HBase——Snapshot（快照）

最新推荐文章于 2024-05-06 11:18:51 发布

转载最新推荐文章于 2024-05-06 11:18:51 发布 · 2.7k 阅读

6 ·

CC 4.0 BY-SA版权

原文链接：https://blog.youkuaiyun.com/weixin_42073408/article/details/107937246

文章标签：

#hbase

Hadoop 专栏收录该内容

5 篇文章

订阅专栏

本文介绍HBase中使用快照进行数据备份的方法，包括快照的创建、删除、克隆新表、表恢复及数据导出等功能，并提供具体操作步骤。

一. 基于Hbase snapshot数据快速备份方法
1)创建快照(查看快照->查看快照snapshot命令相关参数->创建快照—>查看快照)

查看snapshot

hbase(main):002:0> list_snapshots
SNAPSHOT                                                              TABLE + CREATION TIME
0 row(s) in 0.0290 seconds

查看help:

hbase(main):003:0> snapshot

ERROR: wrong number of arguments (0 for 2)
Here is some help for this command:
Take a snapshot of specified table. Examples:

  hbase> snapshot 'sourceTable', 'snapshotName'
  hbase> snapshot 'namespace:sourceTable', 'snapshotName', {SKIP_FLUSH => true}

创建snapshot 例子：

hbase(main):004:0> snapshot 'emp','emp_snapshot'

 0 row(s) in 0.3730 seconds

hbase(main):005:0> list_snapshots
SNAPSHOT                                                              TABLE + CREATION TIME
emp_snapshot                                                         emp (Wed May 16 09:44:53 +0800 2018)
1 row(s) in 0.0190 seconds

=> ["emp_snapshot"]

2)删除并查看快照

hbase(main):006:0> delete_snapshot 'emp_snapshot'
0 row(s) in 0.0390 seconds

hbase(main):007:0> list_snapshots
SNAPSHOT                                                              TABLE + CREATION TIME
0 row(s) in 0.0040 seconds

=> []

3)基于快照，clone一个新表

hbase(main):011:0> clone_snapshot 'emp_snapshot','new_emp'
0 row(s) in 0.5290 seconds

hbase(main):013:0> scan 'new_emp'
ROW                                                                   COLUMN+CELL
1                                                                    column=personal data:city, timestamp=1526269334560, value=hyderabad
1                                                                    column=personal data:name, timestamp=1526269326929, value=raju
1                                                                    column=professional data:designation, timestamp=1526269345044, value=manager
1                                                                    column=professional data:salary, timestamp=1526269352605, value=50000
1 row(s) in 0.1050 seconds

hbase(main):014:0> desc 'new_emp'
Table new_emp is ENABLED
new_emp
COLUMN FAMILIES DESCRIPTION
{NAME => 'personal data', BLOOMFILTER => 'ROW', VERSIONS => '1', IN_MEMORY => 'false', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOCK_ENCODING => 'NONE', TTL => 'FOREVER', COMPRESSION => 'NONE', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE => '65536', REPLICATION_SCOPE
=> '0'}
{NAME => 'professional data', BLOOMFILTER => 'ROW', VERSIONS => '1', IN_MEMORY => 'false', KEEP_DELETED_CELLS => 'FALSE', DATA_BLOCK_ENCODING => 'NONE', TTL => 'FOREVER', COMPRESSION => 'NONE', MIN_VERSIONS => '0', BLOCKCACHE => 'true', BLOCKSIZE => '65536', REPLICATION_S
COPE => '0'}
2 row(s) in 0.0370 seconds

4)基于快照恢复表(原hbase表emp需要删除)

hbase(main):027:0> list
TABLE
new_emp
t1
test
3 row(s) in 0.0130 seconds


=> ["new_emp", "t1", "test"]

hbase(main):028:0> list_snapshots
SNAPSHOT                                                              TABLE + CREATION TIME
emp_snapshot                                                         emp (Wed May 16 09:45:25 +0800 2018)
1 row(s) in 0.0130 seconds


=> ["emp_snapshot"]
hbase(main):029:0> restore_snapshot 'emp_snapshot'
0 row(s) in 0.3700 seconds

hbase(main):030:0> list
TABLE
emp
new_emp
t1
test
4 row(s) in 0.0240 seconds


=> ["emp", "new_emp", "t1", "test"]

5)基于快照将数据导出到另外一个集群中的本地文件中

利用mapreduce job将emp_snapshot这个snapshot 导出到本地目录usr/hdp/2.5.3.0-37/hbase/hbasedata

hbase org.apache.hadoop.hbase.snapshot.ExportSnapshot -snapshot 'emp_snapshot' -copy-to file:usr/hdp/2.5.3.0-37/hbase/hbasedata -mappers 16

6)基于快照将数据导出到另外一个集群中的hdfs上

hbase org.apache.hadoop.hbase.snapshot.ExportSnapshot -snapshot 'emp_snapshot' -copy-to hdfs:///hbase/bak_emp_snapshot -mappers 16