Hbase
Last updated
Last updated
Hbase sink connector
Output data to Hbase
zookeeper_quorum
string
yes
-
table
string
yes
-
rowkey_column
list
yes
-
family_name
config
yes
-
rowkey_delimiter
string
no
""
version_column
string
no
-
null_mode
string
no
skip
wal_write
boolean
yes
false
write_buffer_size
string
no
8 1024 1024
encoding
string
no
utf8
hbase_extra_config
string
no
-
common-options
no
-
ttl
long
no
-
The zookeeper cluster host of hbase, example: "hadoop001:2181,hadoop002:2181,hadoop003:2181"
The table name you want to write, example: "nexus"
The column name list of row keys, example: ["id", "uuid"]
The family name mapping of fields. For example the row from upstream like the following shown:
1
tyrantlucifer
27
id as the row key and other fields written to the different families, you can assign
family_name { name = "info1" age = "info2" }
this means that name
will be written to the family info1
and the age
will be written to the family info2
if you want other fields written to the same family, you can assign
family_name { all_columns = "info" }
this means that all fields will be written to the family info
The delimiter of joining multi row keys, default ""
The version column name, you can use it to assign timestamp for hbase record
The mode of writing null value, support [skip
, empty
], default skip
skip: When the field is null, connector will not write this field to hbase
empty: When the field is null, connector will write generate empty value for this field
The wal log write flag, default false
The write buffer size of hbase client, default 8 * 1024 * 1024
The encoding of string field, support [utf8
, gbk
], default utf8
The extra configuration of hbase
Hbase writes data TTL time, the default is based on the TTL set in the table, unit: milliseconds
Sink plugin common parameters, please refer to Sink Common Options for details