You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hive.apache.org by "Sergey Shelukhin (JIRA)" <ji...@apache.org> on 2017/03/01 23:43:45 UTC
[jira] [Comment Edited] (HIVE-16078) improve abort checking in
Tez/LLAP
[ https://issues.apache.org/jira/browse/HIVE-16078?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15891303#comment-15891303 ]
Sergey Shelukhin edited comment on HIVE-16078 at 3/1/17 11:43 PM:
------------------------------------------------------------------
[~sseth] will test in the cluster eventually; the constant is already double; not sure how nRows can be reversed - can you elaborate? the limit is only changed when nRows is set to 0. As for long ops, not sure. In this case each item is a VRB, not row, so assuming no interrupt checks in the pipeline 1000-row check will be ~1M rows processed between check. I also suspect based on the method names (forwardOverflow etc.) that some batching may be happening internally, making it even more rows. Perhaps [~mmccline] can comment given the callstack above
was (Author: sershe):
[~sseth] will test in the cluster eventually; the constant is already double; not sure how nRows can be reversed - can you elaborate? the count is only changed when nRows is set to 0. As for long ops, not sure. In this case each item is a VRB, not row, so assuming no interrupt checks in the pipeline 1000-row check will be ~1M rows processed between check. I also suspect based on the method names (forwardOverflow etc.) that some batching may be happening internally, making it even more rows. Perhaps [~mmccline] can comment given the callstack above
> improve abort checking in Tez/LLAP
> ----------------------------------
>
> Key: HIVE-16078
> URL: https://issues.apache.org/jira/browse/HIVE-16078
> Project: Hive
> Issue Type: Bug
> Reporter: Sergey Shelukhin
> Assignee: Sergey Shelukhin
> Attachments: HIVE-16078.patch
>
>
> Sometimes, a fragment can run for a long time after a query fails. It looks from logs like the abort/interrupt were called correctly on the thread, yet the thread hangs around minutes after, doing the below. Other tasks for the same job appear to have exited correctly, after the same abort logic (at least, the same log lines, fwiw)
> {noformat}
> at org.apache.hadoop.hive.ql.exec.vector.VectorCopyRow.copyByValue(VectorCopyRow.java:317)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:263)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.forwardOverflow(VectorMapJoinGenerateResultOperator.java:628)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinGenerateResultOperator.generateHashMapResultMultiValue(VectorMapJoinGenerateResultOperator.java:277)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerGenerateResultOperator.finishInner(VectorMapJoinInnerGenerateResultOperator.java:189)
> at org.apache.hadoop.hive.ql.exec.vector.mapjoin.VectorMapJoinInnerLongOperator.process(VectorMapJoinInnerLongOperator.java:389)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.VectorSelectOperator.process(VectorSelectOperator.java:137)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.vector.VectorFilterOperator.process(VectorFilterOperator.java:123)
> at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:897)
> at org.apache.hadoop.hive.ql.exec.TableScanOperator.process(TableScanOperator.java:130)
> at org.apache.hadoop.hive.ql.exec.vector.VectorMapOperator.process(VectorMapOperator.java:783)
> at org.apache.hadoop.hive.ql.exec.tez.MapRecordSource.processRow(MapRecordSource.java:86)
> at org.apache.hadoop.hive.ql.exec.tez.MapRecordSource.pushRecord(MapRecordSource.java:70)
> at org.apache.hadoop.hive.ql.exec.tez.MapRecordProcessor.run(MapRecordProcessor.java:420)
> {noformat}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)