View HBase table partitioning
scan 'hbase:meta'
t1,,1487821235337.353a385f28af52ed47e675f18242bbf8. column=info:regioninfo, timestamp=1487821236518, value={ENCODED => 353a385f28af52ed47e675f18242bbf8, NAME => 't1,,1487821235337.353a385f28af52ed47e675f18242bbf8.', START
KEY => '', ENDKEY => ''}
t1,,1487821235337.353a385f28af52ed47e675f18242bbf8. column=info:seqnumDuringOpen, timestamp=1488359119701, value=\x00\x00\x00\x00\x00\x00\x00\x14
t1,,1487821235337.353a385f28af52ed47e675f18242bbf8. column=info:server, timestamp=1488359119701, value=centos2:16020
t1,,1487821235337.353a385f28af52ed47e675f18242bbf8. column=info:serverstartcode, timestamp=1488359119701, value=1488359111761
full region name
t1, 1487821235337.53a385f28af52ed47e675f18242bbf8.
Tablespace: table name, start rowId, timestamp, hashcode(MD5)
region slice
grammar
split 'tableName'
split 'namespace:tableName'
split 'regionName' # format: 'tableName,startKey,id'
split 'tableName', 'splitKey'
split 'regionName', 'splitKey'
Where 'splitKey' indicates which line to start splitting from
reqion slice merge
hbase> merge_region 'ENCODED_REGIONNAME', 'ENCODED_REGIONNAME'
hbase> merge_region 'ENCODED_REGIONNAME', 'ENCODED_REGIONNAME', true //强制合并
Where 'ENCODED_REGIONNAME' represents the hashcode in the region Id, such as " 353a385f28af52ed47e675f18242bbf8 " in the case of this article
Note: Only regions adjacent to startkey and endkey can be merged, and non-adjacent regions can only be forced to merge
Hbase data query process :
1. The client first connects to zookeeper and looks up the location of the table hbase:meta.
2. Query the table hbase:meta, find a specific region according to the start key and end key of each region, and obtain the server where the region is located, such as " column=info:server, timestamp=1488359119701, value=centos2:16020 "
3. Go to the specified server to find the region.
4. The information in the hbase:meta table will be cached to facilitate the next lookup.