|
| 1 | +# Kudu |
| 2 | + |
| 3 | +> Kudu数据接收器 |
| 4 | +
|
| 5 | +## 支持Kudu版本 |
| 6 | + |
| 7 | +- 1.11.1/1.12.0/1.13.0/1.14.0/1.15.0 |
| 8 | + |
| 9 | +## 支持引擎 |
| 10 | + |
| 11 | +> Spark<br/> |
| 12 | +> Flink<br/> |
| 13 | +> SeaTunnel Zeta<br/> |
| 14 | +
|
| 15 | +## 主要特性 |
| 16 | + |
| 17 | +- [ ] [exactly-once](../../concept/connector-v2-features.md) |
| 18 | +- [x] [cdc](../../concept/connector-v2-features.md) |
| 19 | +- [x] [support multiple table write](../../concept/connector-v2-features.md) |
| 20 | + |
| 21 | +## 数据类型映射 |
| 22 | + |
| 23 | +| SeaTunnel 数据类型 | Kudu 数据类型 | |
| 24 | +|---------------------|--------------------------| |
| 25 | +| BOOLEAN | BOOL | |
| 26 | +| INT | INT8<br/>INT16<br/>INT32 | |
| 27 | +| BIGINT | INT64 | |
| 28 | +| DECIMAL | DECIMAL | |
| 29 | +| FLOAT | FLOAT | |
| 30 | +| DOUBLE | DOUBLE | |
| 31 | +| STRING | STRING | |
| 32 | +| TIMESTAMP | UNIXTIME_MICROS | |
| 33 | +| BYTES | BINARY | |
| 34 | + |
| 35 | +## Sink 选项 |
| 36 | + |
| 37 | +| 名称 | 类型 | 是否必填 | 默认值 | 描述 | |
| 38 | +|-------------------------------------------|--------|----------|------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------| |
| 39 | +| kudu_masters | String | 是 | - | Kudu主地址。用“,”分隔,例如“192.168.88.110:7051”。 | |
| 40 | +| table_name | String | 是 | - | Kudu表的名字。 | |
| 41 | +| client_worker_count | Int | 否 | 2 * Runtime.getRuntime().availableProcessors() | Kudu工人数。默认值是当前cpu核数的两倍。 | |
| 42 | +| client_default_operation_timeout_ms | Long | 否 | 30000 | Kudu正常运行超时。 | |
| 43 | +| client_default_admin_operation_timeout_ms | Long | 否 | 30000 | Kudu管理员操作超时。 | |
| 44 | +| enable_kerberos | Bool | 否 | false | 启用Kerberos主体。 | |
| 45 | +| kerberos_principal | String | 否 | - | Kerberos主体。请注意,所有zeta节点都需要此文件。 | |
| 46 | +| kerberos_keytab | String | 否 | - | Kerberos密钥表。请注意,所有zeta节点都需要此文件。 | |
| 47 | +| kerberos_krb5conf | String | 否 | - | Kerberos krb5 conf.请注意,所有zeta节点都需要此文件。 | |
| 48 | +| save_mode | String | 否 | - | 存储模式,支持 `overwrite` 和 `append`. | |
| 49 | +| session_flush_mode | String | 否 | AUTO_FLUSH_SYNC | Kudu刷新模式。默认AUTO_FLUSH_SYNC。 | |
| 50 | +| batch_size | Int | 否 | 1024 | 超过此记录数的刷新最大大小(包括所有追加、追加和删除记录)将刷新数据。默认值为100 | |
| 51 | +| buffer_flush_interval | Int | 否 | 10000 | 刷新间隔期间,异步线程将刷新数据。 | |
| 52 | +| ignore_not_found | Bool | 否 | false | 如果为true,则忽略所有未找到的行。 | |
| 53 | +| ignore_not_duplicate | Bool | 否 | false | 如果为true,则忽略所有dulicate行。 | |
| 54 | +| common-options | | 否 | - |源插件常用参数,详见[Source common Options](../sink common-Options.md)。 | |
| 55 | + |
| 56 | +## 任务示例 |
| 57 | + |
| 58 | +### 简单示例: |
| 59 | + |
| 60 | +> 以下示例引用了FakeSource kudu写入表kudu_sink_table |
| 61 | +
|
| 62 | +```hocon |
| 63 | +
|
| 64 | +env { |
| 65 | + parallelism = 1 |
| 66 | + job.mode = "BATCH" |
| 67 | +} |
| 68 | + source { |
| 69 | + FakeSource { |
| 70 | + plugin_output = "kudu" |
| 71 | + schema = { |
| 72 | + fields { |
| 73 | + id = int |
| 74 | + val_bool = boolean |
| 75 | + val_int8 = tinyint |
| 76 | + val_int16 = smallint |
| 77 | + val_int32 = int |
| 78 | + val_int64 = bigint |
| 79 | + val_float = float |
| 80 | + val_double = double |
| 81 | + val_decimal = "decimal(16, 1)" |
| 82 | + val_string = string |
| 83 | + val_unixtime_micros = timestamp |
| 84 | + } |
| 85 | + } |
| 86 | + rows = [ |
| 87 | + { |
