运行查询时,从配置单元中的表中选择count(*),Map任务正在完成,而reduce任务尚未完成

时间:2018-06-22 06:10:07

标签: hadoop hive mapreduce cloudera reducers

我正在使用cloudera 5.9版本。我尝试了从Hive CLI的表查询中选择一个简单的select count(*)。这项工作永远不会返回结果。根据日志,Map任务已100%完成,但reducer任务甚至没有开始。

Hadoop job information for Stage-1: number of mappers: 1; number of reducers: 1
2018-06-22 05:56:03,019 Stage-1 map = 0%,  reduce = 0%
2018-06-22 05:56:22,789 Stage-1 map = 16%,  reduce = 0%, Cumulative CPU 11.67 sec
2018-06-22 05:56:41,472 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 05:57:41,547 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 05:58:42,532 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 05:59:43,423 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 06:00:44,304 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 06:01:45,161 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 06:02:45,938 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec
2018-06-22 06:03:46,749 Stage-1 map = 100%,  reduce = 0%, Cumulative CPU 31.45 sec

我尝试运行了几个小时,但在reduce任务上仍然没有任何进展。我检查了日志。日志中没有可用的信息。

有人可以帮助我了解这种情况下可能发生的情况吗。

谢谢。

0 个答案:

没有答案