You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by Doug Steigerwald <ds...@mcclatchyinteractive.com> on 2007/06/07 18:16:36 UTC
Has anyone experienced major kaha failure?
Greetings.
We're leaning heavily towards kaha for our persistent storage. Has anyone had any major issues with
data corruption or total kaha failure when ActiveMQ crashes (if/when it does, or for any other reason).
We're just running through some DR scenarios.
--
Doug Steigerwald
Software Developer
McClatchy Interactive
dsteigerwald@mcclatchyinteractive.com
919.861.1287
Re: Has anyone experienced major kaha failure?
Posted by Tom Samplonius <to...@samplonius.org>.
I have. Just "kill -9" the java process to simulate a power failure. Try various producer and consumer combinations.
In certain circumstances ActiveMQ will not be able to auto-recover, because it tries to access a file that should not be there (ie data--2). There is a bug in the code, where an extra hyphen is added to the pathname (it should be data-2) in some cases.
----- "Doug Steigerwald" <ds...@mcclatchyinteractive.com> wrote:
> Greetings.
>
> We're leaning heavily towards kaha for our persistent storage. Has
> anyone had any major issues with
> data corruption or total kaha failure when ActiveMQ crashes (if/when
> it does, or for any other reason).
>
> We're just running through some DR scenarios.
>
> --
> Doug Steigerwald
> Software Developer
> McClatchy Interactive
> dsteigerwald@mcclatchyinteractive.com
> 919.861.1287