Skip to content

Immuta v2020.3.0 Errata

Immuta v2020.3.0

v2020.3.0 Known Bugs

  • Aggregate performance issues with Native Presto.
  • R UDFs are broken.
  • Some databases fail when using k-anonymization with character columns.
  • SCIM: Okta can't delete synced users from Immuta.
  • Native Snowflake doesn't sync when attributes are removed from a user in an external IAM.
  • Policy updates can break Databricks data sources.
  • Scala spark submit jobs allow reading files using a distributed iterator.
  • Changing the current project as a Spark function is not available.

Immuta v2020.3.1

Immuta version 2020.3.1 was released November 30, 2020.

v2020.3.1 New Features

  • Ability to change projects using UDFs in Spark.
  • Reverse pushdown for Teradata.

v2020.3.1 Bug Fixes

  • Error calculating high cardinality column in MySQL data source when the data source contained a regex.
  • Issues using Databricks v5.5 in certain left join scenarios.
  • Issues with the fingerprint when a data source had a date outside "normal" bounds.
  • Improved high cardinality column detection query.
  • Reduced number of occasions that perform row counts.
  • Use ephemeral overrides for Databricks stats and fingerprint calls if fingerprint is kicked off manually.
  • Improved the search performance.
  • Improved Presto aggregate query performance.
  • Native Snowflake didn't sync attributes and groups for external IAMs on login.
  • Failed to delete IAM due to certificate issue.
  • Spark Submit allowed reading arbitrary files via distributed iterator.
  • Databricks:
    • Policy updates broke data sources.
    • Underlying error for SCIM call in currentUserIsAdmin was not being logged.
    • E2 was unable to create delta tables (encountering 'Only trusted classes can access package' message).
    • Other performance issues.
  • SCIM: Okta couldn't delete synced user from Immuta.
  • Fingerprint concat issues.
  • Buttons for group requests for access didn't work on the profile requests tab.
  • R UDFs were broken.
  • App Settings page save button wasn't always clickable after making changes in Chrome.
  • Encountered 500 error when generating Policy Type: Purpose Restriction Governance reports.
  • Athena was distributed and auto-enabled.
  • Couldn't create a Hive data source with hyphens.
  • Native Presto:
    • HTTP connection was not closed if a call to the Immuta service failed.
    • 500 errors occurred when viewing sources natively and unsupported policies weren't being blocked.
  • Numeric columns were not displayed in column dropdown for FPE policies (only text-based columns).

v2020.3.1 Known Bugs

  • Native Snowflake integration does not work with Snowflake ODBC driver versions > 2.21.6.

Immuta v2020.3.2

Immuta version 2020.3.2 was released December 2, 2020.

v2020.3.2 Bug Fixes

  • In Presto, conditionally masking non-character based columns resulted in a stale view.
  • In Databricks, left joins were broken on runtimes 6.4 and above.
  • The SCIM bearer token regenerated each time the configuration was saved.
  • Query Engine:
    • Incorrect Hive plan was generated.
    • Stats were not updated if row count was grabbed from explain info.

v2020.3.2 Known Bugs

  • Native Snowflake integration does not work with Snowflake ODBC driver versions > 2.21.6.
  • Query Editor becomes unresponsive when importing 10000 tables.
  • Bulk subscribing to over 200 data sources throws an error.
  • Okta can get in a bad state when it tries to provision a user that already exists.
  • On Databricks runtime 5.5, VACUUM, CONVERT TO DELTA, and FSCK REPAIR are blocked from using reflection.

Immuta v2020.3.3

Immuta version 2020.3.3 was released December 15, 2020.

