Mailing-List: contact notifications-help@accumulo.apache.org; run by ezmlm
Precedence: bulk
Reply-To: jira@apache.org
Date: Mon, 26 Jan 2015 23:40:34 +0000 (UTC)
From: "Ed Coleman (JIRA)" <jira@apache.org>
To: notifications@accumulo.apache.org
Message-ID: <JIRA.12709289.1397837162000.178236.1422315634738@Atlassian.JIRA>
In-Reply-To: <JIRA.12709289.1397837162000@Atlassian.JIRA>
References: <JIRA.12709289.1397837162000@Atlassian.JIRA>
 <JIRA.12709289.1397837162950@arcas>
Subject: [jira] [Updated] (ACCUMULO-2696) Master has misleading log message
 in case of failure for status thread
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/ACCUMULO-2696?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ed Coleman updated ACCUMULO-2696:
---------------------------------
    Status: Patch Available  (was: Open)

Had some time so I took a crack at this - can you see if this is what you were looking for?

> Master has misleading log message in case of failure for status thread
> ----------------------------------------------------------------------
>
>                 Key: ACCUMULO-2696
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-2696
>             Project: Accumulo
>          Issue Type: Bug
>          Components: master
>    Affects Versions: 1.6.0, 1.5.0, 1.4.0
>            Reporter: Sean Busbey
>            Assignee: Al Krinker
>            Priority: Minor
>              Labels: logging, newbie
>             Fix For: 1.7.0, 1.6.3
>
>
> the status thread is a worker within the Master that handles internal state transitions. It contains this handler in a catch-all block to ensure it keeps working:
> {noformat}
>         } catch (Throwable t) {
>           log.error("Error balancing tablets", t);
>           UtilWaitThread.sleep(WAIT_BETWEEN_ERRORS);
>         }
> {noformat}
> while balancing does happen in this thread (in a call to updateStatus before this catch), the try block encompasses a great deal of additional code. so this message can be misleading about where to look for problems. It should be updated to state that there was a problem iwth transitioning states and should probably indicate both that it will retry and how long it will pause before doing so.


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