Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Date: Fri, 27 Nov 2015 07:06:11 +0000 (UTC)
From: "Yu Li (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Message-ID: <JIRA.12916547.1448602500000.207341.1448607971083@Atlassian.JIRA>
In-Reply-To: <JIRA.12916547.1448602500000@Atlassian.JIRA>
References: <JIRA.12916547.1448602500000@Atlassian.JIRA>
 <JIRA.12916547.1448602500596@arcas>
Subject: [jira] [Updated] (HBASE-14891) Add log for uncaught exception in
 RegionServerMetricsWrapperRunnable
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/HBASE-14891?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Yu Li updated HBASE-14891:
--------------------------
    Attachment: HBASE-14891.v2.patch

v2 patch fixes indent and line wrapping following code format

> Add log for uncaught exception in RegionServerMetricsWrapperRunnable
> --------------------------------------------------------------------
>
>                 Key: HBASE-14891
>                 URL: https://issues.apache.org/jira/browse/HBASE-14891
>             Project: HBase
>          Issue Type: Improvement
>          Components: metrics
>            Reporter: Yu Li
>            Assignee: Yu Li
>            Priority: Minor
>         Attachments: HBASE-14891.patch, HBASE-14891.v2.patch
>
>
> We are using ScheduledExecutorService#scheduleWithFixedDelay to periodically reporting metrics. From javadoc of this method
> {quote}
> If any execution of the task encounters an exception, subsequent executions are suppressed.
> {quote}
> We could see if any uncaught exception like NPE thrown in the task thread, there won't be any error thrown but the task would be silently terminated.
> In our real case we were making some modification to the metrics and causing NPE carelessly, then we found metrics disappeared but cannot get a clue from RS log.
> Suggest to catch Throwable in MetricsRegionServerWrapperImpl$RegionServerMetricsWrapperRunnable, add some log there and suppress the exception to give it another try at next scheduled run.


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