You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@storm.apache.org by "ASF GitHub Bot (JIRA)" <ji...@apache.org> on 2015/09/29 04:36:04 UTC

[jira] [Commented] (STORM-1073) SequenceFileBolt can end up in an unrecoverable state

    [ https://issues.apache.org/jira/browse/STORM-1073?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14934520#comment-14934520 ] 

ASF GitHub Bot commented on STORM-1073:
---------------------------------------

GitHub user dossett opened a pull request:

    https://github.com/apache/storm/pull/767

    STORM-1073: Refactor AbstractHdfsBolt

    STORM-969 changed the execution of HdfsBolt to be more robust to errors.  This change moves that logic up to AbstractHdfsBolt and adds two new abstract methods to AbstractHdfsBolt: writeTuple and syncTuples.  In other words AbstractHdfsBolt contains all of the error handling and retry logic for writing to HDFS, and its implementing classes will specify how to write specific content to HDFS.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/dossett/storm HdfsBolt

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/storm/pull/767.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #767
    
----
commit 375d70a4134ba93be9ff8040334d63551542ca78
Author: Aaron Dossett <aa...@target.com>
Date:   2015-09-29T02:30:14Z

    STORM-1073: Refactor AbstractHdfsBolt

----


> SequenceFileBolt can end up in an unrecoverable state
> -----------------------------------------------------
>
>                 Key: STORM-1073
>                 URL: https://issues.apache.org/jira/browse/STORM-1073
>             Project: Apache Storm
>          Issue Type: Bug
>          Components: storm-hdfs
>            Reporter: Aaron Dossett
>            Assignee: Aaron Dossett
>
> SequenceFileBolt has the same issues that HdfsBolt has in STORM-969.  This is also an opportunity to refactor AbstractHdfsBolt to most efficiently include these changes:
> Abstract HdfsBolt should define a concrete execute method and define abstract methods for:
> - writing a tuple
> - syncronizing file output
> - initiating a file rotation



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