Kudu 表数据迁移

迈不过友情╰ 2023-02-22 03:59 14阅读 0赞

###### 使用 Kudu Command Line Tools 将表数据复制到另一个表 ######

这两个表可在同一个集群中，也可在不同集群。但是这两个表必须具有相同的表模式，可以具有不同的分区模式。该工具可以使用与源表相同的表和分区模式创建新表。

**用法:**  
`kudu table copy <master_addresses> <table_name> <dest_master_addresses> [-nocreate_table] [-dst_table=<table>] [-num_threads=<threads>] [-predicates=<predicates>] [-tablets=<tablets>] [-write_type=<type>]`

参数:

<table> 
 <thead> 
  <tr> 
   <th align="center">名称</th> 
   <th align="center">描述</th> 
   <th align="center">类型</th> 
   <th align="center">默认</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td align="center">master_addresses</td> 
   <td align="center">以逗号分隔的源 Kudu 主机地址列表，其中每个地址的形式是"hostname:port"，也可以使用集群名</td> 
   <td align="center">string</td> 
   <td align="center">none</td> 
  </tr> 
  <tr> 
   <td align="center">table_name</td> 
   <td align="center">源表名称</td> 
   <td align="center">string</td> 
   <td align="center">none</td> 
  </tr> 
  <tr> 
   <td align="center">dest_master_addresses</td> 
   <td align="center">以逗号分隔的目标 Kudu 主机地址列表，其中每个地址的形式是"hostname:port"，也可以使用集群名</td> 
   <td align="center">string</td> 
   <td align="center">none</td> 
  </tr> 
  <tr> 
   <td align="center">create_table (optional)</td> 
   <td align="center">如果目标表不存在,是否创建目标表。</td> 
   <td align="center">bool</td> 
   <td align="center">true</td> 
  </tr> 
  <tr> 
   <td align="center">dst_table (optional)</td> 
   <td align="center">将数据复制到的目标表的名称。如果为空字符串，则使用与源表相同的名称。</td> 
   <td align="center">string</td> 
   <td align="center">none</td> 
  </tr> 
  <tr> 
   <td align="center">num_threads (optional)</td> 
   <td align="center">要运行的线程数。每个线程运行自己的KuduSession。</td> 
   <td align="center">int32</td> 
   <td align="center">2</td> 
  </tr> 
  <tr> 
   <td align="center">predicates (optional)</td> 
   <td align="center">支持三种类型的谓词，包括"Comparison", “InList” and “IsNull”。</td> 
   <td align="center">string</td> 
   <td align="center">none</td> 
  </tr> 
  <tr> 
   <td align="center">tablets (optional)</td> 
   <td align="center">如果没有指定要检查的 Tablets (以逗号分隔的id列表)，则检查所有的 Tablets。</td> 
   <td align="center">string</td> 
   <td align="center">none</td> 
  </tr> 
  <tr> 
   <td align="center">write_type (optional)</td> 
   <td align="center">如何将数据复制到目标表。“insert”、"upsert"或空字符串。如果字符串为空，则不会复制数据(create_table为true时很有用)。</td> 
   <td align="center">string</td> 
   <td align="center">insert</td> 
  </tr> 
 </tbody> 
</table>

**使用示例:**

`Configuration error: Could not connect to the cluster: no leader master found. Client configured with 1 master(s) (192.168.1.101:7051) but cluster indicates it expects 3 master(s) (cdh01:7051,cdh02:7051,cdh03:7051)`  
**注意:** master\_addresses 需写全 master 地址。

[root@node01 ~]# kudu table copy node01:7051,node02:7051,node03:7051 my_database.my_table cdh01:7051,cdh02:7051,cdh03:7051 -nocreate_table -dst_table=my_database.my_table -num_threads=8 -write_type=upsert
    I0701 13:56:29.095507  1551 table_scanner.cc:532] Scanned count: 1302487
    I0701 13:56:34.099725  1551 table_scanner.cc:532] Scanned count: 2425488
    I0701 13:56:39.105008  1551 table_scanner.cc:532] Scanned count: 3468528
    I0701 13:56:44.109151  1551 table_scanner.cc:532] Scanned count: 4453114
    I0701 13:56:49.113626  1551 table_scanner.cc:532] Scanned count: 5425016
    I0701 13:56:54.117858  1551 table_scanner.cc:532] Scanned count: 6464445
    ......
    I0701 14:55:03.737257  1551 table_scanner.cc:532] Scanned count: 535158673
    I0701 14:55:08.741680  1551 table_scanner.cc:532] Scanned count: 535302695
    I0701 14:55:13.745841  1551 table_scanner.cc:532] Scanned count: 535538444
    I0701 14:55:18.750126  1551 table_scanner.cc:532] Scanned count: 535752970
    I0701 14:55:23.754225  1551 table_scanner.cc:532] Scanned count: 535949697
    I0701 14:55:28.758299  1551 table_scanner.cc:532] Scanned count: 536157109
    I0701 14:55:33.762697  1551 table_scanner.cc:532] Scanned count: 536302601
    I0701 14:55:38.767024  1551 table_scanner.cc:532] Scanned count: 536385095
    I0701 14:55:43.771351  1551 table_scanner.cc:532] Scanned count: 536455367
    T 76ccd465c1a345a6808a5731da858044 scanned count 53658586 cost 1300.65 seconds
    Total count 536573786 cost 3563.88 seconds

参考:  
https://kudu.apache.org/docs/command\_line\_tools\_reference.html\#remote\_replica  
https://kudu.apache.org/docs/hive\_metastore.html