[
https://issues.apache.org/jira/browse/TEZ-2251?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14391890#comment-14391890
]
Rajesh Balamohan commented on TEZ-2251:
---------------------------------------
I haven't personally tried it, will give it a try with readLock.lock().
However, With readLock.lock() in "public synchronized List<InputSpec>", there
is a possibility of getting into a deadlock.
"App Shared Pool" threads could be holding on to the "writeLock" while
invoking setParallelism(). In the mean time, "dispatcher" thread could be
processing TaskAttemptEventDispatcher.handle(), which internally invokes
Edge.getSourceSpec()-->VertexImpl.getTotalTasks().
This requests for readLock() from VertexImpl, which would block (as write lock
is being held by "App Shared Pool" threads).
> Enabling auto reduce parallelism in certain jobs causes DAG to hang
> -------------------------------------------------------------------
>
> Key: TEZ-2251
> URL: https://issues.apache.org/jira/browse/TEZ-2251
> Project: Apache Tez
> Issue Type: Bug
> Reporter: Rajesh Balamohan
> Attachments: TEZ-2251.VertexImpl.patch,
> TEZ-2251.fix_but_slows_down.patch, hive_console.png,
> tez-2251.vertexpatch.am.log.gz, tez_2251_dag.png
>
>
> Scenario:
> - Run TPCH query20
> (https://github.com/cartershanklin/hive-testbench/blob/master/sample-queries-tpch/tpch_query20.sql)
> at 1 TB scale (tez-master branch, hive trunk)
> - Enable auto reduce parallelism
> - DAG didn't complete and got stuck in "Reducer 6"
> Vertex parallelism of "Reducer 5 & 6" happens within a span of 3
> milliseconds, and tasks of "reducer 5" ends up producing wrong partition
> details as it sees the updated task numbers of reducer 6 when scheduled.
> This causes, job to hang.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
