HBase（二） HBase Shell

Hbase Shell

Shell命令规则

ü 表名、列名用单引号或双引号包裹
ü 支持二进制、八进制、十六进制输入输出，但是必须用双引号，否则认为是文本
ü 参数用逗号分隔
ü 键值属性用Ruby散列的形式，键值间用”=>”分隔，整体被包在大括号中，键一般不需要双引号，是关键字，如 {COLUMN =>
‘c1′, TIMERANGE => [ts1, ts2], VERSIONS => 4}
ü 限制输出长度，MAXLENGTH，如get ‘t1’,’r1’, MAXLENGTH=>60
ü Shell中删除字符，ctrl+backspace （和sqlplus一样)

交互式与命令式:

交互式

进入hbase shell，可以执行单行命令，也可以利用JRuby语法写一些简单的脚本：
hbase(main):005:0> java.text.SimpleDateFormat.new("yyyy/MM/dd HH:mm:ss").parse( "2011/05/30 20:56:29").getTime();
=> 1306760189000;
hbase(main):006:0> for i in 'a'..'z' do for j in 'a'..'z' do put 't1', "row-#{i}#{j}", "f1:#{j}", "#{j}" end end;
0 row(s) in 0.0030 seconds;
……;
=> "a".."z"
require 'date';
import java.lang.Long;
import org.apache.hadoop.hbase.util.Bytes;
(Date.new(2015, 01, 01)..Date.today).each {\;
|x| put "t1", "daily", "f1:" + x.strftime("%Y%m%d"),\
Bytes.toBytes(Long.new(rand * \
4000).longValue).to_a.pack("CCCCCCCC") }

命令式

可以使用管道：
pmapp2:/opt/netwatcher/pm4h2/app/opt/hbase-0.94.9/conf> echo "status" | hbase shell
HBase Shell; enter 'help' for list of supported commands.
Type "exit" to leave the HBase Shell
Version 0.94.9, r1496217, Mon Jun 24 20:57:30 UTC 2013
status
3 servers, 0 dead, 26.0000 average load
也可以使用文件

命令全集

COMMAND GROUPS:
Group name: general
Commands: status, version, whoami
Group name: ddl
Commands: alter, alter_async, alter_status, create, describe, disable, disable_all, drop, drop_all, enable, enable_all, exists, is_disabled,
is_enabled, list, show_filters
Group name: dml
Commands: count, delete, deleteall, get, get_counter, incr, put, scan, truncate
Group name: tools
Commands: assign, balance_switch, balancer, close_region, compact, flush, hlog_roll, major_compact, move, split, unassign, zk_dump
Group name: replication
Commands: add_peer, disable_peer, enable_peer, list_peers, remove_peer, start_replication, stop_replication
Group name: snapshot
Commands: clone_snapshot, delete_snapshot, list_snapshots, restore_snapshot, snapshot
Group name: security
Commands: grant, revoke, user_permission

General命令

help

进入shell，输入help命令，会得到帮助信息
help ‘’获取单独命令的详细帮助，如help ‘status’

debug

默认shell的日志级别为ERROR，通过这个命令切换调试模式
hbase(main):002:0> debug
Debug mode is ON
hbase(main):003:0> debug
Debug mode is OFF

status

status 命令返回ClusterStatus类中的各级别信息，分三种级别simple/summary/detailed
hbase(main):008:0> status
3 servers, 0 dead, 25.3333 average load
hbase(main):009:0> status 'summary'
3 servers, 0 dead, 25.3333 average load
hbase(main):010:0> status 'simple'
3 live servers
pmapp3:15241 1437987453096
requestsPerSecond=0, numberOfOnlineRegions=31, usedHeapMB=1404, maxHeapMB=4044
pmapp2:15241 1437987453160
requestsPerSecond=0, numberOfOnlineRegions=19, usedHeapMB=959, maxHeapMB=4044
pmeam1:15241 1437987452789
requestsPerSecond=0, numberOfOnlineRegions=26, usedHeapMB=589, maxHeapMB=4044
0 dead servers Aggregate load: 0, regions: 76
hbase(main):007:0> status 'detailed'
version 0.94.9
0 regionsInTransition
master coprocessors: []
3 live servers
pmapp3:15241 1437987453096
requestsPerSecond=0, numberOfOnlineRegions=31, usedHeapMB=1399, maxHeapMB=4044
12H_20150725_0000,,1437797703299.31e4dc69db44be43274ab80f0b0355e5.
numberOfStores=1, numberOfStorefiles=1, storefileUncompressedSizeMB=28, storefileSizeMB=28, compressionRatio=1.0000,
memstoreSizeMB=0, storefileIndexSizeMB=0, readRequestsCount=0, writeRequestsCount=0, rootIndexSizeKB=23, totalStaticIndexSizeKB=16,
totalStaticBloomSizeKB=0, totalCompactingKVs=0, currentCompactedKVs=0, compactionProgressPct=NaN
……

