Machine Data Lake prerequisites

Verify the Splunk Cloud Platform, Data Management, Ingest Processor, search, service-account, user-permission, and routing requirements before onboarding Machine Data Lake.

Before you begin checklist

Verify each requirement before you create a raw table. Use the observable result, not only the existence of a configuration item, to decide whether the environment is ready.

  • Splunk Cloud Platform version and Machine Data Lake availability

    How to verify: Confirm that the target Splunk Cloud Platform environment runs a supported version and is enabled for Machine Data Lake. For current requirements, see Release notes.

    Expected result: The target environment supports Machine Data Lake raw table creation, data landing, raw search, promotion preview, and promotion workflows.

    If verification fails: Contact Splunk Support before onboarding.

  • Search-head target

    How to verify: Identify the search head unit (SHU) or search head cluster where you plan to use Machine Data Lake. Use that same target for the Data Management connection and the service-account role. In most environments, use the default search head.

    Expected result: The Data Management connection and service-account role use the same search-head target as your Machine Data Lake workflows.

    If verification fails: If your environment has multiple SHUs or you need a different search head, contact Splunk Support before changing the target.

  • Data Management connection

    How to verify: Open Data Management and confirm that the intended Splunk Cloud Platform deployment is available as the connected target. If you have not completed first-time setup, follow First-time setup instructions for the Ingest Processor solution before you provision Machine Data Lake.

    To review the connection settings, open Settings > System connections in Data Management. Confirm that the scpbridge connection uses the service account you configured during first-time setup and shows a connected status. If you changed the service-account role, refresh the connection.

    Expected result: The scpbridge connection shows a connected status and manage resources for raw tables, data landing, preview, and promotion inuses the intended service account.

    If verification fails: Review the connection settings and service-account credentials by following the Ingest Processor setup before you create a raw tableinstructions. If the connection still fails, contact Splunk Support.

    If Ingest Processor setup succeeds but Machine Data Lake provisioning fails, see Machine Data Lake shows an access or 403 errorNo Content found for /db/organizations/splunk/repositories/splunkcloud-10_5_2605/content/documents/Platform/SplunkCloud/SplunkCloud_MachineDataLake/tr_mdl_access_error.dita.

  • Search head target

    How to verify: Confirm that the Data Management connection and service account use the same search head unit (SHU) or search head cluster that customers use for Machine Data Lake workflows.

    Expected result: Users create and manage raw tables, data landing, raw search, promotion preview, and promotion workflows from the SHU linked to the Data Management connection. In most environments, use the default search head.

    If verification fails: If the environment has multiple SHUs or you need a different search head, such as a search head that runs Splunk Enterprise Security, contact Splunk Support before changing the target.

  • Service account and required capabilities

    How to verify: Confirm that you have created a service account and assigned it the role described in First-time setup instructions for the Ingest Processor solution. Machine Data Lake uses this same account through the scpbridge connection. Verify that the service account's effective capabilities include all of the following:

    • list_search_head_clustering
    • edit_spl2_datasets
    • edit_connections
    • edit_datasets
    • read_datasets
    • write_datasets
    • read_connections
    • indexes_edit
    • search

    Expected result: The service account has every capability listed above before you provision Machine Data Lake.

    If verification fails: Ask your Splunk Cloud Platform administrator to create the account or update its assigned role by following the setup instructions.

  • Ingest Processor resources

    How to verify: Verify that Ingest Processor resources are available and healthy for dataset landing and promotion flows. Review applicable limits in Service limits and constraints.

    Expected result: Data landing and streaming promotion workflows have available Ingest Processor capacity and required services report healthy status.

    If verification fails: Resolve service health or capacity issues before creating raw tables or streaming promotions. If the required health or capacity status is not available to you, contact Splunk Support.

  • SPL2 and data orchestration services

    How to verify: Verify that SPL2 modules and data orchestration services are available for raw table search and analytics search.

    Expected result: Raw table search and analytics search paths are available after data lands or promotion completes.

    If verification fails: Ask an administrator to verify SPL2 modules and data orchestration service health before you retry the workflow. If the services remain unavailable, contact Splunk Support.

  • User permissions

    How to verify: Confirm that the user has the dataset access and role capabilities required for the intended action. For details, see Access control, roles, and capabilities.

    Expected result: The user can see the required dataset, creation, search, promotion, sharing, edit, or delete action in the workflow.

    If verification fails: Ask an administrator to grant or update the required role, dataset access level, or sharing capability.

  • Input and routing strategy

    How to verify: Confirm the supported input path, such as HEC, Universal Forwarder, or Heavy Forwarder, and decide whether matching events land only in Machine Data Lake or land in Machine Data Lake and also continue to an existing Splunk index path where supported.

    Expected result: Events are already reaching the Machine Data Lake-enabled Splunk Cloud Platform environment, and the expected routing model is documented before raw table creation.

    If verification fails: Prepare the source first. For details, see Prepare a data source for Machine Data Lake.