执行Redshift卸载操作时内存不足

Question

这是我使用的命令：

unload ('select * from SPEC_BFO.CASE_HISTORY where (INTEGRATION_ID,LAST_OPERATION_DATE) IN (SELECT INTEGRATION_ID,max(LAST_OPERATION_DATE) from SPEC_BFO.CASE_HISTORY group by INTEGRATION_ID)') to 's3://use-s3-dwnam-qa/NAM/SPEC_BFO/CASE_HISTORY/VIEW_CASE_HISTORY/VIEW_CASE_HISTORY.' iam_role 'arn:aws:iam::111111111111:role/use-redshift-dwnam-qa' delimiter '|' PARALLEL OFF header ALLOWOVERWRITE gzip;

获取此错误：

('There is a problem:', InternalError('S3 Query Exception (Fetch)\nDETAIL:  \n  -----------------------------------------------\n  error:  S3 Query Exception (Fetch)\n  code:      15001\n  context:   Request ran out of memory in the S3 query layer.\n  query:     8163346\n  location:  dory_util.cpp:1083\n  process:   asyncrequest_thread [pid=112556]\n  -----------------------------------------------\n\n',))

远程脚本调用失败，此脚本将暂停。

Answer 1

查询很可能太复杂，无法与UNLOAD操作结合。

首先尝试使用CREATE TABLE AS命令创建输出表，然后通过SELECT * UNLOAD该表。