v2020.3.3 Bug Fixes

  • Added "Link SQL Account" functionality to SAML IAMs. This allows support of LDAP authentication in the query engine in the case that an LDAP server and a SAML provider represent the same directory and subsequently, yield the same user ID's. This can particularly be useful for Okta users who leverage Okta's LDAP Interface.
  • Improved the frontend performance when a user has thousands of data sources.
  • Fixed bug with STRUCTS in Athena.
  • Fixed default state of the auto-subscribe checkbox after a migration.
  • Fixed issue loading attributes when an Active Directory user record contained child nodes.
  • Databricks
    • Moved table access checks/raw table resolution to PreHoc resolution.
    • Fixed error in TPC-DS query with CTE.
    • Fixed error when setting current database and querying data source in that database.
    • Reading Databricks IO cache directory directly is blocked on runtime 5.5.
  • Fixed issue that caused Okta to get in a bad state when it tried to provision a user that already existed.
  • Querying a K-anonymized data source with several duplicate columns failed over Postgres.
  • Users weren't able to create Snowflake workspaces if the HDFS handler wasn't enabled.
  • pg-boss 5.0.6 migrations were missing.
  • Users could not set their current project if they were subscribed to the project through a group.
  • SCIM didn't sync attributes if attribute mapping was added after user was already provisioned (in Azure).

v2020.3.3 Known Bugs

  • Native Snowflake integration does not work with Snowflake ODBC driver versions > 2.21.6.
  • Query Editor becomes unresponsive when importing 10000 tables.
  • Bulk subscribing to over 200 data sources throws an error.
  • SparkFunctions: Select set_current_project(0) returns a 404 until the default project is set to another project.
  • Databricks
    • On Databricks runtime 5.5, VACUUM, CONVERT TO DELTA, and FSCK REPAIR are blocked from using reflection.
    • CONVERT TO DELTA doesn't work with Scala.
    • Converting a table to Delta through Delta API is blocked in workspaces.
  • In Presto, conditionally masking non-character-based columns results in stale view.

Immuta v2020.3.4

Immuta version 2020.3.4 was released December 21, 2020.

v2020.3.4 Bug Fixes

  • Fixed issues authenticating with LDAP.

v2020.3.4 Known Bugs

  • Native Snowflake integration does not work with Snowflake ODBC driver versions > 2.21.6.
  • Query Editor becomes unresponsive when importing 10000 tables.
  • Bulk subscribing to over 200 data sources throws an error.
  • SparkFunctions: Select set_current_project(0) returns a 404 until the default project is set to another project.
  • Databricks
    • On Databricks runtime 5.5, VACUUM, CONVERT TO DELTA, and FSCK REPAIR are blocked from using reflection.
    • CONVERT TO DELTA doesn't work with Scala.
    • Converting a table to Delta through Delta API is blocked in workspaces.
  • In Presto, conditionally masking non-character-based columns results in stale view.

Immuta v2020.3.5

Immuta version 2020.3.5 was released January 6, 2021.

v2020.3.5 Bug Fixes

  • Fixed case sensitivity issue in Databricks.

v2020.3.5 Known Bugs

  • Native Snowflake integration does not work with Snowflake ODBC driver versions > 2.21.6.
  • Query Editor becomes unresponsive when importing 10000 tables.
  • Bulk subscribing to over 200 data sources throws an error.
  • SparkFunctions: Select set_current_project(0) returns a 404 until the default project is set to another project.
  • Databricks
    • On Databricks runtime 5.5, VACUUM, CONVERT TO DELTA, and FSCK REPAIR are blocked from using reflection.
    • CONVERT TO DELTA doesn't work with Scala.
    • Converting a table to Delta through Delta API is blocked in workspaces.
  • In Presto, conditionally masking non-character-based columns results in stale view.

Immuta v2020.3.6

Immuta version 2020.3.6 was released February 2, 2021.

v2020.3.6 Bug Fixes

  • Fixed auto-scaling issue in Databricks for Python and SQL notebooks.

v2020.3.6 Known Bugs

  • Native Snowflake integration does not work with Snowflake ODBC driver versions > 2.21.6.
  • Query Editor becomes unresponsive when importing 10000 tables.
  • Bulk subscribing to over 200 data sources throws an error.
  • SparkFunctions: Select set_current_project(0) returns a 404 until the default project is set to another project.
  • Databricks
    • Auto-scaling can still cause issues in R and Scala notebooks.
    • On Databricks runtime 5.5, VACUUM, CONVERT TO DELTA, and FSCK REPAIR are blocked from using reflection.
    • CONVERT TO DELTA doesn't work with Scala.
    • Converting a table to Delta through Delta API is blocked in workspaces.
  • In Presto, conditionally masking non-character-based columns results in stale view.