Skip to content

Immuta v2.1.0 Release Notes

Immuta v2.1.0

Immuta version 2.1.0 was released January 25, 2018.

v2.1.0 Features

  • Spark Access Pattern
  • Join Pushdown
  • Governor Reports
  • Add license keys
  • Simplified data source creation workflow
  • Added where clause policy
  • More function pushdown (dates and casts specifically)
  • Ability to copy data sources
  • Ability to copy users from another data source
  • Upload UI for persisted data sources
  • TTL for data sources
  • Azure Blob Store handler
  • MemSQL handler
  • Okta/SAML IAM support
  • Edit and re-crawl for ingested data sources
  • Governor Notification Page
  • Sparkmagic + Kerberos in Notebooks (partial support in 2.0)
  • Git tag available via /health command
  • Specify rounding format for ingested data sources
  • Postgres configuration comes from Immuta web service instead of config file
  • Removed boaudit database
  • User notifications via email
  • More information in audit, removed unnecessary audit records
  • HA Postgres support (with background workers to refresh tokens and kinit)

v2.1.0 Migration Notes

The boaudit database will go away. The migration runs as part of the normal migration All the FDW server objects in Postgres need to be decrypted and re-encrypted to allow join pushdown. The migration runs when the Immuta web service starts. Kerberos settings moved in the configuration file from under HdfsHandler to it's own higher level block. Any existing customers using Kerberos will need to update their configuration.

Immuta v2.1.4

v2.1.4 Bug Fixes

  • Various Hive Issues:
    • Hive doesn't support integer as a data type only int so trying to do something like SELECT CAST(count(1) AS INTEGER) FROM table throws a function sequence error.
    • Hive date minimization policies don't work when the column in the comparison is of type date.
    • Hive hash() prevents minimization policies from minimizing very randomly.
    • CHARACTER_LENGTH/CHAR_LENGTH is not supported in the CDH hive distro.
    • Hive does not support GROUP BY <col_index> unless hive.groupby.orderby.position.alias is set to true (default is false).
    • JOIN pushdown breaks in some scenarios.
    • ORDER BY doesn't work when using JOIN.
    • CDH Hive doesn't support regression aggregates (they exist in Apache Hive 2.2.0 and up).
  • Unable to view files in FileSystem with eventTime in the blobId.

Immuta v2.1.3

v2.1.3 Deprecation Notes

  • Script sharing and Project Scripts have been removed from the Immuta UI.
  • The built-in Scripts environment is deprecated as of v2.1.3, and will be removed in an upcoming release. An official plugin is planned be available for users to use in any external Jupyter Notebook environment of their choice.

Version 2.1.2

Immuta version 2.1.2 was released February 20, 2018.

v2.1.2 Features

  • Netezza Connector

v2.1.2 Bug Fixes

  • User cannot specify their own extraJavaOptions when using spark-submit
  • Using date_trunc('minute', ...) in a where clause generates an error about invalid masking configuration
  • Heartbeat errors when connecting to Impala over JDBC
  • Switch to Amazon Redshift Driver
  • MySQL has certain SSL issues
  • FDW: Support substr and regexp_replace
  • Impala array columns do not work
  • SQL Query with Joins and aggregates, the aggregate does not get pushed down
  • Our ImmutaContext handles temp tables differently than the HiveContext (we resolve data sources before temp tables currently)
  • Users cannot query Impala/Hive data sources from spark if they are backed by a view

v2.1.2 Migration Notes

See v2.1.0 Migration Notes for migration from releases prior to 2.1.

Immuta v2.1.1

Immuta version 2.1.1 was released February 8, 2018.

v2.1.1 Features

No new features.

v2.1.1 Bug Fixes

  • Spark: Encountering errors when trying to view data in spark that's masked (with a join that has a masked policy or select when is masked)
  • Purposes: Sorting by Newest or Oldest throws a red toaster error
  • Projects: Unable to sort project members by Role
  • Custom policy text entry box too small
  • Add Java 8 support for Spark Context

v2.1.1 Migration Notes

See v2.1.0 Migration Notes for migration from releases prior to 2.1.