version

返回当前版本信息、仓库版本、编译信息，即ClusterStatus.getHBaseVersion()
hbase(main):012:0> version
0.94.9, r1496217, Mon Jun 24 20:57:30 UTC 2013

exit

退出shell

DDL命令

create

即createTable()方法，可以选择多个参数，但表名和列簇名是必须的参数，其它参数还包括版本数、TTL以及预分Region建表的
key数组等，如：
hbase> create 't1', {NAME => 'f1', VERSIONS => 5}
hbase> create 't1', {NAME => 'f1'}, {NAME => 'f2'}, {NAME => 'f3'}
hbase> # The above in shorthand would be the following:
hbase> create 't1', 'f1', 'f2', 'f3'
hbase> create 't1', {NAME => 'f1', VERSIONS => 1, TTL => 2592000, BLOCKCACHE => true}
hbase> create 't1', 'f1', {SPLITS => ['10', '20', '30', '40']}
hbase> create 't1', 'f1', {SPLITS_FILE => 'splits.txt'}
hbase> # Optionally pre-split the table into NUMREGIONS, using
hbase> # SPLITALGO ("HexStringSplit", "UniformSplit" or classname)
hbase> create 't1', 'f1', {NUMREGIONS => 15, SPLITALGO => 'HexStringSplit'}
hbase(main):013:0> create 't1','r1'
0 row(s) in 1.1480 seconds

alter

修改表结构，即modifyTable()，修改前需要disable表
hbase(main):033:0> disable 't1'
0 row(s) in 2.1280 seconds
hbase(main):034:0> alter 't1', NAME => 'f1', VERSIONS => 5
Updating all regions with the new schema...
1/1 regions updated.
Done.
0 row(s) in 1.2150 seconds
hbase(main):040:0> alter 't1', NAME => 'r1', METHOD => 'delete'
Updating all regions with the new schema...
1/1 regions updated.
Done.
0 row(s) in 1.1820 seconds
hbase(main):042:0> describe 't1'
DESCRIPTION ENABLED
't1', {NAME => 'f1', DATA_BLOCK_ENCODING => 'NONE', BLOOMFILTER => 'NONE', REPLICATION_SCOPE => '0', COMPRESSION => 'NONE',
VERSIONS => '5', TTL => '21 false
47483647', MIN_VERSIONS => '0', KEEP_DELETED_CELLS => 'false', BLOCKSIZE => '65536', ENCODE_ON_DISK => 'true', IN_MEMORY =>
'false', BLOCKCACHE => 'tru
e'}
1 row(s) in 0.1510 seconds

describe

查看表机构：打印HTableDescriptor，类似oracle的desc。上面已有示例

disable

禁用表，调用disableTable()方法。上面已有示例

enable

启用表，调用enableTable()。上面已有示例

exists

查看表是否存在。tableExists()
hbase(main):043:0> exists 't1'
Table t1 does exist
0 row(s) in 0.0240 seconds

is_disabled

表是否启用isTableEnabled()
hbase(main):046:0> is_enabled 't1'
false
0 row(s) in 0.0180 seconds

list

列出所有表listTables()
hbase(main):001:0> list
TABLE
12H_20150722_0000
t1
……
55 row(s) in 0.7340 seconds

drop

删除表deleteTable()
hbase(main):002:0> drop 't1'
0 row(s) in 1.2100 seconds
hbase(main):003:0> exists 't1'
Table t1 does not exist

DML命令

put

使用java api的Put类，插入数据。必选参数是表名、RowKey、列名（包括列簇和列名）和值，可选参数包括时间戳put 't1', 'r1',
'c1', 'value', ts1
hbase(main):016:0> put 't1','row1','f1:c2','value2'
0 row(s) in 0.0060 seconds

