activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rob Davies (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (AMQ-3432) Slave broker able to start together with Master broker (shared file system master/slave setup)
Date Thu, 16 Jan 2014 15:46:21 GMT

     [ https://issues.apache.org/jira/browse/AMQ-3432?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Rob Davies updated AMQ-3432:
----------------------------

    Priority: Major  (was: Critical)

Looks like distributed lock not being honoured by the file system  - need information on the
shared file system to progress this.

> Slave broker able to start together with Master broker (shared file system master/slave
setup)
> ----------------------------------------------------------------------------------------------
>
>                 Key: AMQ-3432
>                 URL: https://issues.apache.org/jira/browse/AMQ-3432
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Message Store
>    Affects Versions: 5.5.0
>         Environment: Windows 2003, Weblogic Application Server 
>            Reporter: PK Tan
>
> When deploying activemq server in shared file system master/slave setup on a cluster
of 2 servers without any of the MQ data files (journal folder, kr-store folder and lock file)
[i.e. a fresh state], the slave broker is able to startup along with the master server. Both
are active as verified by telnetting to the servers at the activemq port .  
> If you were to restart both servers, both will not able to obtain the lock. If you stop
both servers, delete the lock file and restart both the servers, we get back to the initial
state where both servers are started and running (i.e. telnettable).
> This issue might be related to AMQ-3273.  I'm logging another case for it because the
environment and the situation is different . 



--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message