Drill 1.10.0-1808 Release Notes
This section provides reference information for fixes in Drill 1.10.0-1808.
The following release notes apply to the 1.10.0-1808 version of the Drill component:
Version | 1.10.0-1808 |
Release Date | September 2018 |
MapR Version Interoperability | Component Versions for Released EEPs |
Package Names | Navigate to https://package.ezmeral.hpe.com/releases/MEP/, and select your EEP and OS to view
the list of package names, for example:
|
Fixes
Drill 1.10.0-1808 provides the following fixes:
Fix Tracking Numbers | Description |
---|---|
MD-3943 | The io.dropwizard.metrics have been removed from Drill. |
MD-3716 | Queries on complex nested data in Parquet files no longer fail when selecting on all columns. |
MD-3715 | A Drillbit no longer hangs or loses connectivity. |
MD-3544 | Queries on complex nested data in Parquet files no longer fail. |
MD-3471 | Queries on views no longer intermittently return permission denied errors. |
MD-3409 | Drill returns the correct row count for queries on MapR Database JSON tables. |
MD-3214 | Drill reruns queries on Parquet data sources if they fail due to Parquet input stream read errors. |
MD-3196 | Queries with multiple concatenated items no longer hang. |
MD-2758 | Performance improvements for queries with the LIKE operator. |
MD-2528 | Drill can read Parquet columns with INT64 types. |
DRILL-6140 | Drill correctly lists operators in query profiles. |
DRILL-6028 | Drill splits the generated code in ChainedHashTable into blocks to avoid "code too large" error. |
DRILL-5978 | Hive libraries are updated to version 2.1.2-mapr-1710. |
DRILL-5972 | Improved performance for queries with filters on INFORMATION_SCHEMA. |
DRILL-5971 | Drill can read Parquet logical types INT64 and INT32. |
DRILL-5906 | The Hive version in Drill is upgraded to 1.2.0-mapr-1707 hive.version to avoid NullPointerException when querying Hive ORC tables. |
DRILL-5857 | Hive unit tests include a fix for NumberFormatException. |
DRILL-5757 | Queries with the CONVERT_TO_JSON function no longer fail when using fields that do not exist as a parameter. |
DRILL-5697 | Improved performance for queries with the LIKE operator to filter on pattern matching. |
DRILL-5420 | The ParquetAsyncPgReader no longer goes into an infinite loop during cleanup. |
DRILL-5083 | To avoid an infinite loop in the Merge Join, the status.getOutcome() method returns a failure if one of the batches has a “stop” status. |
DRILL-4120 | The Avro storage format allows implicit columns. |
DRILL-4039 | Queries no longer fail when non-ascii characters are used in string literals. |
DRILL-3250 | Drill no longer fails when comparing multi-byte characters from a Hive table. |