DROP TABLE filtered_online_march_customers;
            --creating bucketed table with customer id
            CREATE TABLE IF NOT EXISTS filtered_online_march_customers(
              customer_id           string,
              order_id              string
            )
            CLUSTERED BY(customer_id) INTO 32 BUCKETS;

--populating the table
set hive.enforce.bucketing = true;
FROM filtered_march_online_transactions
INSERT OVERWRITE TABLE filtered_online_march_customers
SELECT
  *

我创建了这个表，它按客户id进行聚类。但是，当我实际尝试使用bucket时，它不起作用。

CREATE TABLE randomized_filtered_march_customers
AS
SELECT
  *
FROM
 filtered_online_march_customers
TABLESAMPLE(BUCKET 1 OUT OF 32 ON customer_id)

我在pathtopartitioninfo:[maprfs:/hive/v0k0020.db/filtered\u online\u march\u customers/000000\u 0]org.apache.hadoop.hive.io.hivefileformatutils.getpartitiondescfrompathRecursive（hivefileformatutils）中发现错误：cannot find dir=maprfs:///hive/v0k0020.db/filtered\u online\u march\u customers/000000\u 0。java:344)在org.apache.hadoop.hive.ql.io.hivefileformatutils.getPartitionDescrFromPathRecursive（hivefileformatutils）。java:306)位于org.apache.hadoop.hive.ql.io.combinehiveinputformat$combinehiveinputsplit.（combinehiveinputformat。java:108)位于org.apache.hadoop.hive.ql.io.combinehiveinputformat.getsplits（combinehiveinputformat）。java:455)在org.apache.hadoop.mapred.jobclient.writeoldsplits（jobclient。java:1098)在org.apache.hadoop.mapred.jobclient.writeslits（jobclient。java:1090)访问org.apache.hadoop.mapred.jobclient.access$500（jobclient。java:176)在org.apache.hadoop.mapred.jobclient$2.run（jobclient。java:931)在org.apache.hadoop.mapred.jobclient$2.run（jobclient。java:882)在javax.security.auth.subject.doas（主题）中的java.security.accesscontroller.doprivileged（本机方法）。java:415)在org.apache.hadoop.security.usergroupinformation.doas（usergroupinformation。java:1595)在org.apache.hadoop.mapred.jobclient.submitjobinternal（jobclient。java:882)在org.apache.hadoop.mapred.jobclient.submitjob（jobclient。java:856)在org.apache.hadoop.hive.ql.exec.mr.execdriver.execute（execdriver。java:420)位于org.apache.hadoop.hive.ql.exec.mr.mapredtask.execute（mapredtask）。java:136)在org.apache.hadoop.hive.ql.exec.task.executetask（任务。java:153)位于org.apache.hadoop.hive.ql.exec.taskrunner.runsequential（taskrunner）。java:85)在org.apache.hadoop.hive.ql.driver.launchtask（驱动程序。java:1503)在org.apache.hadoop.hive.ql.driver.execute（driver。java:1270)在org.apache.hadoop.hive.ql.driver.runinternal（driver。java:1088)在org.apache.hadoop.hive.ql.driver.run（driver。java:911)在org.apache.hadoop.hive.ql.driver.run（driver。java:901)在org.apache.hadoop.hive.cli.clidriver.processlocalcmd（clidriver。java:268)在org.apache.hadoop.hive.cli.clidriver.processcmd（clidriver。java:220)在org.apache.hadoop.hive.cli.clidriver.processline（clidriver。java:423)在org.apache.hadoop.hive.cli.clidriver.processline（clidriver。java:359)在org.apache.hadoop.hive.cli.clidriver.processreader（clidriver。java:456)在org.apache.hadoop.hive.cli.clidriver.processfile（clidriver。java:466)在org.apache.hadoop.hive.cli.clidriver.executedriver（clidriver。java:748)在org.apache.hadoop.hive.cli.clidriver.run（clidriver。java:686)位于org.apache.hadoop.hive.cli.clidriver.main（clidriver。java:625)在sun.reflect.nativemethodaccessorimpl.invoke0（本机方法）位于sun.reflect.nativemethodaccessorimpl.invoke（nativemethodaccessorimpl）。java:57)在sun.reflect.delegatingmethodaccessorimpl.invoke（delegatingmethodaccessorimpl。java:43)在java.lang.reflect.method.invoke（方法。java:606)在org.apache.hadoop.util.runjar.run（runjar。java:221)在org.apache.hadoop.util.runjar.main（runjar。java:136)作业提交失败，出现异常“java.io.ioexception（在路径分区信息：[maprfs:/hive/v0k0020.db/filtered\u online\u march\u customers/000000\u 0]中找不到dir=maprfs:///hive/v0k0020.db/filtered\u online\u march\u customers/000000\u 0]）”
如果我把查询改为

CREATE TABLE randomized_filtered_march_customers
    AS
    SELECT
      *
    FROM
     filtered_online_march_customers
    TABLESAMPLE(BUCKET 1 OUT OF 32 ON rand())

很好用。你知道怎么解决吗？