Veritas Data Insight Administrator's Guide

To apply metadata tags

1. Create a CSV file with the tagging information. You can create more than one CSV file with tagging information for paths.

   To assign tags to the files, ensure that the CSV file name starts with File_ (for example, File_tags.csv). Enter paths for different files with the tag name and tag values.

   To assign tags to the devices, ensure that the CSV file name starts with Device_ (for example, Device_tags.csv). Enter device with the tag name and tag values.

   To assign tags to the share/folder level, ensure that the CSV file name starts with Folder_ (for example, Folder_tags.csv). Enter share paths with the tag name and tag values.

   Note:

   i18n and special characters are not supported in tag names.

2. Save the CSV files in the data/console/tags folder in the Data Insight installation directory on the Management Server.
3. A scheduled job TagsConsumerJob parses the CSV file and creates a Tags database for each share. The job imports the tags for the paths into Data Insight. The job runs once in a day by default.

   If the job is executed manually using the configcli command, the job forcefully consumes all the CSV files under Tags folder.

   Whenever the job runs, it checks if the modified time of any of the CSV files under the Tags folder is greater than the time of the previous execution of job. If the job finds any such CSV, it processes all the CSV files under Tags folder. If the CSV file(s) have not been modified after the job was last executed, the job does not take any action.

   The job does not accept any tag name that starts with mx_ because they are reserved for Data Insight internal tags usage. Whenever the job processes the CSV, Data Insight deletes all existing tags (except tags starting with "mx_") from all files and folders and attaches new tags.

   Note:

   If a path is tagged in two different CSV files with the same tag name but with a different value, then the value in the last CSV file that is processed is applied. For device tags, multiple CSV is supported and all tags are applied. For device level tags, you can associate multiple tag values with single tag.

4. To replace existing tags, update the CSV with new tags. The scheduled job replaces existing tags with the new tags. If any paths are discarded during the last run of the job, these are logged in a database $DATADIR/console/generictags_scan_status_5.0.db that stores the discarded paths during the last run of the job.

   If any devices are discarded during the last run of the job, these are logged in a database $DATADIR/console/devicetags_scan_status_5.0.db. that stores the discarded device during the last run of the job.

5. Create a DQL report to retrieve the tags from the database.

   Here are a few example queries that you can use:

   • To fetch all paths in your storage environment along with the tags (my_tag) assigned to them.

     FROM path GET name, TAG my_tag

   • To get all paths owned by user Joe Camel tagged with the needs_assesment tag.

     FROM owner GET TAG owner.path.needs_assessment, owner.path.name 
     IF user.name="joe_camel"

   • To fetch the device in your storage environment along with the tags (my_tag) assigned to the device.

     FROM device GET name, Tag device.my_tag

   • To fetch the share in your storage environment along with the tags (my_tag) assigned to the device.

      FROM msu GET name, Tag msu.my_tag

   Note:

   The DQL report output does not return any tag if the content does not match any predefined classification tag.

6. To verify the names of tags that are stored for a share, run the idxreader command on the indexer node.

   idxreader  - i $MATRIX_DATA_DIR/indexer/default/99/99 
    - gettags all

7. To remove all the tags, delete the CSV file but if it is the last CSV file in the Tag folder, remove the content from the CSV file. Do not delete the last CSV in the folder.