Are you using the capacity scheduler or fifo scheduler without multi resource scheduling by any chance?

On Thu, Feb 12, 2015 at 1:51 PM, Anders Arpteg <arpteg@spotify.com> wrote:
The nm logs only seems to contain similar to the following. Nothing else in the same time range. Any help?

2015-02-12 20:47:31,245 WARN org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl: Event EventType: KILL_CONTAINER sent to absent container container_1422406067005_0053_01_000002
2015-02-12 20:47:31,246 WARN org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl: Event EventType: KILL_CONTAINER sent to absent container container_1422406067005_0053_01_000012
2015-02-12 20:47:31,246 WARN org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl: Event EventType: KILL_CONTAINER sent to absent container container_1422406067005_0053_01_000022
2015-02-12 20:47:31,246 WARN org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl: Event EventType: KILL_CONTAINER sent to absent container container_1422406067005_0053_01_000032
2015-02-12 20:47:31,246 WARN org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl: Event EventType: KILL_CONTAINER sent to absent container container_1422406067005_0053_01_000042
2015-02-12 21:24:30,515 WARN org.apache.hadoop.yarn.server.nodemanager.containermanager.ContainerManagerImpl: Event EventType: FINISH_APPLICATION sent to absent application application_1422406067005_0053


On Thu, Feb 12, 2015 at 10:38 PM, Sandy Ryza <sandy.ryza@cloudera.com> wrote:
It seems unlikely to me that it would be a 2.2 issue, though not entirely impossible.  Are you able to find any of the container logs?  Is the NodeManager launching containers and reporting some exit code?

-Sandy

On Thu, Feb 12, 2015 at 1:21 PM, Anders Arpteg <arpteg@spotify.com> wrote:
No, not submitting from windows, from a debian distribution. Had a quick look at the rm logs, and it seems some containers are allocated but then released again for some reason. Not easy to make sense of the logs, but here is a snippet from the logs (from a test in our small test cluster) if you'd like to have a closer look: http://pastebin.com/8WU9ivqC

Sandy, sounds like it could possible be a 2.2 issue then, or what do you think?

Thanks,
Anders

On Thu, Feb 12, 2015 at 3:11 PM, Aniket Bhatnagar <aniket.bhatnagar@gmail.com> wrote:

This is tricky to debug. Check logs of node and resource manager of YARN to see if you can trace the error. In the past I have to closely look at arguments getting passed to YARN container (they get logged before attempting to launch containers). If I still don't get a clue, I had to check the script generated by YARN to execute the container and even run manually to trace at what line the error has occurred.

BTW are you submitting the job from windows?


On Thu, Feb 12, 2015, 3:34 PM Anders Arpteg <arpteg@spotify.com> wrote:
Interesting to hear that it works for you. Are you using Yarn 2.2 as well? No strange log message during startup, and can't see any other log messages since no executer gets launched. Does not seems to work in yarn-client mode either, failing with the exception below.

Exception in thread "main" org.apache.spark.SparkException: Yarn application has already ended! It might have been killed or unable to
launch application master.
        at org.apache.spark.scheduler.cluster.YarnClientSchedulerBackend.waitForApplication(YarnClientSchedulerBackend.scala:119)
        at org.apache.spark.scheduler.cluster.YarnClientSchedulerBackend.start(YarnClientSchedulerBackend.scala:59)
        at org.apache.spark.scheduler.TaskSchedulerImpl.start(TaskSchedulerImpl.scala:141)
        at org.apache.spark.SparkContext.<init>(SparkContext.scala:370)
        at com.spotify.analytics.AnalyticsSparkContext.<init>(AnalyticsSparkContext.scala:8)
        at com.spotify.analytics.DataSampler$.main(DataSampler.scala:42)
        at com.spotify.analytics.DataSampler.main(DataSampler.scala)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:39)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:597)
        at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:551)
        at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:155)
        at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:178)
        at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:99)
        at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)

/Anders


On Thu, Feb 12, 2015 at 1:33 AM, Sandy Ryza <sandy.ryza@cloudera.com> wrote:
Hi Anders,

I just tried this out and was able to successfully acquire executors.  Any strange log messages or additional color you can provide on your setup?  Does yarn-client mode work?

-Sandy

On Wed, Feb 11, 2015 at 1:28 PM, Anders Arpteg <arpteg@spotify.com> wrote:
Hi,

Compiled the latest master of Spark yesterday (2015-02-10) for Hadoop 2.2 and failed executing jobs in yarn-cluster mode for that build. Works successfully with spark 1.2 (and also master from 2015-01-16), so something has changed since then that prevents the job from receiving any executors on the cluster.

Basic symptoms are that the jobs fires up the AM, but after examining the "executors" page in the web ui, only the driver is listed, no executors are ever received, and the driver keep waiting forever. Has anyone seemed similar problems?

Thanks for any insights,
Anders