logo
publist
写文章

简介

该用户还未填写简介

擅长的技术栈

可提供的服务

暂无可提供的服务

Spark导出数据文件到HDFS

test/maktval res2 =spark.sql(s"""

#javascript#ajax#前端
JAVA http请求报错:unable to find valid certification path to requested target

JAVA http请求报错:unable to find valid certification path to requested target

#http#网络协议#网络 +1
spark 设置hive.exec.max.dynamic.partition不生效

spark脚本和程序中设置ive.exec.max.dynamic.partition不生效。

文章图片
#spark#hive#大数据
SparkStreaming直连kafka报错:requirement failed: Failed to get records for spark-executor-xxx topic_xxxx

依次排查后,确认是kafka服务端网络问题,因为sparkStreaming连接kafka走得是公网,数据量大网络不稳定是常有的情况,由于超过spark.streaming.kafka.consumer.poll.ms设置的120s时长,task没获取到对应的topic分区的数据,就会导致类似连接超时的异常,那这个task就fail掉了。:奇了怪了,这个任务之前一直在生产上跑的好好的,突然今天发现

文章图片
#kafka#spark
spark报错ERROR client.TransportResponseHandler: Still have 1 requests outstanding when connection from

spark任务报错(不过计算后的结果是对的):21/04/29 14:38:38 ERROR client.TransportResponseHandler: Still have 1 requests outstanding when connection from /172.16.4.156:37528 is closed21/04/29 14:38:39 ERROR cluster.Yarn

#spark
hive SQL struct类型和Array<struct>类型转json字符串

数据格式:default_address是struct类型,address是Array<struct>类型:例如:array<struct<id:string,customer_id:string>>目标格式:[{"id": "1111","customer_id": "2222"},{"id": "3333","customer_id": "444

#sql#hive#大数据
到底了