[AMBARI-22125] Druid process fails to start with OOME on small machines with default ambari configs

• Type: Bug
• Status: Resolved
• Priority: Blocker
• Resolution: Fixed
• Fix Version/s: trunk, 2.6.0
• Assignee: Nishant Bangarwa
• Reporter: Nishant Bangarwa
• Created: 03/Oct/17 22:18
• Updated: 04/Oct/17 20:38
• Resolved: 04/Oct/17 19:58

Description

Druid fails to start with ambari default configs when all the druid processes are allocated on a single node.

stack trace -

 
Caused by: java.lang.OutOfMemoryError
        at sun.misc.Unsafe.allocateMemory(Native Method)
        at java.nio.DirectByteBuffer.<init>(DirectByteBuffer.java:127)
        at java.nio.ByteBuffer.allocateDirect(ByteBuffer.java:311)
        at io.druid.offheap.OffheapBufferGenerator.get(OffheapBufferGenerator.java:53)
        at io.druid.offheap.OffheapBufferGenerator.get(OffheapBufferGenerator.java:29)
        at io.druid.collections.DefaultBlockingPool.<init>(DefaultBlockingPool.java:58)
        at io.druid.guice.DruidProcessingModule.getMergeBufferPool(DruidProcessingModule.java:126)
        at io.druid.guice.DruidProcessingModule$$FastClassByGuice$$8e266e5c.invoke(<generated>)
        at com.google.inject.internal.ProviderMethod$FastClassProviderMethod.doProvision(ProviderMethod.java:264)
        at com.google.inject.internal.ProviderMethod$Factory.provision(ProviderMethod.java:401)
        at com.google.inject.internal.ProviderMethod$Factory.get(ProviderMethod.java:376)
        at com.google.inject.internal.ProviderToInternalFactoryAdapter$1.call(ProviderToInternalFactoryAdapter.java:46)

Issue Links

• links to: review link