Mailing-List: contact commits-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cassandra.apache.org
Date: Thu, 11 Feb 2016 04:28:18 +0000 (UTC)
From: "Stefania (JIRA)" <jira@apache.org>
To: commits@cassandra.apache.org
Message-ID: <JIRA.12938418.1455160809000.23346.1455164898049@Atlassian.JIRA>
In-Reply-To: <JIRA.12938418.1455160809000@Atlassian.JIRA>
References: <JIRA.12938418.1455160809000@Atlassian.JIRA>
 <JIRA.12938418.1455160809580@arcas>
Subject: [jira] [Commented] (CASSANDRA-11156) AssertionError when listing
 sstable files on inconsistent disk state
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/CASSANDRA-11156?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15142232#comment-15142232 ] 

Stefania commented on CASSANDRA-11156:
--------------------------------------

3.0 patch is [here|https://github.com/stef1927/cassandra/commits/11156-3.0] and merges cleanly upwards. There is already a unit test capable of covering this code path but it expects an exception thrown when listing files in an inconsistent state and it did not check for assertion errors.

Tests:

* http://cassci.datastax.com/view/Dev/view/stef1927/job/stef1927-11156-3.0-testall
* http://cassci.datastax.com/view/Dev/view/stef1927/job/stef1927-11156-3.0-dtest


> AssertionError when listing sstable files on inconsistent disk state
> --------------------------------------------------------------------
>
>                 Key: CASSANDRA-11156
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-11156
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Local Write-Read Paths
>            Reporter: Stefania
>            Assignee: Stefania
>             Fix For: 3.0.x, 3.x
>
>
> We've had the following failure which, although very hard to reproduce, could happen if we find an inconsistent disk state:
> http://cassci.datastax.com/view/Dev/view/krummas/job/krummas-marcuse-11148-trunk-dtest/1/testReport/junit/consistency_test/TestConsistency/readrepair_test_2/
> To fix it we just need to reset the records in {{LogFile}} before reading it again in {{LogFileLister}}:
> {code}
>     // otherwise read the file again to see if it is completed now
>     readTxnLog(txnFile);
> {code}


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