Cloudera Employee Ana
Cloudera Employee
Posts: 93
Registered: ‎10-14-2015

[ANNOUNCE] RecordService 0.3.0 Released

We are pleased to announce the release of RecordService 0.3.0. RecordService is a core security layer that centrally enforces fine-grained access control policy, using a new abstraction layer between compute frameworks and data storage. Complementing Apache Sentry, which provides unified policy management, Cloudera now delivers unified row- and column-based security, and dynamic data masking, to every Hadoop access path. This combination of RecordService and Sentry allows security administrators to define fine-grained access control policies that are uniformly enforced for Impala, Spark, Pig, Hive, and MapReduce with no performance loss.  

RecordService 0.3.0 includes the following new features:

  • Planner Auto Discovery / HA Enhancements

  • Improved Resource Consumption using Dynamic Fetch Size

  • Pig Support via HCatalog Integration

  • CSD User Experience Improvements for Sentry Configuration


In addition, this release includes important bug fixes such as:

  • Short circuit reads not enabled (RS-114)

  • Flakiness in RecordService VM (RS-60)

  • CHAR/VARCHAR data type fixes for Spark (RS-116)

  • Enforce restriction on ZK paths (RS-3)


Links to the product and documentation are listed below:


As always, we are happy to hear your feedback. Please post your questions and comments to our mailing list or using our community forum.

Posts: 20
Registered: ‎11-26-2015

Re: [ANNOUNCE] RecordService 0.3.0 Released



I can't see anything regarding row level security. In which version will it be implemented?





Cloudera Employee Ana
Cloudera Employee
Posts: 93
Registered: ‎10-14-2015

Re: [ANNOUNCE] RecordService 0.3.0 Released

Row level security is already implemented via views (and the fact that you can use these views in Spark , MR, Pig, etc).