spark-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Sea" <261810...@qq.com>
Subject 回复: Jobs hang for a long time
Date Wed, 13 Apr 2016 09:50:16 GMT
Hi, Ted:
    My spark version is 1.6.1. 
    The logs in bigdata-arch-hdp1078:


16/04/12 20:33:25 ERROR MapOutputTracker: Missing an output location for shuffle 565
16/04/12 20:33:25 ERROR MapOutputTracker: Missing an output location for shuffle 565
16/04/12 20:33:25 ERROR MapOutputTracker: Missing an output location for shuffle 565
16/04/12 22:43:06 ERROR CoarseGrainedExecutorBackend: RECEIVED SIGNAL 15: SIGTERM





------------------ 原始邮件 ------------------
发件人: "Ted Yu";<yuzhihong@gmail.com>;
发送时间: 2016年4月13日(星期三) 下午5:19
收件人: "Sea"<261810726@qq.com>; 
抄送: "user"<user@spark.apache.org>; "joshrosen"<joshrosen@databricks.com>;

主题: Re: Jobs hang for a long time



Can you provide some more information ?

Release of Spark.

Have you checked the logs from node bigdata-arch-hdp1078 which was mentioned in the failed
stage ?



On Wed, Apr 13, 2016 at 2:03 AM, Sea <261810726@qq.com> wrote:
I have a lot of jobs hang for a long time.





This is one job, one stage failed and three stages are pending...







The stage 498 do not start.





Anyone can help?
Mime
View raw message