accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mike Drob (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (ACCUMULO-767) write-ahead logs prematurely recovered
Date Tue, 22 Apr 2014 18:10:16 GMT

    [ https://issues.apache.org/jira/browse/ACCUMULO-767?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13977146#comment-13977146
] 

Mike Drob commented on ACCUMULO-767:
------------------------------------

Since ACCUMULO-722 was closed as Won't Fix, is this still relevant outside of that context?

> write-ahead logs prematurely recovered
> --------------------------------------
>
>                 Key: ACCUMULO-767
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-767
>             Project: Accumulo
>          Issue Type: Bug
>          Components: tserver
>            Reporter: Eric Newton
>            Assignee: Keith Turner
>            Priority: Minor
>
> While using the Distributed Name Node implementation of HDFS (see ACCUMULO-722), a WAL
was unnecessarily recovered before it was closed.  This happened because a log reference was
written to the METADATA table before the location was set.  With HDFS, the NameNode recovers
the lease and prevents further writes to the log.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Mime
View raw message