Mailing-List: contact dev-help@falcon.incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@falcon.incubator.apache.org
Date: Tue, 30 Sep 2014 13:43:33 +0000 (UTC)
From: "Shaik Idris Ali (JIRA)" <jira@apache.org>
To: dev@falcon.incubator.apache.org
Message-ID: <JIRA.12744952.1412084562000.155116.1412084613614@Atlassian.JIRA>
In-Reply-To: <JIRA.12744952.1412084562000@Atlassian.JIRA>
References: <JIRA.12744952.1412084562000@Atlassian.JIRA>
 <JIRA.12744952.1412084562853@arcas>
Subject: [jira] [Created] (FALCON-765) Add TTL to falcon.late.queue and
 check the late-cutoff before retrying after dequeue
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit

Shaik Idris Ali created FALCON-765:
--------------------------------------

             Summary: Add TTL to falcon.late.queue and check the late-cutoff before retrying after dequeue
                 Key: FALCON-765
                 URL: https://issues.apache.org/jira/browse/FALCON-765
             Project: Falcon
          Issue Type: Bug
            Reporter: Shaik Idris Ali
            Assignee: Shaik Idris Ali


Currently, we do not have TTL set on falcon.late.queue used for late-data processing.

Ideally, the queue should be continuously processed by the Falcon LateConsumer thread, but in case the thread is stuck, the messages are just stuck in JMS (activeMQ) increasing the load on engine.

Also, when the thread becomes live after the quite a long stuck period it should simply discard the messages after dequeuing if the late-cutoff is expired. Otherwise very old messages might be dequeued and Falcon simply starts rerun of old those process instances.


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