replication, which is a different process3. Therefore, option C is the correct answer, and options A, B, and D are incorrect.

1: How knowledge bundle replication works 2: Start and stop Splunk Enterprise 3: Restart the KV store

When using ingest-based licensing, there are no Splunk roles that require the license manager to scale, because the license manager does not need to handle any additional load or complexity. Ingest-based licensing is a new licensing model that allows customers to pay for the data they ingest into Splunk, regardless of the data source, volume, or use case. Ingest-based licensing simplifies the licensing process and eliminates the need for license pools, license stacks, license slaves, and license warnings. The license manager is still responsible for enforcing the license quota and generating license usage reports, but it does not need to communicate with any other Splunk instances or monitor their license usage. Therefore, option C is the correct answer. Option A is incorrect because search peers are indexers that participate in a distributed search. They do not affect the license manager’s scalability, because they do not report their license usage to the license manager. Option B is incorrect because search heads are Splunk instances that coordinate searches across multiple indexers. They do not affect the license manager’s scalability, because they do not report their license usage to the license manager. Option D is incorrect because deployment clients are Splunk instances that receive configuration updates and apps from a deployment server. They do not affect the license manager’s scalability, because they do not report their license usage to the license manager12

1: https://docs.splunk.com/Documentation/Splunk/9.1.2/Admin/AboutSplunklicensing 2: https://docs.splunk.com/Documentation/Splunk/9.1.2/Admin/HowSplunklicensingworks

The splunk rebuild command is used for thawing the archive bucket. Thawing is the process of restoring frozen data back to Splunk for searching. Frozen data is data that has been archived or deleted from Splunk after reaching the end of its retention period. To thaw a bucket, the user needs to copy the bucket from the archive location to the thaweddb directory under SPLUNK_HOME/var/lib/splunk and run the splunk rebuild command to rebuild the .tsidx files for the bucket. The splunk collect command is used for collecting diagnostic data from a Splunk instance. The splunk convert command is used for converting configuration files from one format to another. The splunk dbinspect command is used for inspecting the status and properties of the buckets in an index.

According to the Splunk blog1, the Universal Forwarder is ideal for collecting data from high-velocity data sources, such as a syslog server, due to its smaller footprint and faster performance. The Universal Forwarder performs minimal processing and sends raw or unparsed data to the indexers, reducing the network traffic and the load on the forwarders. The other options are false because:

When most of the data requires masking, a Heavy Forwarder is needed, as it can perform advanced filtering and data transformation before forwarding the data2.

When data comes directly from a database server, a Heavy Forwarder is needed, as it can run modular inputs such as DB Connect to collect data from various databases2.

When a modular input is needed, a Heavy Forwarder is needed, as the Universal Forwarder does not include a bundled version of Python, which is required for most modular inputs2.

 The algorithm used to determine captaincy in a Splunk search head cluster is Raft distributed consensus. Raft is a consensus algorithm that is used to elect a leader among a group of nodes in a distributed system. In a Splunk search head cluster, Raft is used to elect a captain among the cluster members. The captain is the cluster member that is responsible for coordinating the search activities, replicating the configurations and apps, and pushing the knowledge bundles to the search peers. The captain is dynamically elected based on various criteria, such as CPU load, network latency, and search load. The captain can change over time, depending on the availability and performance of the cluster members. Rapt, Rift, and Round-robin are not valid algorithms for determining captaincy in a Splunk search head cluster

When installing Enterprise Security on a Search Head Cluster (SHC), the following steps should be done: Install Enterprise Security on the deployer, and use the deployer to deploy Enterprise Security to the cluster members. Enterprise Security is a premium app that provides security analytics and monitoring capabilities for Splunk. Enterprise Security can be installed on a SHC by using the deployer, which is a standalone instance that distributes apps and other configurations to the SHC members. Enterprise Security should be installed on the deployer first, and then deployed to the cluster members using the splunk apply shcluster-bundle command. Enterprise Security should not be installed on a staging instance, because a staging instance is not part of the SHC deployment process. Enterprise Security configurations should not be copied to the deployer, because they are already included in the Enterprise Security app package.

Per the Splunk Enterprise Licensing Documentation, a license peer (such as an indexer or search head) must regularly communicate with its license manager to report data usage and verify license validity. Splunk allows a 72-hour grace period during which the peer continues operating normally even if communication with the license manager fails.

If this communication is not re-established within 72 hours, the peer enters a “license violation” state. In this state, the system blocks all search activities, including ad-hoc and scheduled searches, but continues to ingest and index data. Administrative and licensing-related searches may still run for diagnostic purposes, but user searches are restricted.

The intent of this design is to prevent prolonged unlicensed data ingestion while ensuring the environment remains compliant. The 72-hour rule is hard-coded in Splunk Enterprise and applies uniformly across license types (Enterprise or Distributed). This ensures consistent licensing enforcement across distributed deployments.

Warnings are generated during the grace period, but after 72 hours, searches are automatically blocked until the peer successfully reconnects to its license manager.

