Ah. Missed the part where we could configure parameters at a job level.
Thanks a lot for clarifying, Yaniv.
Cheers,
Arun
On Sun, Jun 3, 2018 at 6:37 PM Yaniv Rodenski <ya...@shinto.io> wrote:
> Hi Arun,
>
> Fair point, but I think it can be configured via env/[env_name]/spark.yml
> am I wrong?
> Anyway, ideally I think we should try and have per job configurations in
> the environment rather than in the amaterasu.properties
>
> Cheers,
> Yaniv
>
> On Sun, 3 Jun 2018 at 1:32 am, Arun Manivannan <a...@arunma.com> wrote:
>
> > Gentlemen,
> >
> > Apologies for coming back late. The issue was just the minimum container
> > size that was configured in my cluster
> > (yarn.scheduler.maximum-allocation-mb). It was set at 1 GB.
> >
> > I didn't specify any spark specific memory parameters during my run (the
> > memory defaults that the SparkSetupProvider was looking at) and to top it
> > the code was setting the Xmx at 1 GB causing the overallocation and
> > failure.
> >
> > I have one minor proposal. If this is agreeable, I can raise a quick PR.
> >
> > Can we pull out the executor java options as a property in the
> > amaterasu.properties?
> >
> > amaterasu.executor.extra.java.opts = "-Xmx1G -Dscala.usejavacp=true
> > -Dhdp.version=2.6.5.0-292"
> >
> >
> > As a side effect, we must provide the flexibility to allow quotes around
> > the parameter but passing the quotes to the java command would fail. I
> > have stripped off the extra quotes in a dirty way at the moment. Should
> we
> > consider proper command parsing (and possibly convert them to be bash
> > compatible strings)?
> >
> > s"java -cp
> > spark/jars/*:executor.jar:spark/conf/:${config.YARN.hadoopHomeDir}/conf/
> > " +
> > s" ${config.amaterasuExecutorJavaOpts.replaceAll("\"","")} "+
> >
> >
> > Meanwhile, I'll also update by PR for Amaterasu-24 after pulling the
> latest
> > from the branch.
> >
> > Cheers,
> > Arun
> >
> > On Wed, May 30, 2018 at 1:25 PM Arun Manivannan <a...@arunma.com> wrote:
> >
> > > Thanks a lot, Nadav. Will get home and spend some more time on this. I
> > was
> > > in a rush and did this poor workaround. My VM is just 8 GB.
> > >
> > > Cheers
> > > Arun
> > >
> > >
> > > On Wed, May 30, 2018, 12:27 Nadav Har Tzvi <nadavhart...@gmail.com>
> > wrote:
> > >
> > >> Yaniv and I just tested it. It worked flawlessly on my end (HDP docker
> > on
> > >> AWS). Both Spark-Scala and PySpark.
> > >> It worked on Yaniv's HDP cluster as well.
> > >> Worth noting:
> > >> 1. HDP 2.6.4
> > >> 2. Cluster has total of 32GB memory available
> > >> 3. Each container is allocated 1G memory.
> > >> 4. Amaterasu.properties:
> > >>
> > >> zk=sandbox-hdp.hortonworks.com
> > >> version=0.2.0-incubating-rc3
> > >> master=192.168.33.11
> > >> user=root
> > >> mode=yarn
> > >> webserver.port=8000
> > >> webserver.root=dist
> > >> spark.version=2.6.4.0-91
> > >> yarn.queue=default
> > >> yarn.jarspath=hdfs:///apps/amaterasu
> > >> spark.home=/usr/hdp/current/spark2-client
> > >>
> > >>
> >
> #spark.home=/opt/cloudera/parcels/SPARK2-2.1.0.cloudera2-1.cdh5.7.0.p0.171658/lib/spark2
> > >> yarn.hadoop.home.dir=/etc/hadoop
> > >> spark.opts.spark.yarn.am.extraJavaOptions="-Dhdp.version=2.6.4.0-91"
> > >> spark.opts.spark.driver.extraJavaOptions="-Dhdp.version=2.6.4.0-91"
> > >>
> > >>
> > >> Arun, please share:
> > >> 1. YARN memory configurations
> > >> 2. amaterasu.properties content
> > >> 3. HDP version.
> > >>
> > >> Cheers,
> > >> Nadav
> > >>
> > >>
> > >> On 30 May 2018 at 07:11, Arun Manivannan <a...@arunma.com> wrote:
> > >>
> > >> > The pmem disabling is just temporary. I'll do a detailed analysis
> and
> > >> get
> > >> > back with a proper solution.
> > >> >
> > >> > Any hints on this front is highly appreciated.
> > >> >
> > >> > Cheers
> > >> > Arun
> > >> >
> > >> > On Wed, May 30, 2018, 01:10 Nadav Har Tzvi <nadavhart...@gmail.com>
> > >> wrote:
> > >> >
> > >> > > Yaniv, Eyal, this might be related to the same issue you faced
> with
> > >> HDP.
