[
https://issues.apache.org/jira/browse/HIVE-16233?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sergey Shelukhin updated HIVE-16233:
------------------------------------
Attachment: HIVE-16233.06.patch
Minimal patch for discarding memory. I've run some perf tests and it seems to
be ok. Defragmentation thread is actually not as useful because usually the
cache is full. In certain cases it can be useful, esp. if there's gap between
queries, but it cannot defragment fast enough during the query. I will attach
it to a separate jira on top of this patch. This is ready for review... most of
the code is similar/same as the old patch, or reused from it.
> 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, HIVE-16233.04.patch, HIVE-16233.05.patch,
> HIVE-16233.06.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)