References (Splunk Enterprise Documentation):

• Managing Licenses in a Distributed Environment

• License Manager and Peer Communication Workflow

• Splunk License Enforcement and Violation Behavior

• Splunk Enterprise Admin Manual – License Usage and Reporting Policies

The TailingProcessor channel in the splunkd.log file would help troubleshoot this issue, because it contains information about the files that Splunk monitors and indexes, such as the file path, size, modification time, and CRC checksum. It also logs any errors or warnings that occur during the file monitoring process, such as permission issues, file rotation, or file truncation. The TailingProcessor channel can help identify if Splunk is reading the new data from the monitor input file or not, and what might be causing the problem. Option B is the correct answer. Option A is incorrect because the ModularInputs channel logs information about the modular inputs that Splunk uses to collect data from external sources, such as scripts, APIs, or custom applications. It does not log information about the monitor input file. Option C is incorrect because the ChunkedLBProcessor channel logs information about the load balancing process that Splunk uses to distribute data among multiple indexers. It does not log information about the monitor input file. Option D is incorrect because the ArchiveProcessor channel logs information about the archive process that Splunk uses to move data from the hot/warm buckets to the cold/frozen buckets. It does not log information about the monitor input file12

1: https://docs.splunk.com/Documentation/Splunk/9.1.2/Troubleshooting/WhatSplunklogsaboutitself#splunkd.log 2: https://docs.splunk.com/Documentation/Splunk/9.1.2/Troubleshooting/Didyouloseyourfishbucket#Check_the_splunkd.log_file

Splunk Enterprise Search Optimization documentation consistently emphasizes that filtering data as early as possible in the search pipeline is the most effective way to improve search performance. The base search (the part before the first pipe |) determines the volume of raw events Splunk retrieves from the indexers. Therefore, by applying restrictive conditions early—such as time ranges, indexed fields, and metadata filters—you can drastically reduce the number of events that need to be fetched and processed downstream.

The best practice is to use indexed field filters (e.g., index=security sourcetype=syslog host=server01) combined with search or where clauses at the start of the query. This minimizes unnecessary data movement between indexers and the search head, improving both search speed and system efficiency.

Using non-streaming commands early (Option C) can degrade performance because they require full result sets before producing output. Likewise, focusing solely on shortening queries (Option A) or excessive use of the not operator (Option D) does not guarantee efficiency, as both may still process large datasets.

Filtering early leverages Splunk’s distributed search architecture to limit data at the indexer level, reducing processing load and network transfer.

References (Splunk Enterprise Documentation):

• Search Performance Tuning and Optimization Guide

• Best Practices for Writing Efficient SPL Queries

• Understanding Streaming and Non-Streaming Commands

• Search Job Inspector: Analyzing Execution Costs

Comprehensive and Detailed Explanation (From Splunk Enterprise Documentation)Splunk’s official planning guidance explains that the deployment cycle begins with defining the system architecture, capacity planning, platform design, and topologies. Splunk states that before any component is installed, administrators must complete “infrastructure planning and buildout,” which includes determining indexer capacity, search head roles, clustering strategy, storage layout, and performance requirements. This foundational step ensures that all Splunk components have the proper hardware, network design, and scaling expectations.

After the environment is built, Splunk documentation states that the next stage is “deployment and data onboarding,” which includes configuring indexers, forwarders, parsing rules, event processing pipelines, data source validation, and enrichment steps such as field extractions, tagging, event types, and CIM alignment. Splunk describes this as the phase where you bring in data and confirm correctness, completeness, and normalization.

Only after the system is stable and populated with data does Splunk recommend “user planning and rollout”, which includes developing dashboards, roles, knowledge objects, search best practices, and enabling user access. Splunk emphasizes that user onboarding should occur last, once infrastructure and data pipelines are fully validated.

 The Splunk instance with the given settings in SPLUNK_HOME/etc/system/local/server.conf is missing the master_uri attribute and needs to be restarted. The master_uri attribute is required for the master node to communicate with the peer nodes and the search head cluster. The master_uri attribute specifies the host name and port number of the master node. Without this attribute, the master node cannot function properly. The Splunk instance also needs to be restarted for the changes in the server.conf file to take effect. The replication_factor setting determines how many copies of each bucket are maintained across the peer nodes. The search factor is a separate setting that determines how many searchable copies of each bucket are maintained across the peer nodes. The search factor is not specified in the given settings, so it defaults to the same value as the replication factor, which is 2. This is not a multi-site cluster, because the site attribute is not specified in the clustering stanza. A multi-site cluster is a cluster that spans multiple geographic locations, or sites, and has different replication and search factors for each site.

According to the Indexer Clustering Administration Guide, configuration bundles pushed from the Cluster Manager (Master Node) overwrite the contents of the $SPLUNK_HOME/etc/slave-apps/ directory on each search peer (indexer). However, Splunk provides a special persistent location — the _cluster app’s local directory — for files that must survive bundle redeployments.

Specifically, any configuration files placed in:

$SPLUNK_HOME/etc/slave-apps/_cluster/local/

will persist after future bundle pushes because this directory is excluded from the automatic overwrite process.

This is particularly useful for maintaining local overrides or custom configurations that should not be replaced by the Cluster Manager, such as environment-specific inputs, temporary test settings, or monitoring configurations unique to that peer.

Other directories under slave-apps are overwritten each time a configuration bundle is pushed, ensuring consistency across the cluster. Likewise, master-apps exists only on the Cluster Manager and is used for deployment, not persistence.

Thus, the _cluster/local folder is the only safe, Splunk-documented location for configurations that need to survive bundle redeployment.

References (Splunk Enterprise Documentation):

• Indexer Clustering: How Configuration Bundles Work

• Maintaining Local Configurations on Clustered Indexers

• slave-apps and _cluster App Structure and Behavior

• Splunk Enterprise Admin Manual – Cluster Configuration Management Best Practices