spark源码编译命令输入之后,等了20多分钟没反应,貌似卡住了。这样正常吗?

来源:3-3 补录:Spark源码编译中的坑

qq_鸣神剑二_0

2018-08-31

http://img.mukewang.com/szimg/5b88e067000137dc12160511.jpg

写回答

2回答

慕粉4424978

2018-09-04

i also stuck here,  it is normal  taking a long time for downloading the jar , after the hang, it will produce some error or  good thing : build success;  

if errors appear, please check the scala version and make sure the maven central and cloudera mirror are set in the maven setting.xml; the missing of jars because of network problem will make some error which is hard to diagnose ; many error is caused by jar missing;

the most important solution for me in this build : check the jars in your maven repository and download the maven jars manually by thunder which is also mentioned by the teacher ;

the second: skip the tests by -Dskiptests , i missed this and it caused the out of memory of jvm


   

0
1
Michael_PK
要想快,就把make文件中的scala版本啥的改了,就不会根据系统环境变量来处理了
2018-09-04
共1条回复

Michael_PK

2018-08-31

现在是在检查环境变量,你make-dis....中那几个参数改了吗? 改了之后就不会检查了

0
2
Michael_PK
有几个手工改的,比如scala版本啥的
2018-08-31
共2条回复

以慕课网日志分析为例 进入大数据Spark SQL的世界

快速转型大数据:Hadoop,Hive,SparkSQL步步为赢

1644 学习 · 1129 问题

查看课程