Support Questions
Find answers, ask questions, and share your expertise
Announcements
Alert: Welcome to the Unified Cloudera Community. Former HCC members be sure to read and learn how to activate your account here.

Big-Bench run failed on CDH 5.5.2

Big-Bench run failed on CDH 5.5.2

Explorer

I try to use BigBench for benchmarking on my CDH installation. It is stucked in Data generation state. I have seen the job on the cluster and have found that:

[java, -Xmx800m, -cp, /etc/hadoop/conf:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop/lib/:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop/.//:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop-hdfs/./:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop-hdfs/lib/:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop-hdfs/.//:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop-yarn/lib/:/opt/cloudera/parcels/CDH-5.5.2-1.cdh5.5.2.p0.4/lib/hadoop/libexec/../../hadoop-yarn/.//:/opt/cloudera/parcels/CDH/lib/hadoop-mapreduce/lib/:/opt/cloudera/parcels/CDH/lib/hadoop-mapreduce/.//:pdgf.jar, -Dpdgf.log.folder=/tmp/pdgfLog/HadoopClusterExec.taskNumber, -Dcore-site.xml=/etc/hadoop/conf.cloudera.hdfs/core-site.xml, -Dhdfs-site.xml=/etc/hadoop/conf.cloudera.hdfs/hdfs-site.xml, -Djava.library.path=/opt/cloudera/parcels/CDH/lib/hadoop/lib/native, -DFileChannelProvider=pdgf.util.caching.fileWriter.HDFSChannelProvider, -Ddfs.replication.override=1, pdgf.Controller, -nc, HadoopClusterExec.tasks, -nn, HadoopClusterExec.taskNumber, -ns, -c, -sp, REFRESH_PHASE, 0, -o, '/user/sensor/benchmarks/bigbench/data/'+table.getName()+'/', -workers, 1, -ap, 3000, -s, -sf, 1]

State:  ACCEPTED
FinalStatus:    UNDEFINED
Started:    Fri Mar 11 09:12:04 +0100 2016
Elapsed:    1hrs, 3mins, 45sec
Tracking URL:   UNASSIGNED

 

It is maybe because of overusage of resources. I have already checked and modified yarn.scheduler.maximum-allocation-mb, mapreduce.map.memory.mb, mapreduce.map.java.opts, mapreduce.reduce.memory.mb, mapreduce.reduce.java.opts values.

 

I have found the rule that:

mapreduce.map.memory.mb < yarn.nodemanager.resource.memory-mb < yarn.scheduler.maximum-allocation-mb

Cluster is running on 2 KVM virtualized VM. Each have 4 CPU cores, first VM have 36 GB RAM and 80 GB disc space, second have 16 GB RAM and 40 GB disc space.

How should values above configured to run jobs on this infrastructure? All the jobs are waiting for execution in ACCEPTED state. (I have tried also Terasort and TestDFSIO with various memory limit values). I have stucked in this problem.

1 REPLY 1

Re: Big-Bench run failed on CDH 5.5.2

Explorer

I use these values.

 

values