activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Dan Rudman (JIRA)" <j...@apache.org>
Subject [jira] Issue Comment Edited: (AMQ-1313) DefaultDatabaseLocker broken on MySQL 4.1
Date Tue, 11 Sep 2007 21:32:22 GMT

    [ https://issues.apache.org/activemq/browse/AMQ-1313?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_40126
] 

dantelope edited comment on AMQ-1313 at 9/11/07 2:32 PM:
----------------------------------------------------------

This is a major, major issue and I am very concerned that it appears to be up for fix on the
5.2.0 release.  As my company will not deploy the 5.x release of ActiveMQ for quite some time
as it is nowhere near stable (or performant, it seems) as 4.1.x, I would like to recommend
that a follow-up release to 4.1.1 (4.1.2?) be made as soon as possible.

Or, at the very least, provide us with a patch to fix this issue.

Thanks for your consideration.

P.S. The defect affects all versions of MySQL.

      was (Author: dantelope):
    This is a major, major issue and I am very concerned that it appears to be up for fix
on the 5.2.0 release.  As my company will not deploy the 5.x release of ActiveMQ for quite
some time as it is nowhere near stable (or performant, it seems) as 4.1.x, I would like to
recommend that a follow-up release to 4.1.1 (4.1.2?) be made as soon as possible.

Or, at the very least, provide us with a patch to fix this issue.

Thanks for your consideration.
  
> DefaultDatabaseLocker broken on MySQL 4.1
> -----------------------------------------
>
>                 Key: AMQ-1313
>                 URL: https://issues.apache.org/activemq/browse/AMQ-1313
>             Project: ActiveMQ
>          Issue Type: Bug
>          Components: Message Store
>    Affects Versions: 4.1.1
>            Reporter: Kenny MacLeod
>             Fix For: 5.2.0
>
>
> AMQ-992 introduced a change in 4.1.1 where the DefaultDatabaseLocker issued a LOCK TABLE
command instead of a SELECT FOR UPDATE.
> This may work with MySQL 5 ( I haven't had the chance to test that yet), but it certainly
doesn't work with MySQL 4.1.  The "LOCK TABLE" statement does not return a result set in 4.1,
so the execute() method returns false, and the locking loop never terminates.
> Given that AMQ-992 was unnecessary (see comments there for the reason why), I suggest
that the change is rolled back to using SELECT FOR UPDATE.

-- 
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