{
"job": {
"setting": {
"speed": {
"channel": 3,
"bytes": 0
},
"errorLimit": {
"record": 10000,
"percentage": 100
}
},
"content": [
{
"reader": {
"name": "mysqlreader",
"parameter": {
"username": "root",
"password": "111111",
"column": [
"v",
"id"
],
"connection": [
{
"table": [
"tt"
],
"jdbcUrl": [
"jdbc:mysql://rdos1:3306/hyf?useCursorFetch=true"
]
}
],
"splitPk": "id"
}
},
"writer": {
"name": "carbondatawriter",
"parameter": {
"path": "hdfs://ns1/user/hive/warehouse/carbon.store1/sb/sb500",
"hadoopConfig": {
"dfs.ha.namenodes.ns1": "nn1,nn2",
"dfs.namenode.rpc-address.ns1.nn2": "rdos2:9000",
"dfs.client.failover.proxy.provider.ns1": "org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider",
"dfs.namenode.rpc-address.ns1.nn1": "rdos1:9000",
"dfs.nameservices": "ns1"
},
"defaultFS": "hdfs://ns1",
"table": "sb500",
"database": "sb",
"writeMode": "overwrite",
"column": [
"a",
"b"
],
"batchSize": 1
}
}
}
]
}
}
-
name
- 描述:插件名,此处只能填carbondatawriter,否则Flinkx将无法正常加载该插件包。
- 必选:是
- 默认值:无
-
path
-
描述:carbondata表的存储路径
-
必选:是
-
默认值:无
-
-
table
-
描述:carbondata表名
-
必选:否
-
默认值:无
-
-
database
-
描述:carbondata库名
-
必选:否
-
默认值:无
-
-
filter
-
描述:简单过滤器,目前只支持单条件的简单过滤,形式为 col op value
col为列名;
op为关系运算符,包括=,>,>=,<,<=;
value为字面值,如1234, "ssss" -
必选:否
-
默认值:无
-
-
column
-
描述:所配置的表中需要同步的字段名列表
-
必选:是
-
默认值:无
-
-
hadoopConfig
- 描述:hadoopConfig里可以配置与Hadoop相关的一些高级参数,比如HA的配置。
{ "hadoopConfig": { "dfs.ha.namenodes.ns1": "nn1,nn2", "dfs.namenode.rpc-address.ns1.nn2": "rdos2:9000", "dfs.client.failover.proxy.provider.ns1": "org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider", "dfs.namenode.rpc-address.ns1.nn1": "rdos1:9000", "dfs.nameservices": "ns1", "fs.defaultFS": "hdfs://ns1" } }
- 描述:hadoopConfig里可以配置与Hadoop相关的一些高级参数,比如HA的配置。
-
defaultFS
-
描述:Hadoop hdfs文件系统namenode节点地址。
-
必选:是
-
默认值:无
-
支持如下数据类型
- SMALLINT
- INT/INTEGER
- BIGINT
- DOUBLE
- DECIMAL
- FLOAT
- BYTE
- BOOLEAN
- STRING
- CHAR
- VARCHAR
- DATE
- TIMESTAMP
不支持如下数据类型
- arrays: ARRAY<data_type>
- structs: STRUCT<col_name : data_type COMMENT col_comment, ...>
- maps: MAP<primitive_type, data_type>