Mailing-List: contact hbase-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hbase-dev@hadoop.apache.org
Message-ID: <339314475.1228287284186.JavaMail.jira@brutus>
Date: Tue, 2 Dec 2008 22:54:44 -0800 (PST)
From: "stack (JIRA)" <jira@apache.org>
To: hbase-dev@hadoop.apache.org
Subject: [jira] Updated: (HBASE-1042) OOME but we don't abort
In-Reply-To: <1154319851.1228261844213.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


     [ https://issues.apache.org/jira/browse/HBASE-1042?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

stack updated HBASE-1042:
-------------------------

    Attachment: 1042-v2.patch

Patch regularizes exception throwing in HRS.  Changes all IOException catches to Throwable catches so we get a whack at things like OOMEs.

Also removed throw of IOE from HRS#close.  Whats client going to do with an exception on close.

Still need to test.

> OOME but we don't abort
> -----------------------
>
>                 Key: HBASE-1042
>                 URL: https://issues.apache.org/jira/browse/HBASE-1042
>             Project: Hadoop HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: stack
>             Fix For: 0.18.2
>
>         Attachments: 1042-v2.patch, 1042.patch
>
>
> On streamy cluster saw case where graceful shutdown had been triggered rather than an abort on OOME.  On graceful shutdown, we wait on leases to expire or be closed.  Server wouldn't go down because it was waiting on leases to expire only an OOME in Leases had killed the thread so it wasn't ever going to expire anything.   Node was stuck for four hours till someone noticed it.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.