activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gary Tully (JIRA)" <j...@apache.org>
Subject [jira] Resolved: (AMQ-1885) The jdbc lockKeepAlive thread should be started only if current broker instance successfully started and became a master
Date Thu, 14 Aug 2008 18:37:52 GMT

     [ https://issues.apache.org/activemq/browse/AMQ-1885?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Gary Tully resolved AMQ-1885.
-----------------------------

    Resolution: Fixed

fixed in revision 685988. 
Added two jdbc master slave tests cases, normal failover after master dies and failover after
temporary db outage.

> The jdbc lockKeepAlive thread should be started only if current broker instance successfully
started and became a master
> ------------------------------------------------------------------------------------------------------------------------
>
>                 Key: AMQ-1885
>                 URL: https://issues.apache.org/activemq/browse/AMQ-1885
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Broker
>    Affects Versions: 5.1.0
>         Environment: jdbc persistence adapter with master/slave
>            Reporter: Gary Tully
>            Assignee: Gary Tully
>             Fix For: 5.2.0
>
>
> using jdbc persistence with master/slave sharing the database. If the db dies, the master
dies when it sees its lock expire. unfortunately the slave also has a lockKeepalive thread
that detects the db failure, even if it does not have the lock.
> The lockKeepAlive thread should be used once a lock has been acquired. in this way, the
slave can remain trying to acquire the lock and eventually succeed and promote its self to
master when the db comes back up.

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


Mime
View raw message