标签：Es id2 write SparkSql table nodes id es
| key |
备注 |
| es.write.operation |
index(默认)添加新数据，旧数据被替换，重新索引;create添加新数据，数据存在抛出异常;update 更新现有数据，不存在抛出异常，upsert 插入及更新 |
| es.mapping.id |
_id es的doc_id |
出仓脚本exg:
CREATE TEMPORARY VIEW table_name
(
`id` STRING COMMENT 'id',
`id2` BIGINT COMMENT 'id2'
) USING org.elasticsearch.spark.sql OPTIONS (
resource 'index_version/_doc',
nodes 'xxx'
,port '9200'
,scroll_size '50'
,nodes_client_only 'true'
,nodes_data_only 'false'
,es.mapping.id 'id'
,es.write.operation 'upsert'
,es.batch.write.retry.count '6'
,es.batch.write.retry.wait '20s'
);
-- 只出仓维度
insert into table table_name
(select id,
id2
from table_name_source
where xxx
)
标签：Es,
id2,
write,
SparkSql,
table,
nodes,
id,
es
From： https://www.cnblogs.com/eiffelzero/p/17722979.html