get

Get类。单行查询，必选参数是表名和RowKey，可选参数包括列名（包括列簇和列名）、时间戳、版本数等语法：
hbase> get 't1', 'r1'
hbase> get 't1', 'r1', {TIMERANGE => [ts1, ts2]}
hbase> get 't1', 'r1', {COLUMN => 'c1'}
hbase> get 't1', 'r1', {COLUMN => ['c1', 'c2', 'c3']}
hbase> get 't1', 'r1', {COLUMN => 'c1', TIMESTAMP => ts1}
hbase> get 't1', 'r1', {COLUMN => 'c1', TIMERANGE => [ts1, ts2], VERSIONS => 4}
hbase> get 't1', 'r1', {COLUMN => 'c1', TIMESTAMP => ts1, VERSIONS => 4}
hbase> get 't1', 'r1', {FILTER => "ValueFilter(=, 'binary:abc')"}
hbase> get 't1', 'r1', 'c1'
hbase> get 't1', 'r1', 'c1', 'c2'
hbase> get 't1', 'r1', ['c1', 'c2']
hbase(main):019:0> get 't1','row1'
COLUMN CELL
f1:c2 timestamp=1438152790698,
value=value2
f1:column1 timestamp=1438151985660,
value=value1
2 row(s) in 0.0140 seconds

scan

多行查询，依赖Scan类。必选参数是表名，可选参数包括列名（包括列簇和列名）、起止key、Filter，语法：
hbase> scan '.META.'
hbase> scan '.META.', {COLUMNS => 'info:regioninfo'}
hbase> scan 't1', {COLUMNS => ['c1', 'c2'], LIMIT => 10, STARTROW => 'xyz'}
hbase> scan 't1', {COLUMNS => 'c1', TIMERANGE => [1303668804, 1303668904]}
hbase> scan 't1', {FILTER => "(PrefixFilter ('row2') AND (QualifierFilter (>=, 'binary:xyz'))) AND (TimestampsFilter ( 123, 456))"}
hbase> scan 't1', {FILTER => org.apache.hadoop.hbase.filter.ColumnPaginationFilter.new(1, 0)}
hbase(main):021:0> scan 't1'
ROW COLUMN+CELL
row1 column=f1:c2, timestamp=1438152790698,
value=value2
row1 column=f1:column1, timestamp=1438151985660,
value=value1
row2 column=f1:c2, timestamp=1438152770375, value=value2

delete

单列删除，依赖Delete类。必选参数table/row/column，可选参数时间戳
语法：delete 't1', 'r1', 'c1', ts1
hbase(main):042:0> delete 't1','row1','f1:column1'
0 row(s) in 0.0170 seconds

deleteall

可以删除一个列族或一列，也依赖Delete类。必选参数表、行，可选参数列、时间戳
hbase> deleteall 't1', 'r1'
hbase> deleteall 't1', 'r1', 'c1'
hbase> deleteall 't1', 'r1', 'c1', ts1

truncate

清理表中数据，相当于disable drop create在同一个模式下的顺序使用
hbase(main):045:0> truncate 't1'
Truncating 't1' table (it may take a while):
- Disabling table...
- Dropping table...
- Creating table...
0 row(s) in 4.4000 seconds

count

统计一个表的行数，内部使用了Scan
hbase(main):047:0> count 't1'
0 row(s) in 0.0550 seconds

incr

操作计数器，默认加一，可以操作正数、负数、零。依赖Increment类。下面的命令会创建新的r1行和f1:1列，不用实现创建好。
hbase(main):051:0> incr 't1', 'r1', 'f1:1'
COUNTER VALUE = 1
hbase(main):052:0> incr 't1', 'r1', 'f1:1'
COUNTER VALUE = 2
hbase(main):055:0> incr 't1', 'r1', 'f1:1' ,5
COUNTER VALUE = 7

get_counter

返回一个计数器的值，和get类似，但是把计数器的值转换为可读的形式，也依赖Get类
hbase(main):004:0> get_counter 't1','r1','f1:1'
COUNTER VALUE = 7
hbase(main):005:0> get 't1','r1','f1:1'
COLUMN CELL
f1:1 timestamp=1438153769249, value=\x00\x00\x00\x00\x00\x00\x00\x07
1 row(s) in 0.0350 seconds

Tool命令

assign

java api 的assign() 方法，分配一个region到一台服务器中。

balance_switch

balanceSwitch() 方法，切换负载均衡开关

balancer

balancer() ，启动负载均衡

close_region

closeRegion() ，关闭一个region

compact

compact()，开启某个region或一张表的异步合并

flush

flush()，开启某个region或一张表的异步flush

major_compact

majorCompact()，开启某个region或一张表的major合并

move

move()，移动一个region到不同的服务器中

split

split()，拆分一个region或一个表

unassign

unassign()，下线一个region

zk_dump

把zookeeper的信息dump到hbase里，这是内部类提供的特殊功能，Hbase Master 的Web UI也有这个功能