[jira] [Commented] (FLINK-4123) CassandraWriteAheadSink can hang on cassandra failure

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-4123?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15355321#comment-15355321 ] 

ASF GitHub Bot commented on FLINK-4123:
---------------------------------------

GitHub user zentol opened a pull request:

    https://github.com/apache/flink/pull/2183

    [FLINK-4123] Cassandra sink checks for exceptions in ack phase

    This PR prevents the Cassandra Sink from locking up during the acknowledgement phase.
    
    We know check for exceptions in the ack phase. Should an exception have occurred we will retry the submission on the next notifyCheckpointComplete.
    
    In addition, separate sendValues() calls are now properly isolated. Previously the callbacks of an aborted checkpoint could interfere with the exception field and response counter for other checkpoints.

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

    $ git pull https://github.com/zentol/flink 4123_cass_hang

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

    https://github.com/apache/flink/pull/2183.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 #2183
    
----
commit 0ec0220e794e6b71053cc1061b2b70ee2865016a
Author: zentol <ch...@apache.org>
Date:   2016-06-29T14:21:21Z

    [FLINK-4123] Cassandra sink checks for exceptions in ack phase

----


> CassandraWriteAheadSink can hang on cassandra failure
> -----------------------------------------------------
>
>                 Key: FLINK-4123
>                 URL: https://issues.apache.org/jira/browse/FLINK-4123
>             Project: Flink
>          Issue Type: Bug
>          Components: Cassandra Connector
>    Affects Versions: 1.1.0
>            Reporter: Chesnay Schepler
>            Assignee: Chesnay Schepler
>            Priority: Blocker
>             Fix For: 1.1.0
>
>
> The CassandraWriteAheadSink verifies that all writes send to cassandra have been applied by counting how many were sent and how many callbacks were activated. Once all writes were sent the sink enters into a loop that is only exited once both counts are equal.
> Thus, should cassandra crash after all writes were sent, without having acknowledged all writes, the sink will deadlock in the loop.



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