This job will convert Kylin's cuboid (sequence files) to HBases HFile
format; It will use HBase classes in the MR; This step is stable as we
didn't see error for some time; The only issue I know for this step is, if
the cube is very large, the convert may take long time and makes Kylin job
engine time out; Any way, you can investigate from another direction: is
there any change be made on this env before this error?

2015-05-26 16:08 GMT+08:00 dong wang <[email protected]>:

> Hi shaofeng, I checked the log for a long time, this may be the only hints:
>
> 15/05/26 16:03:07 INFO hs.HistoryFileManager: Moving hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done_intermediate/root/job_1432623918955_0018_conf.xml
> to hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done/2015/05/26/000000/job_1432623918955_0018_conf.xml
> 15/05/26 16:05:17 INFO hs.JobHistory: Starting scan to move intermediate
> done files
> 15/05/26 16:05:28 INFO hs.CompletedJob: Loading job: job_1432623918955_0019
> from file: hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done_intermediate/root/job_1432623918955_0019-1432627478584-root-Kylin_HFile_Generator_tbl1-1432627521052-5-0-FAILED-root.root-1432627485193.jhist
> 15/05/26 16:05:28 INFO hs.CompletedJob: Loading history file: [hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done_intermediate/root/job_1432623918955_0019-1432627478584-root-Kylin_HFile_Generator_tbl1-1432627521052-5-0-FAILED-root.root-1432627485193.jhist
> ]
> 15/05/26 16:05:28 INFO jobhistory.JobSummary:
>
> jobId=job_1432623918955_0019,submitTime=1432627478584,launchTime=1432627485193,firstMapTaskLaunchTime=1432627487604,firstReduceTaskLaunchTime=1432627495325,finishTime=1432627521052,resourcesPerMap=4096,resourcesPerReduce=8192,numMaps=5,numReduces=1,user=root,queue=default,status=FAILED,mapSlotSeconds=57,reduceSlotSeconds=75,jobName=Kylin_HFile_Generator_tbl1_1_Step
> 15/05/26 16:05:28 INFO hs.HistoryFileManager: Deleting JobSummary file:
> [hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done_intermediate/root/job_1432623918955_0019.summary
> ]
> 15/05/26 16:05:28 INFO hs.HistoryFileManager: Moving hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done_intermediate/root/job_1432623918955_0019-1432627478584-root-Kylin_HFile_Generator_tbl1-1432627521052-5-0-FAILED-root.root-1432627485193.jhist
> to hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done/2015/05/26/000000/job_1432623918955_0019-1432627478584-root-Kylin_HFile_Generator_tbl1-1432627521052-5-0-FAILED-root.root-1432627485193.jhist
> 15/05/26 16:05:28 INFO hs.HistoryFileManager: Moving hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done_intermediate/root/job_1432623918955_0019_conf.xml
> to hdfs://
>
> abc-master.test.com:8020/tmp/hadoop-yarn/staging/history/done/2015/05/26/000000/job_1432623918955_0019_conf.xml
>
> 2015-05-26 16:02 GMT+08:00 dong wang <[email protected]>:
>
> > does anyone know that is there any stuff related to "HBase" itself when
> ""Convert
> > Cuboid Data to HFile"?
> >
> > 2015-05-26 15:58 GMT+08:00 dong wang <[email protected]>:
> >
> >> do anyone meet with the same problem with CDH 5.4.2 +
> kylin-0.7.1-staging
> >> source codes?
> >>
> >> 2015-05-26 13:54 GMT+08:00 dong wang <[email protected]>:
> >>
> >>> sorry, I mis-click the log information button, I will check the MR log
> >>> first
> >>>
> >>> 2015-05-26 13:51 GMT+08:00 dong wang <[email protected]>:
> >>>
> >>>> today, I update the environment, and when building the cube, the error
> >>>> looks the following:
> >>>> 2015-05-25 22:40:04.388 - State of Hadoop job:
> >>>> job_1432568508250_0142:ACCEPTED - UNDEFINED
> >>>> 2015-05-25 22:40:14.405 - State of Hadoop job:
> >>>> job_1432568508250_0142:RUNNING - UNDEFINED
> >>>> 2015-05-25 22:40:24.424 - State of Hadoop job:
> >>>> job_1432568508250_0142:RUNNING - UNDEFINED
> >>>> 2015-05-25 22:40:34.438 - State of Hadoop job:
> >>>> job_1432568508250_0142:RUNNING - UNDEFINED
> >>>> 2015-05-25 22:40:44.451 - State of Hadoop job:
> >>>> job_1432568508250_0142:RUNNING - UNDEFINED
> >>>> 2015-05-25 22:40:54.465 - State of Hadoop job:
> >>>> job_1432568508250_0142:FINISHED - FAILED
> >>>> no counters for job job_1432568508250_0142
> >>>>
> >>>>
> >>>> and when looking into the MR log, it says:
> >>>>
> >>>> Total Vmem allocated for Containers 29.40 GB
> >>>> Vmem enforcement enabled false
> >>>> Total Pmem allocated for Container 14 GB
> >>>> Pmem enforcement enabled true
> >>>> Total VCores allocated for Containers 8
> >>>> NodeHealthyStatus true
> >>>> LastNodeHealthTime Tue May 26 13:20:13 CST 2015
> >>>> NodeHealthReport
> >>>> Node Manager Version: 2.6.0-cdh5.4.2 from
> >>>> 15b703c8725733b7b2813d2325659eb7d57e7a3f by jenkins source checksum
> >>>> e7a085479aa1989b5cecfabea403549 on 2015-05-20T00:09Z
> >>>> Hadoop Version: 2.6.0-cdh5.4.2 from
> >>>> 15b703c8725733b7b2813d2325659eb7d57e7a3f by jenkins source checksum
> >>>> de74f1adb3744f8ee85d9a5b98f90d on 2015-05-20T00:03Z
> >>>>
> >>>
> >>>
> >>
> >
>

Reply via email to