| 88 | + kind = INSERT |
| 89 | + fields = [1, true, 1, 2, 3, 4, 4.3,5.3,6.3, "NEW", "2020-02-02T02:02:02"] |
| 90 | + }, |
| 91 | + { |
| 92 | + kind = INSERT |
| 93 | + fields = [2, true, 1, 2, 3, 4, 4.3,5.3,6.3, "NEW", "2020-02-02T02:02:02"] |
| 94 | + }, |
| 95 | + { |
| 96 | + kind = INSERT |
| 97 | + fields = [3, true, 1, 2, 3, 4, 4.3,5.3,6.3, "NEW", "2020-02-02T02:02:02"] |
| 98 | + }, |
| 99 | + { |
| 100 | + kind = UPDATE_BEFORE |
| 101 | + fields = [1, true, 1, 2, 3, 4, 4.3,5.3,6.3, "NEW", "2020-02-02T02:02:02"] |
| 102 | + }, |
| 103 | + { |
| 104 | + kind = UPDATE_AFTER |
| 105 | + fields = [1, true, 2, 2, 3, 4, 4.3,5.3,6.3, "NEW", "2020-02-02T02:02:02"] |
| 106 | + }, |
| 107 | + { |
| 108 | + kind = DELETE |
| 109 | + fields = [2, true, 1, 2, 3, 4, 4.3,5.3,6.3, "NEW", "2020-02-02T02:02:02"] |
| 110 | + } |
| 111 | + ] |
| 112 | + } |
| 113 | + } |
| 114 | +
|
| 115 | +sink { |
| 116 | + kudu{ |
| 117 | + plugin_input = "kudu" |
| 118 | + kudu_masters = "kudu-master-cdc:7051" |
| 119 | + table_name = "kudu_sink_table" |
| 120 | + enable_kerberos = true |
| 121 | + kerberos_principal = "[email protected]" |
| 122 | + kerberos_keytab = "xx.keytab" |
| 123 | + } |
| 124 | +} |
| 125 | +``` |
| 126 | + |
| 127 | +### 多表 |
| 128 | + |
| 129 | +#### 示例1 |
| 130 | + |
| 131 | +```hocon |
| 132 | +env { |
| 133 | + parallelism = 1 |
| 134 | + job.mode = "STREAMING" |
| 135 | + checkpoint.interval = 5000 |
| 136 | +} |
| 137 | +
|
| 138 | +source { |
| 139 | + Mysql-CDC { |
| 140 | + base-url = "jdbc:mysql://127.0.0.1:3306/seatunnel" |
| 141 | + username = "root" |
| 142 | + password = "******" |
| 143 | + |
| 144 | + table-names = ["seatunnel.role","seatunnel.user","galileo.Bucket"] |
| 145 | + } |
| 146 | +} |
| 147 | +
|
| 148 | +transform { |
| 149 | +} |
| 150 | +
|
| 151 | +sink { |
| 152 | + kudu{ |
| 153 | + kudu_masters = "kudu-master-cdc:7051" |
| 154 | + table_name = "${database_name}_${table_name}_test" |
| 155 | + } |
| 156 | +} |
| 157 | +``` |
| 158 | + |
| 159 | +#### 示例2 |
| 160 | + |
| 161 | +```hocon |
| 162 | +env { |
| 163 | + parallelism = 1 |
| 164 | + job.mode = "BATCH" |
| 165 | +} |
| 166 | +
|
| 167 | +source { |
| 168 | + Jdbc { |
| 169 | + driver = oracle.jdbc.driver.OracleDriver |
| 170 | + url = "jdbc:oracle:thin:@localhost:1521/XE" |
| 171 | + user = testUser |
| 172 | + password = testPassword |
| 173 | +
|
| 174 | + table_list = [ |
| 175 | + { |
| 176 | + table_path = "TESTSCHEMA.TABLE_1" |
| 177 | + }, |
| 178 | + { |
| 179 | + table_path = "TESTSCHEMA.TABLE_2" |
| 180 | + } |
| 181 | + ] |
| 182 | + } |
| 183 | +} |
| 184 | +
|
| 185 | +transform { |
| 186 | +} |
| 187 | +
|
| 188 | +sink { |
| 189 | + kudu{ |
| 190 | + kudu_masters = "kudu-master-cdc:7051" |
| 191 | + table_name = "${schema_name}_${table_name}_test" |
| 192 | + } |
| 193 | +} |
| 194 | +``` |
| 195 | + |
| 196 | +## 更改日志 |
| 197 | + |
| 198 | +### 2.2.0-beta 2022-09-26 |
| 199 | + |
| 200 | +- 添加Kudu数据接收器 |
| 201 | + |
| 202 | +### 2.3.0-beta 2022-10-20 |
| 203 | + |
| 204 | +- [改进]Kudu Sink连接器支持追加销售行([2881](https://github.com/apache/seatunnel/pull/2881)) |
| 205 | + |
| 206 | +### 下个版本 |
| 207 | + |
| 208 | +- 将插件名称从"KuduSink"更改为"Kudu"[3432](https://github.com/apache/seatunnel/pull/3432) |
| 209 | + |
0 commit comments