[
https://issues.apache.org/jira/browse/HIVE-16233?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15993688#comment-15993688
]
Sergey Shelukhin commented on HIVE-16233:
-----------------------------------------
After some discussion, this patch probably can be simplified (and also made
better wrt task thread load).
The code that finds contiguous areas and the code that moves stuff need to be
separated. The former will run in-line with allocation when the block is not
available. It will not move anything but rather evict everything and take the
space. Sad, but unavoidable.
The defrag code will run on a background thread and will not need to bother
with contiguous areas, it will merely try to coalesce similar sized blocks
together and/or just move stuff left in every arena.
> llap: Query failed with AllocatorOutOfMemoryException
> -----------------------------------------------------
>
> Key: HIVE-16233
> URL: https://issues.apache.org/jira/browse/HIVE-16233
> Project: Hive
> Issue Type: Bug
> Components: llap
> Reporter: Siddharth Seth
> Assignee: Sergey Shelukhin
> Attachments: HIVE-16233.01.patch, HIVE-16233.02.patch,
> HIVE-16233.03.patch
>
>
> {code}
> TaskAttempt 5 failed, info=[Error: Error while running task ( failure ) :
> attempt_1488231257387_2288_25_05_000056_5:java.lang.RuntimeException:
> org.apache.hadoop.hive.ql.metadata.HiveException: java.io.IOException:
> java.io.IOException:
> org.apache.hadoop.hive.common.io.Allocator$AllocatorOutOfMemoryException:
> Failed to allocate 262144; at 0 out of 1
> at
> org.apache.hadoop.hive.ql.exec.tez.TezProcessor.initializeAndRunProcessor(TezProcessor.java:211)
> at
> org.apache.hadoop.hive.ql.exec.tez.TezProcessor.run(TezProcessor.java:168)
> at
> org.apache.tez.runtime.LogicalIOProcessorRuntimeTask.run(LogicalIOProcessorRuntimeTask.java:370)
> at
> org.apache.tez.runtime.task.TaskRunner2Callable$1.run(TaskRunner2Callable.java:73)
> at
> org.apache.tez.runtime.task.TaskRunner2Callable$1.run(TaskRunner2Callable.java:61)
> at java.security.AccessController.doPrivileged(Native Method)
> at javax.security.auth.Subject.doAs(Subject.java:422)
> at
> org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1724)
> at
> org.apache.tez.runtime.task.TaskRunner2Callable.callInternal(TaskRunner2Callable.java:61)
> at
> org.apache.tez.runtime.task.TaskRunner2Callable.callInternal(TaskRunner2Callable.java:37)
> at org.apache.tez.common.CallableWithNdc.call(CallableWithNdc.java:36)
> at
> org.apache.hadoop.hive.llap.daemon.impl.StatsRecordingThreadPool$WrappedCallable.call(StatsRecordingThreadPool.java:110)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
> at java.lang.Thread.run(Thread.java:745)
> Caused by: org.apache.hadoop.hive.ql.metadata.HiveException:
> java.io.IOException: java.io.IOException:
> org.apache.hadoop.hive.common.io.Allocator$AllocatorOutOfMemoryException:
> Failed to allocate 262144; at 0 out of 1
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordSource.pushRecord(MapRecordSource.java:74)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.run(MapRecordProcessor.java:419)
> at
> org.apache.hadoop.hive.ql.exec.tez.TezProcessor.initializeAndRunProcessor(TezProcessor.java:185)
> ... 15 more
> Caused by: java.io.IOException: java.io.IOException:
> org.apache.hadoop.hive.common.io.Allocator$AllocatorOutOfMemoryException:
> Failed to allocate 262144; at 0 out of 1
> at
> org.apache.hadoop.hive.io.HiveIOExceptionHandlerChain.handleRecordReaderNextException(HiveIOExceptionHandlerChain.java:121)
> at
> org.apache.hadoop.hive.io.HiveIOExceptionHandlerUtil.handleRecordReaderNextException(HiveIOExceptionHandlerUtil.java:77)
> at
> org.apache.hadoop.hive.ql.io.HiveContextAwareRecordReader.doNext(HiveContextAwareRecordReader.java:365)
> at
> org.apache.hadoop.hive.ql.io.HiveRecordReader.doNext(HiveRecordReader.java:79)
> at
> org.apache.hadoop.hive.ql.io.HiveRecordReader.doNext(HiveRecordReader.java:33)
> at
> org.apache.hadoop.hive.ql.io.HiveContextAwareRecordReader.next(HiveContextAwareRecordReader.java:116)
> at
> org.apache.hadoop.mapred.split.TezGroupedSplitsInputFormat$TezGroupedSplitsRecordReader.next(TezGroupedSplitsInputFormat.java:151)
> at
> org.apache.tez.mapreduce.lib.MRReaderMapred.next(MRReaderMapred.java:116)
> at
> org.apache.hadoop.hive.ql.exec.tez.MapRecordSource.pushRecord(MapRecordSource.java:62)
> ... 17 more
> Caused by: java.io.IOException:
> org.apache.hadoop.hive.common.io.Allocator$AllocatorOutOfMemoryException:
> Failed to allocate 262144; at 0 out of 1
> at
> org.apache.hadoop.hive.ql.io.orc.encoded.EncodedReaderImpl.readEncodedColumns(EncodedReaderImpl.java:425)
> at
> org.apache.hadoop.hive.llap.io.encoded.OrcEncodedDataReader.performDataRead(OrcEncodedDataReader.java:413)
> at
> org.apache.hadoop.hive.llap.io.encoded.OrcEncodedDataReader$4.run(OrcEncodedDataReader.java:235)
> at
> org.apache.hadoop.hive.llap.io.encoded.OrcEncodedDataReader$4.run(OrcEncodedDataReader.java:232)
> at java.security.AccessController.doPrivileged(Native Method)
> at javax.security.auth.Subject.doAs(Subject.java:422)
> at
> org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1724)
> at
> org.apache.hadoop.hive.llap.io.encoded.OrcEncodedDataReader.callInternal(OrcEncodedDataReader.java:232)
> at
> org.apache.hadoop.hive.llap.io.encoded.OrcEncodedDataReader.callInternal(OrcEncodedDataReader.java:94)
> ... 6 more
> Caused by:
> org.apache.hadoop.hive.common.io.Allocator$AllocatorOutOfMemoryException:
> Failed to allocate 262144; at 0 out of 1
> at
> org.apache.hadoop.hive.llap.cache.BuddyAllocator.allocateMultiple(BuddyAllocator.java:275)
> at
> org.apache.hadoop.hive.ql.io.orc.encoded.EncodedReaderImpl.readEncodedStream(EncodedReaderImpl.java:675)
> at
> org.apache.hadoop.hive.ql.io.orc.encoded.EncodedReaderImpl.readEncodedColumns(EncodedReaderImpl.java:412)
> ... 14 more
> {code}
> Also reported by [~selinazh] while using older bits.
> This JIRA will be used to track a long-term fix.
> We were initially planning to write a different allocator with compaction,
> but now we are stuck with buddyallocator.
> There are two approaches - compaction on top of buddy allocator (should be
> fairly simple actually, aside from indirection... luckily for the latter, we
> can take the buffer out in most places and we can reuse the existing
> locking-to-prevent-eviction mechanism to also prevent reallocation); or
> writing a better allocator. Probably the former is preferred, need to think
> about the latter.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
