Immuta v2.2.0 Release Notes
Immuta version 2.2.0 was released June 15, 2018.
v2.2.0 New Features
- Global Policies: This allows users with Governance permission to build policies in a semantic manner which are tied to data via metadata rather than literal columns/tables.
- Tags: Support the creation of curated tags that can be added to tables and
columns to improve search and drive Global Policies. This also includes
support for ad-hoc
#tags(hashtags) which is what was called tags in prior releases.
- External Catalogs: Tags can be populated through external catalogs through this pluggable interface. Implementation exists for Apache Atlas, planned implementations with Collibra and Waterline Data.
- Now support Spark
- Additional predicate pushdown:
- Math functions
- Time functions
- Spatial functions
- Further database support:
- Amazon Athena
- Azure SQL Data Warehouse
- Custom Data Access Questionnaire: allows the creation of custom questions when requesting access to data sources.
- Additional Governance Reports related to Global Policies.
- Enhanced Identity Management
- The ability to attach authorizations to groups.
- The ability to import groups from external identity management system yet manage authorizations within Immuta or vice versa.
- Improved search through the search panel and search results screen to include searching column.
- Spark cluster awareness: Ability to run Spark jobs that can reference both local and remote HIVE and Impala tables.
- No exception policies: ability to build policies without exceptions, e.g. "Mask column x for everyone"; before only exception policies were supported, e.g. "Mask column x for everyone except members of group y".
- Support database certificates for authentication when creating Immuta data sources.
- Simplified WHERE policy creation to enable rapid creation of "Right to be Forgotten" policies.
- Notification when data source goes to unhealthy state.
- Support additional
- Allow co-mingling of differential privacy policies with purpose restriction policies.
- Setting is now available to require all data sources to have a purpose restriction (in support of GDPR).
- Exposed the Immuta policy logic endpoint (protected with an Immuta permission) to enable external software to leverage Immuta as a policy decision point.
- Support for S3 public buckets (no authentication).
- Health and policy status has been added to the data source user interface.
- Added variables to row level security WHERE policies to allow for more advanced policies.
- Improved visibility of the user interface for low contrast monitors.
v2.2.0 Major Bug Fixes
- CSD for CDH will now appropriately refresh the Immuta partition service kerberos ticket.
- Differential Privacy caching will now appropriately cache to the specified latency tolerance.
- Fixed security bug where masked values could come back unmasked if they were part of a query with both an aggregate and join.
v2.2.0 Known Bugs
- If you edit a tag, while the edit does propagate down to the data sources, it does not propagate to the global policies, you must edit those manually if they reference an edited tag.
v2.2.0 Deprecation Notices
- Masking policies now require appropriate column types and will no longer dynamically change the column type. For example, you can no longer hash a numeric type, you can generalize it (rounding) or convert to text before applying hashing.
- Joining on masked columns is no longer supported due to privacy enforcement concerns. We plan to make this configurable in the future, but will advise against doing so.
- Storefront has been broken into My Data Sources and My Projects with enhanced search capabilities through the search panel rather than having a dedicated Storefront page.
- Removed the required "organization" and "category" metadata from data sources in favor of tagging (new feature described above).
- No longer allow the generation of a policy impact report on the "Users" tab due to its lack of usability and usefulness. We plan to address with a new enhancement in the future.
- Removed the concept of child data sources. We plan to implement this differently in the future due to lack of customer demand/feedback.
Immuta version 2.2.3 was released July 17, 2018.
v2.2.3 Major Bug Fixes
- Sufficiently complicated plans cause file access errors when attempting to write a DataFrame to HDFS
Immuta version 2.2.2 was released July 17, 2018.
v2.2.2 New Features
- New database support:
- Starburst Presto
v2.2.2 Major Bug Fixes
- Unable to create table based Data Sources against Oracle 11G
- The Immuta
SparkSessionwould resolve non-Immuta tables
- Missing audit event for Acknowledge Purpose when creating a project
- Certain Hive queries push down
Immuta version 2.2.1 was released June 29, 2018.
v2.2.1 New Features
- New external catalog support:
- Waterline Data
v2.2.1 Major Bug Fixes
- Unable to create purposes in Internet Explorer
- Email notification config may create invalid "from" address
- Unable to connect to Oracle schema that use mixed-case