Hadoop 数据迁移工具，Apache Sqoop 1.4.6 发布

Sqoop是一个用来将Hadoop和关系型数据库中的数据相互转移的工具，可以将一个关系型数据库（例如 ： MySQL ,Oracle ,Postgres等）中的数据导入到Hadoop的HDFS中，也可以将HDFS的数据导入到关系型数据库中。

http://www.apache.org/dyn/closer.cgi/sqoop/ 

更新内容如下：

Bug 修复

• [SQOOP-1125] - Out of memory errors when number of records to import < 0.5 * splitSize

• [SQOOP-1368] - the configuration properties are reset in HBaseImportJob

• [SQOOP-1387] - Incorrect permissions on manager.d directory can lead to NPE

• [SQOOP-1400] - Failed to import data using mysql-connector-java-5.1.17.jar

• [SQOOP-1411] - The number of tasks is not set properly in PGBulkloadExportManager

• [SQOOP-1412] - Text splitter should also consider NCHAR and NVARCHAR fields

• [SQOOP-1422] - Integration tests for Oracle connector fail as not using direct option

• [SQOOP-1423] - hcatalog export with --map-column-java fails

• [SQOOP-1429] - Fix native characters usage for SqlServer object names

• [SQOOP-1433] - Oracle direct connector fails with Oracle 12c JDBC driver

• [SQOOP-1437] - 'Text' reserved word in compliation

• [SQOOP-1472] - Use Properties.load() method to load property files under conf/manager.d

• [SQOOP-1474] - Fix usage of StringUtils.repeat method

• [SQOOP-1490] - Connectors documentation doesn't build on CentOS 5

• [SQOOP-1494] - Fix generateArvoSchema in DataDrivenImportJob

• [SQOOP-1519] - Enable HCat/HBase/Accumulo operations with OraOop connection manager

• [SQOOP-1520] - The table is null when using import into hive as parquet file and query option

• [SQOOP-1524] - Error to import into hive as textfile on hive 0.13.0

• [SQOOP-1539] - Empty partition keys and values in multiple partition key hcatalog usage should be validated

• [SQOOP-1540] - Accumulo unit tests fail with Accumulo 1.6.1 because of conflicts in libthrift libraries

• [SQOOP-1617] - MySQL fetch-size behavior changed with SQOOP-1400

• [SQOOP-1627] - Fix Hadoop100  and Hadoop20 profile

• [SQOOP-1631] - Drop confusing use of --clean-staging-table parameter from PGBulkloadManager

• [SQOOP-1663] - OraOop test cases are not logging any output

• [SQOOP-1682] - Test cases *LobAvroImportTest are failing

• [SQOOP-1684] - Use pre-existing HBase delegation token

• [SQOOP-1685] - HCatalog integration is not working on JDK8

• [SQOOP-1759] - TestIncrementalImport fails with NPE on Windows

• [SQOOP-1764] - Numeric Overflow when getting extent map

• [SQOOP-1779] - Add support for --hive-database when importing Parquet files into Hive

• [SQOOP-1826] - NPE in ImportTool.lastModifiedMerge during postgres import

• [SQOOP-1890] - Properly escape table name in generated queries

• [SQOOP-1970] - Add warning about trailing whitespace characters when using password file to User guide

• [SQOOP-2017] - Print out loaded columns and their type in verbose mode

• [SQOOP-2024] - Hive import doesn't remove target directory in hive

• [SQOOP-2055] - Run only one map task attempt during export

• [SQOOP-2057] - Skip delegation token generation flag during hbase import

• [SQOOP-2128] - HBaseImportJob should close connection from HBaseAdmin to HBase

• [SQOOP-2130] - BaseSqoopTestCase should use manager.escapeTable instead of directly hardcoding double quotes

• [SQOOP-2132] - Remove test TestMainframeImportTool.testPrintHelp

• [SQOOP-2136] - Test case SqlServerUpsertOutputFormatTest is failing

• [SQOOP-2137] - Sqoop tests and documentation refer to as-avrofile (instead of as-avrodatafile)

• [SQOOP-2145] - Default Hive home is not being set properly under certain circumstances

• [SQOOP-2164] - Enhance the Netezza Connector for Sqoop

• [SQOOP-2170] - MySQL specific tests are not properly cleaning up created tables

• [SQOOP-2191] - Provide an option automatically choose one mapper when neither primary key is defined nor split by column is provided

• [SQOOP-2254] - Failed to build release notes

• [SQOOP-2257] - Parquet target for imports with Hive overwrite option does not work

• [SQOOP-2263] - Sqoop1 has some files without a copyright header

• [SQOOP-2264] - Exclude and remove SqoopUserGuide.xml from git repository

• [SQOOP-2281] - Set overwrite on kite dataset

• [SQOOP-2282] - Add validation check for --hive-import and --append

• [SQOOP-2283] - Support usage of --exec and --password-alias

• [SQOOP-2286] - Ensure Sqoop generates valid avro column names

• [SQOOP-2290] - java.lang.ArrayIndexOutOfBoundsException thrown when malformed column mapping is provided

• [SQOOP-2294] - Change to Avro schema name breaks some use cases

• [SQOOP-2324] - Remove extra license handling for consistency

改进

• [SQOOP-1330] - Ignore blank newlines in managers.d property files

• [SQOOP-1391] - Compression codec handling

• [SQOOP-1392] - Create the temporary directory inside task working dir rather then in tmp

• [SQOOP-1421] - Automated patch script

• [SQOOP-1471] - Use Hadoop CredentialProvider API to encyrpt passwords at rest

• [SQOOP-1489] - Propagate cubrid properties to the test VM

• [SQOOP-1567] - Auto-Configure JTDS Driver From JDBCUrl

• [SQOOP-1622] - Copying from staging table should be in single transaction for pg_bulkload connector

• [SQOOP-1632] - Add support for index organized tables to direct connector

• [SQOOP-2149] - Update Kite dependency to 1.0.0

• [SQOOP-2252] - Add default to Avro Schema

新特性

• [SQOOP-1272] - Support importing mainframe sequential datasets

• [SQOOP-1309] - Expand the Sqoop to support CUBRID database.

• [SQOOP-1366] - Propose to add Parquet support

• [SQOOP-1403] - Upsert export for SQL Server

• [SQOOP-1405] - Add arg to enable SQL Server identity insert on export

• [SQOOP-1450] - Copy the avro schema file to hdfs for AVRO based import

更多更新内容请看：

https://issues.apache.org/jira/secure/ReleaseNote.jspa?version=12327469&projectId=12311320