You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@accumulo.apache.org by "ASF subversion and git services (JIRA)" <ji...@apache.org> on 2014/03/20 22:49:44 UTC

[jira] [Commented] (ACCUMULO-2512) MAC stop should waitFor processes

    [ https://issues.apache.org/jira/browse/ACCUMULO-2512?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13942352#comment-13942352 ] 

ASF subversion and git services commented on ACCUMULO-2512:
-----------------------------------------------------------

Commit 0753a754b0a1751ab7f2bb4ddd922aa9dd492252 in accumulo's branch refs/heads/1.4.5-SNAPSHOT from [~elserj]
[ https://git-wip-us.apache.org/repos/asf?p=accumulo.git;h=0753a75 ]

ACCUMULO-2512 Wait for each MAC process to exit before returning from MAC.stop()


> MAC stop should waitFor processes
> ---------------------------------
>
>                 Key: ACCUMULO-2512
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-2512
>             Project: Accumulo
>          Issue Type: Bug
>    Affects Versions: 1.4.4, 1.5.0
>            Reporter: Keith Turner
>            Assignee: Josh Elser
>             Fix For: 1.4.5, 1.5.2, 1.6.0
>
>
> [~elserj] ran into an issue w/ VolumeIT.testDirtyReplaceVolumes() while working on ACCUMULO-2061 where a file was not being a moved.  The test would rename a directory after stopping MAC, however MAC would create a file in the old location.  We determined the cause of the problem seems to be that MAC.stop() does not call waitFor() after destroying processes.   Calling Process.destroy() seems to asynchronously kill a process (via SIGTERM).  So after calling MAC.stop() tserver processes were still running for a bit.



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