Spark job failed during runtime. Please check stacktrace for the root cause.报错解释

select
    mid_id,
    dt,
    date_diff,
    count(*)over(partition by mid_id,date_diff) diff_count
from (
     select mid_id,
            dt,
            date_add(dt, -rk) date_diff
     from (
              select mid_id,
                     dt,
                     rank() over (partition by mid_id order by dt) rk
              from (
                       select mid_id,
                              dt
                       from dws_uv_detail_daycount
                       where dt >= date_add('2020-06-20', -6)
                       group by mid_id, dt--日期相关。一个设备在一个日期使用。
                   ) t1
          ) t2
)t3

求最近七天内连续三天活跃用户数时，报错。信息如下：
Error while processing statement: FAILED: Execution Error, return code 3 from org.apache.hadoop.hive.ql.exec.spark.SparkTask. Spark job failed during runtime. Please check stacktrace for the root cause.
原因如下：
①我们把hive的计算引擎替换为spark，有可能是内存不足，。
②有可能是源码不兼容。
如果在计算引擎为MapReduce的hive上运行，是不会报错的。

希望大家可以关注下公众号，会定期分享自己从业经历、技术积累及踩坑经验，支持一下，鞠躬感谢~

关注公众号，回复“资料全集”，不定期最新大数据业内资讯。