giraph-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hudson (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (GIRAPH-1036) Allow mappers to fail early on exceptions
Date Thu, 22 Oct 2015 03:22:27 GMT

    [ https://issues.apache.org/jira/browse/GIRAPH-1036?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14968439#comment-14968439
] 

Hudson commented on GIRAPH-1036:
--------------------------------

FAILURE: Integrated in Giraph-trunk-Commit #1572 (See [https://builds.apache.org/job/Giraph-trunk-Commit/1572/])
GIRAPH-1036: Allow mappers to fail early on exceptions (majakabiljo: [http://git-wip-us.apache.org/repos/asf?p=giraph.git&a=commit&h=81d5badf7b76e9f1efde1cebe2150bee70e4cf58])
* giraph-core/src/main/java/org/apache/giraph/utils/ThreadUtils.java
* giraph-core/src/main/java/org/apache/giraph/utils/JMapHistoDumper.java
* giraph-core/src/main/java/org/apache/giraph/utils/ProgressableUtils.java
* giraph-core/src/main/java/org/apache/giraph/worker/WorkerProgressWriter.java
* giraph-core/src/main/java/org/apache/giraph/utils/ReactiveJMapHistoDumper.java
* giraph-core/src/main/java/org/apache/giraph/comm/messages/queue/AsyncMessageStoreWrapper.java
* giraph-core/src/main/java/org/apache/giraph/ooc/AdaptiveOutOfCoreEngine.java


> Allow mappers to fail early on exceptions
> -----------------------------------------
>
>                 Key: GIRAPH-1036
>                 URL: https://issues.apache.org/jira/browse/GIRAPH-1036
>             Project: Giraph
>          Issue Type: Bug
>            Reporter: Maja Kabiljo
>            Assignee: Maja Kabiljo
>
> Often when something fails in a mapper we see it stuck until its timeout passes. Digging
through this issue I found two root causes:
> - Many threads we are creating were not daemon, preventing process to exit, only main
thread should be daemon
> - When calling submit on ExecutorService, exceptions are not propagated back to the caller,
unless get is called on the future. In ProgressableUtils.getResultsWithNCallables we were
calling get on one by one future, causing us to have to wait for previous futures to finish
before getting exception which happened in later one.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message