> > >> > > Can you confirm?
> > >> > >
> > >> > > On Tue, May 29, 2018, 17:58 Arun Manivannan <a...@arunma.com>
> > wrote:
> > >> > >
> > >> > > > +1 from me
> > >> > > >
> > >> > > > Unit Tests and Build ran fine.
> > >> > > >
> > >> > > > Tested on HDP (VM) but had trouble allocating containers (didn't
> > >> have
> > >> > > that
> > >> > > > before). Apparently Centos VMs are known to have this problem.
> > >> > Disabled
> > >> > > > physical memory check (yarn.nodemanager.pmem-check-enabled) and
> > ran
> > >> > jobs
> > >> > > > successfully.
> > >> > > >
> > >> > > >
> > >> > > >
> > >> > > >
> > >> > > >
> > >> > > > On Tue, May 29, 2018 at 10:42 PM Kirupa Devarajan <
> > >> > > kirupagara...@gmail.com
> > >> > > > >
> > >> > > > wrote:
> > >> > > >
> > >> > > > > Unit tests passing and build was successful on the branch
> > >> > > > > "version-0.2.0-incubating-rc3"
> > >> > > > >
> > >> > > > > +1 from me
> > >> > > > >
> > >> > > > > Cheers,
> > >> > > > > Kirupa
> > >> > > > >
> > >> > > > >
> > >> > > > > On Tue, May 29, 2018 at 3:06 PM, guy peleg <
> whisr...@gmail.com>
> > >> > wrote:
> > >> > > > >
> > >> > > > > > +1 looks good to me
> > >> > > > > >
> > >> > > > > > On Tue, May 29, 2018, 14:39 Nadav Har Tzvi <
> > >> nadavhart...@gmail.com
> > >> > >
> > >> > > > > wrote:
> > >> > > > > >
> > >> > > > > > > +1 approve. Tested multiple times and after a long round
> of
> > >> > fixing
> > >> > > > and
> > >> > > > > > > testing over and over.
> > >> > > > > > >
> > >> > > > > > > Cheers,
> > >> > > > > > > Nadav
> > >> > > > > > >
> > >> > > > > > >
> > >> > > > > > > On 29 May 2018 at 07:38, Yaniv Rodenski <ya...@shinto.io>
> > >> wrote:
> > >> > > > > > >
> > >> > > > > > > > Hi everyone,
> > >> > > > > > > >
> > >> > > > > > > > We have fixed the legal issues, as well as a bug found
> by
> > >> > @Nadav
> > >> > > > > please
> > >> > > > > > > > review and vote on the release candidate #3 for the
> > version
> > >> > > > > > > > 0.2.0-incubating, as follows
> > >> > > > > > > >
> > >> > > > > > > > [ ] +1, Approve the release
> > >> > > > > > > > [ ] -1, Do not approve the release (please provide
> > specific
> > >> > > > comments)
> > >> > > > > > > >
> > >> > > > > > > > The complete staging area is available for your review,
> > >> which
> > >> > > > > includes:
> > >> > > > > > > >
> > >> > > > > > > > * JIRA release notes [1],
> > >> > > > > > > > * the official Apache source release to be deployed to
> > >> > > > > dist.apache.org
> > >> > > > > > > > [2],
> > >> > > > > > > > which is signed with the key with fingerprint [3],
> > >> > > > > > > > * source code tag "version-0.2.0-incubating-rc3" [4],
> > >> > > > > > > > * Java artifacts were built with Gradle 3.1 and
> > >> OpenJDK/Oracle
> > >> > > JDK
> > >> > > > > > > > 1.8.0_151
> > >> > > > > > > >
> > >> > > > > > > > The vote will be open for at least 72 hours. It is
> adopted
> > >> by
> > >> > > > > majority
> > >> > > > > > > > approval, with at least 3 PMC affirmative votes.
> > >> > > > > > > >
> > >> > > > > > > > Thanks,
> > >> > > > > > > > Yaniv
> > >> > > > > > > >
> > >> > > > > > > > [1]
> > https://issues.apache.org/jira/secure/ReleaseNote.jspa?
> > >> > > > > > > > projectId=12321521&version=12342793
> > >> > > > > > > > [2] https://dist.apache.org/repos/
> > >> > dist/dev/incubator/amaterasu/
> > >> > > > > > 0.2.0rc3/
> > >> > > > > > > > [3]
> > >> > > > https://dist.apache.org/repos/dist/dev/incubator/amaterasu/KEYS
> > >> > > > > > > > [4] https://github.com/apache/incubator-amaterasu/tags
> > >> > > > > > > >
> > >> > > > > > >
> > >> > > > > >
> > >> > > > >
> > >> > > >
> > >> > >
> > >> >
> > >>
> > >
> >
> --
> Yaniv Rodenski
>
> +61 477 778 405 <+61%20477%20778%20405>
> ya...@shinto.io
>
