红联Linux门户
Linux帮助

Apache Hive 0.13.1发布,数据仓库平台

发布时间:2014-06-08 09:19:29来源:红联作者:empast
Apache Hive 0.13.1 发布,Hive是一个基于Hadoop的数据仓库平台。通过hive,我们可以方便地进行ETL的工作。hive定义了一个类似于SQL的查询语言:HQL,能 够将用户编写的QL转化为相应的Mapreduce程序基于Hadoop执行。

该版本主要是 bug 修复,包括:

[HIVE-4576] - templeton.hive.properties does not allow values with commas

[HIVE-6783] - Incompatible schema for maps between parquet-hive and parquet-pig

[HIVE-6826] - Hive-tez has issues when different partitions work off of different input types

[HIVE-6828] - Hive tez bucket map join conversion interferes with map join conversion

[HIVE-6862] - add DB schema DDL and upgrade 12to13 scripts for MS SQL Server

[HIVE-6877] - TestOrcRawRecordMerger is deleting test.tmp.dir

[HIVE-6883] - Dynamic partitioning optimization does not honor sort order or order by

[HIVE-6898] - Functions in hive are failing with java.lang.ClassNotFoundException on Tez

[HIVE-6919] - hive sql std auth select query fails on partitioned tables

[HIVE-6921] - index creation fails with sql std auth turned on

[HIVE-6945] - issues with dropping partitions on Oracle

[HIVE-6952] - Hive 0.13 HiveOutputFormat breaks backwards compatibility

[HIVE-6955] - ExprNodeColDesc isSame doesn't account for tabAlias: this affects trait Propagation in Joins

[HIVE-6957] - SQL authorization does not work with HS2 binary mode and Kerberos auth

[HIVE-6961] - Drop partitions treats partition columns as strings

[HIVE-6985] - sql std auth - privileges grants to public role not being honored

[HIVE-6996] - FS based stats broken with indexed tables

[HIVE-7001] - fs.permissions.umask-mode is getting unset when Session is started

[HIVE-7067] - Min() and Max() on Timestamp and Date columns for ORC returns wrong results

[HIVE-7116] - HDFS FileSystem object cache causes permission issues in creating tmp directories

[HIVE-7130] - schematool is broken for minor version upgrades (eg 0.13.x)

[HIVE-7165] - Fix hive-default.xml.template errors & omissions

项目主页:http://hive.apache.org/

下载地址:http://www.apache.org/dyn/closer.cgi/hive/

来自:开源中国社区
文章评论

共有 0 条评论