[DRILL-2274] Unable to allocate sv2 buffer after repeated attempts : JOIN, Order by used in query

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 1.2.0
• Component/s: Execution - Relational Operators
• Assignee: Rahul Kumar Challapalli
• Reporter: Rahul Kumar Challapalli
• Created: 20/Feb/15 00:36
• Updated: 03/Oct/15 00:17
• Resolved: 29/Sep/15 00:58

Description

git.commit.id.abbrev=6676f2d

The below query fails :

select sub1.uid from `data.json` sub1 inner join `data.json` sub2 on sub1.uid = sub2.uid order by sub1.uid;

Error from the logs :

2015-02-20 00:24:08,431 [2b1981b0-149e-981b-f83f-512c587321d7:frag:1:2] ERROR o.a.d.e.w.f.AbstractStatusReporter - Error 66dba4ff-644c-4400-ab84-203256dc2600: Failure while running fragment.
 java.lang.RuntimeException: org.apache.drill.exec.memory.OutOfMemoryException: Unable to allocate sv2 buffer after repeated attempts
 	at org.apache.drill.exec.physical.impl.xsort.ExternalSortBatch.innerNext(ExternalSortBatch.java:307) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.record.AbstractRecordBatch.next(AbstractRecordBatch.java:142) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.validate.IteratorValidatorBatchIterator.next(IteratorValidatorBatchIterator.java:118) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.record.AbstractRecordBatch.next(AbstractRecordBatch.java:99) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.record.AbstractRecordBatch.next(AbstractRecordBatch.java:89) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.record.AbstractSingleRecordBatch.innerNext(AbstractSingleRecordBatch.java:51) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.svremover.RemovingRecordBatch.innerNext(RemovingRecordBatch.java:96) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.record.AbstractRecordBatch.next(AbstractRecordBatch.java:142) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.validate.IteratorValidatorBatchIterator.next(IteratorValidatorBatchIterator.java:118) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.BaseRootExec.next(BaseRootExec.java:67) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.SingleSenderCreator$SingleSenderRootExec.innerNext(SingleSenderCreator.java:97) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.BaseRootExec.next(BaseRootExec.java:57) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.work.fragment.FragmentExecutor.run(FragmentExecutor.java:116) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.work.WorkManager$RunnableWrapper.run(WorkManager.java:303) [drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) [na:1.7.0_71]
 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) [na:1.7.0_71]
 	at java.lang.Thread.run(Thread.java:745) [na:1.7.0_71]
 Caused by: org.apache.drill.exec.memory.OutOfMemoryException: Unable to allocate sv2 buffer after repeated attempts
 	at org.apache.drill.exec.physical.impl.xsort.ExternalSortBatch.newSV2(ExternalSortBatch.java:516) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	at org.apache.drill.exec.physical.impl.xsort.ExternalSortBatch.innerNext(ExternalSortBatch.java:305) ~[drill-java-exec-0.8.0-SNAPSHOT-rebuffed.jar:0.8.0-SNAPSHOT]
 	... 16 common frames omitted

On a different drillbit in the cluster, I found the below message for the same run

2015-02-20 00:24:08,435 [BitServer-6] WARN  o.a.d.exec.rpc.control.WorkEventBus - A fragment message arrived but there was no registered listener for that message: profile {
   state: FAILED
   error {
     error_id: "66dba4ff-644c-4400-ab84-203256dc2600"
     endpoint {
       address: "qa-node191.qa.lab"
       user_port: 31010
       control_port: 31011
       data_port: 31012
     }
     message: "Failure while running fragment., Unable to allocate sv2 buffer after repeated attempts [ 66dba4ff-644c-4400-ab84-203256dc2600 on qa-node191.qa.lab:31010 ]\n"
   }

I attached the data file which only has 2 records. I manually copied over the 2 records 50000 times and ran these queries on top of them.

Let me know if you need anything else

Issue Links

• is duplicated by: DRILL-2074 Queries fail with OutOfMemory Exception when Hash Join & Agg are turned off
• is related to: DRILL-2074 Queries fail with OutOfMemory Exception when Hash Join & Agg are turned off