我是hadoop和yarn的新手,尝试在Hive2直线上运行HiveInsert查询。我得到的信息是:info:tez会话尚未创建。开幕式上,我的系统似乎被挂断了。我试着在yarn ui 8088端口上检查日志,我可以看到我的工作被杀了
User: hive
Name: HIVE-e5f3af53-8861-4095-b6ea-4b6f28300ccc
Application Type: TEZ
Application Tags:
Application Priority: 0 (Higher Integer value indicates higher priority)
YarnApplicationState: KILLED
Queue: default
FinalStatus Reported by AM: KILLED
Started: Wed Jan 23 17:15:07 +0530 2019
Elapsed: 8mins, 33sec
Tracking URL: History
Log Aggregation Status: NOT_START
Application Timeout (Remaining Time): Unlimited
Diagnostics: Application application_1546858571448_0008 was killed by user hive at 10.5.3.113
Unmanaged Application: false
Application Node Label expression: <Not set>
AM container Node Label expression: <DEFAULT_PARTITION>
在hdfs中检查vi gc.log-201901071626文件中的日志状态后,我可以看到gc分配失败
2019-01-08T09:56:45.906+0530: 63044.883: [GC (Allocation Failure) 2019-01-08T09:56:45.906+0530: 63044.883: [ParNew: 106452K->1675K(118016K), 0.0137290 secs] 133048K->28271K(1035520K), 0.0138677 secs] [Times: user=0.03 sys=0.00, real=0.02 secs]
2019-01-08T12:45:45.849+0530: 73184.825: [GC (Allocation Failure) 2019-01-08T12:45:45.849+0530: 73184.826: [ParNew: 106635K->940K(118016K), 0.0119422 secs] 133231K->27986K(1035520K), 0.0121155 secs] [Times: user=0.04 sys=0.00, real=0.02 secs]
请帮我找到合适的解决办法
我试着把引擎换成mr,但不支持
1条答案
按热度按时间bjg7j2ky1#
hiveserver上是否安装了tez客户端?这是一项要求。您也可以尝试降低tez.am.resource.memory.mb