Tamr Documentation

v2019.020 Notes

Tamr release notes.

What's New

  • Improved the user experience by only showing those datasets that can be added to the project based on project's access control policy. If your project's permissions do not include access to some datasets as resources, these datasets no longer show up as candidates for adding to the project.
  • You can use % (remainder) arithmetic operator in transformations, and also use the following added functions: PMOD(), MATH.SIN(), MATH.COS(), MATH.ASIN(), MATH.ACOS().
  • If you haven't upgraded your Tamr projects to the new hash() function introduced in v.2019.010, you can use a script that locates all instances of the new hash() and allows you to go back to the legacy.hash(), until you are ready to upgrade. This measure is temporary and you are strongly encouraged to upgrade your scripts to start using the new hash() function.

Access Control

Improved the behavior of project configuration to only display those datasets that are allowed to be added to a project. In previous releases, if a project's policy did not include access to a dataset as a resource, the datasets still appeared in the list even though you could not add them.

Golden Records

Fixed an issue in custom conditions editing for golden records where syntax errors would persist when users attempted to write custom conditions on golden records and switched between condition types.


  • Added the PMOD() function and the % (remainder) arithmetic operator.
  • Added a scroll bar to the transformations editor and a maximum number of script lines that you can view without expanding the editor window (300). To see the entire log script, maximize the editor window. Use CTRL+F to search for items in the script even if they don't display in the smaller version of the editor (before you expand it to show the entire script). Previously, it was not possible to scroll through long scripts and formulas, making it hard for you to navigate and locate the place for editing on the side panel.
  • Added a tools.jar script that allows you to convert all hash() functions back to legacy.hash() during the grace period while you upgrade to using the hash() function. In v.2019.010.0 Tamr introduced a new implementation for the hash() function in transformations. For information, see Release Notes for Tamr v.2019.010. If your existing projects still rely on transformations with the old hash() behavior, use this grace period and rely on this script to help you switch between hash() and legacy.hash() in all your projects. We strongly encourage you to upgrade your transformation scripts to use the new hash() function. Download the file tools.jar from the same location as the Tamr package. Run toools.jar --h to learn how to use it. For example, to convert instances of hash() to legacy.hash(), run: java -jar tools.jar function-replacer --host http://<tamr-port>:9100 -u admin -p --new “legacy.hash” --old hash. Contact Tamr Support if you need additional guidance. To contact Tamr Support, use the Support icon located at the lower right corner of this page.

Configuration Changes

  • Upgraded ZooKeeper to version 3.4.14. This change does not require any actions during upgrades. The upgrading process in Tamr takes care of this upgrade to one of Tamr dependency packages.
  • Added a new YARN configuration parameter to the Tamr configuration, named TAMR_YARN_SCHEDULER_CAPACITY_MAXIMUM_AM_RESOURCE_PERCENT. See YARN Cluster Manager Jobs.
  • Fixed an issue that allows backing up of Elasticsearch artifacts to GCS. This was enabled by the upgrade to ElasticSearch 6.8.2 that took place in the previous release. To configure a Tamr backup and restore with GCS, set these two properties:
    • TAMR_UNIFY_BACKUP_URI is the path to backup and restore directory in this format gs://<bucket>/<path/to/backup>, such as gs://backup-bucket/backup1
    • TAMR_GOOGLE_APPLICATION_CREDENTIALS is an absolute local path to the service account credentials JSON file, such as /tmp/gcs/creds.json.

Other Improvements

  • Performance improvements. Made pair generation process faster for large datasets.
  • User interface improvements. Fixed an issue where a failed job icon was misplaced on the navigation bar.
  • Fixed the default sorting of records on the clusters page.
  • Updated the documentation file (--help) of the Tamr administrative utility, admin-unify.sh.
  • Upgrade improvements. Fixed a number of internally-reported issues with the Tamr upgrade scripts.

Support Tickets

Fixed the following support tickets.

  • Fixed a bug where you could not select an attribute from the drop-down list of conditions in golden records when its name was also part of another attribute’s name. The fix allows you to select any attribute in conditions and rules for golden records.
  • Fixed an issue where syntax errors would persist when users attempted to write custom conditions on golden records and switched between condition types.
  • Fixed an issue with concurrent jobs failing due to inability to configure the YARN memory limit. Added a new Tamr configuration parameter for YARN, TAMR_YARN_SCHEDULER_CAPACITY_MAXIMUM_AM_RESOURCE_PERCENT.

Known Issues

The following known issues exist in this release.

  • Column resizing on the Users page does not behave as expected.
  • Job submission for chained projects may not appear immediately on the Jobs page after choosing Submit. Submit is not disabled in this case.
  • Cannot update golden records with a geospatial data type present.
  • A warning displays in the browser console for any page or details panel that contains geospatial record types in mastering projects.
  • Error when attempting sorting by a geospatial type attribute.
  • Most frequent values are not showing up for a geospatial type field on a profiled attribute on the Schema Mapping page.
  • The schema mapping suggestion fails if a geospatial record type attribute is present.
  • Export is failing on a dataset with geospatial data.


See upgrading page for instructions.

v2019.020 Notes

Tamr release notes.

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.