Understanding system tasks

System tasks take care of scheduling tasks not related to search sources. There are far fewer system tasks than search data source tasks.

Only administrators with accounts that include the PRIV_TASK_MANAGE privilege can access, schedule, and launch tasks.

You can execute these system tasks:

System Task Description

SD_ASSET_EXTRACTOR

Harvests metadata in the asset repository.

The metadata is harvested from files in the repository and from metadata entered when adding or editing an asset (see Adding a digital asset and Modifying the details of a digital asset).

Portfolio indexes the following types of files:

  • *.pdf
  • *.txt, *.rtf (plain text, rich text)
  • *.doc, *.docx (Microsoft Word)
  • *.xls, *.xlsx (Microsoft Excel)
  • *.jpg
  • *.gif
  • *.tiff
  • *.png
  • *.bmp

After Portfolio runs the task, it then runs SD_ASSET_IMPORT, and the SD_ASSET_INDEX. You can schedule how frequently Portfolio runs this process or you can choose to manually run it as needed. For more information about digital assets, see Digital Assets.

SD_ASSET_FULLTEXT_EXTRACT

Harvests document text, and text scanned from graphics in the asset repository.

Portfolio harvests and indexes text from the following types of files:

  • *.pdf
  • *.txt, *.rtf (plain text, rich text)
  • *.doc, *.docx (Microsoft Word)
  • *.xls, *.xlsx (Microsoft Excel)

Portfolio uses optical character recognition (OCR) scanning to harvest and index text in the following image file types:

  • *.jpg
  • *.gif
  • *.tiff
  • *.png
  • *.bmp

Note: For the best OCR scan results, text in images should be large, straight, clear, and not include graphics or background images. If possible, use images with black text and white backgrounds. If the results are coming back inaccurate, scaling the image size up can help. Currently, OCR is only supported in English.

SUGGEST_TERM_JOB

Updates the search suggestions list, if search suggestions are enabled for search profiles. You can schedule how frequently Portfolio runs this process. Because of the amount of memory needed for this task, SirsiDynix suggests that you schedule it to run after midnight and not concurrent with any of the ILS tasks. For more information about search suggestions, see Search Suggestions.

SD_CONTENT_EXTRACTOR

Harvests content from the rooms in your system, including any SirsiDynix authored content. After Portfolio runs this task, it then runs the SD_CONTENT_IMPORT and SD_CONTENT_INDEX processes in succession. You can schedule how frequently Portfolio runs this process, however, it may be most effective to run the SD_CONTENT_EXTRACTOR as needed. For more information, see Creating and editing room content.

Related topics 

Understanding search source tasks

Scheduling tasks

Fields: Search Source Tasks