Mailing-List: contact issues-help@camel.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@camel.apache.org
Date: Mon, 9 Jul 2012 09:14:35 +0000 (UTC)
From: "Claus Ibsen (JIRA)" <jira@apache.org>
To: issues@camel.apache.org
Message-ID: <1678977791.22018.1341825275026.JavaMail.jiratomcat@issues-vm>
In-Reply-To: <1892100363.22017.1341825274963.JavaMail.jiratomcat@issues-vm>
Subject: [jira] [Commented] (CAMEL-5433) camel-hdfs - Allow to startup async
 as hdfs file system has hardcoded 15 min retry in case no connection
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/CAMEL-5433?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13409279#comment-13409279 ] 

Claus Ibsen commented on CAMEL-5433:
------------------------------------

Looking in the hadoop-core source code it seems that re-connect logic is hardcoded to 45 max attempts with 20 sec delay
{code}
/* The max number of retries is 45,
           * which amounts to 20s*45 = 15 minutes retries.
           */
          handleConnectionFailure(timeoutFailures++, 45, toe);
{code}

                
> camel-hdfs - Allow to startup async as hdfs file system has hardcoded 15 min retry in case no connection
> --------------------------------------------------------------------------------------------------------
>
>                 Key: CAMEL-5433
>                 URL: https://issues.apache.org/jira/browse/CAMEL-5433
>             Project: Camel
>          Issue Type: Improvement
>          Components: camel-hdfs
>    Affects Versions: 2.10.0
>            Reporter: Claus Ibsen
>            Assignee: Claus Ibsen
>             Fix For: 2.10.1, 2.11
>
>
> When start a Camel app with camel-hdfs (either as producer or consumer), then getting a hdfs file system object may have Hadoop block for up till 15 minutes where it tries to obtain a connection to the cluster.
> This makes the app appear as blocking and people dont want their apps to block for 15 min on startup.
> We should let the hadoop hdfs connection happing in an async thread, so the app can startup. Configurable so you can turn back the old behavior.
> Hadoop dosn't allow to configure a lower timeout etc this as its hardcoded.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira