You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@activemq.apache.org by "Matt Pavlovich (Commented) (JIRA)" <ji...@apache.org> on 2011/10/31 20:09:32 UTC

[jira] [Commented] (AMQ-2814) When running multiple brokers on the same box, "activemq-admin stop" command on a downed instance will shutdown other instances

    [ https://issues.apache.org/jira/browse/AMQ-2814?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13140457#comment-13140457 ] 

Matt Pavlovich commented on AMQ-2814:
-------------------------------------

Hi Scott-

The activemq-admin script has an option to point to a specific jmx URL.  I suggest looking at adding the jmx URL as an argument to the activemq-admin program when you execute it.  

There are many different ways folks run multiple ActiveMQ instances installed on a single system, so it is difficult to have those scripts always line up to the correct instance in those scenarios.  The best solution is to align it to the correct JMX URL.

Thanks,
Matt Pavlovich
                
> When running multiple brokers on the same box, "activemq-admin stop" command on a downed instance will shutdown other instances
> -------------------------------------------------------------------------------------------------------------------------------
>
>                 Key: AMQ-2814
>                 URL: https://issues.apache.org/jira/browse/AMQ-2814
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Broker
>    Affects Versions: 5.3.0
>         Environment: Linux 2.6.31-22-generic-pae #60-Ubuntu SMP Thu May 27 01:40:15 UTC 2010 i686 GNU/Linux
>            Reporter: Scott Emmons
>            Priority: Critical
>
> In an environment where multiple ActiveMQ broker servers are running on the same host, the "activemq-admin stop" command will shutdown other instances if the requested instance is already down.
> This is very dangerous behavior in the case where a user inadvertently attempts to shutdown a broker instance which is already down - instead of getting an error message, or the command being a no-op, a different instance than requested will be shutdown instead.
> To reproduce:
> * Install ActiveMQ into SOMEDIR/broker_a and configure port and name.
> * Install ActiveMQ into SOMEDIR/broker_b and configure unique port and name.
> * Start broker_a - broker_a/bin/activemq (or alternatively activemq-admin start)
> * Start broker_b - broker_b/bin/activemq (or alternatively activemq-admin start)
> * Stop broker_a - broker_a/bin/activemq-admin stop
> * Try to stop broker_a again - broker_a/bin/activemq-admin stop -- THIS WILL RESULT IN broker_b SHUTTING DOWN
> If there are three or more instances, the result is the same - "activemq-admin stop" will shutdown instances other than the one requested.

--
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