Collector

This configuration tab contains the classification engine settings. Each configuration option has an associated “i” which describes the nature of the setting.

Option	Description	Comment
General settings
Max Document Size	Sets the maximum size of the document to be processed.	Documents exceeding this size will typically be excluded from processing.
Collect Metadata of Excluded Items	When enabled, the Netwrix Data Classification services will include the document, but from a metadata standpoint only (no text will be extracted from the file).	Used in combination with the "Max Document Size" value. Inactive by default.
Collector Threads	The number of overall background threads to be utilized by the Collector to access content from the source system.	Each thread can be considered a "user" when considering load on the source system. For more information, see this Knowledge Base article.
Collector Domain Threads	The number of threads to be utilized by the Collector to access content from each HTTP domain. (Examples: netwrix.com, google.com, microsoft.com, etc.) The number will be automatically capped by the "Collector Threads" value.	Applies to HTTP source types only. Each thread can be considered a "user" when considering load on the source system. For more information, see this Knowledge Base article.
Collector File Threads	The number of threads used to crawl file system content.	For more information, see this Knowledge Base article.
Process Document Images	If enabled, images will be extracted from supported documents (Office XML files and PDFs). These images will be then collected, and any text found will be included (with the document text) for classification.	For this setting to be applied, OCR must also be enabled at a content type level — to ensure that the extracted images are run through the OCR engine. This setting is inactive by default.
Maximum Images per Document	Maximum amount of images to process through OCR on a per document basis.
Minimum Image Resolution Width	Minimum resolution (in pixels) of images to process through OCR.
Minimum Image Resolution Height	Minimum resolution (in pixels) of images to process through OCR.
Document Set Mode	Specifies how SharePoint document sets will be treated. Possible options: - Process as Folder — Classifications will only be written to the child items. - Process Set and Children— Classifications will be written to both the root item and individual children. - Process Set as Document (default) — Classifications will only be written to the root item.	Applies to SharePoint documents.
Duplicate Detection Scope	Instructs to exclude detected duplicates from processing within the specified scope: Global, Source, Source Group	Applies to File and web sources.
Advanced settings
Collector User Agent	Is used by Collector service as part of each web request made when crawling HTTP sources — to identify itself to the crawled systems.
Encrypt Text (text.cse)	Encrypts all data stored in text.cse (raw document extracts).	Inactive by default. If data already exists in the index, then to enable encryption on that existing data, you should perform re-collection. For that, click Run Cleaner button on the right.
Optimize Text Storage	Reduces storage requirements for stored text.	Enabled by default. At each re-crawl or re-index the program tries to detect whether the document text has changed.
Re-use Text Offsets	Reduces storage requirements for stored text by sharing and reusing the stored text.	May slightly increase the SQL demands — in order to process each de-duplication command.
Collector Delay	The sleep time (in milliseconds) between intensive operations, such as storing crawled text. Default is 1 ms.
Collector Polling	The sleep time (in seconds) between Collector batches.	Only utilized when the Collector queue is empty.
iFilter Processing Mode	Specify where the iFilter processing will run. Possible options: Process as Sub Process — run in a separate process Process Internally — run within Collector process
Collector Reader Process Pool Size	The number of external processes that will be utilized for iFilter conversion.	Each additional process adds additional load on the Netwrix Data Classification server. We recommend leaving this setting on its default value. For more information, see this Knowledge Base article.