activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gary Tully <gary.tu...@gmail.com>
Subject Re: Dropping pure master/slave support from 5.8
Date Thu, 08 Nov 2012 14:13:08 GMT
I agree, it has not had much love over the past few releases and the
implementation over a single channel will never scale, + there is the
recovery problem.

The theory is great though, no infrastructure required. We can maybe
revisit this "feature" with a replicated memory store at some stage.

+1

It does keep popping up on the user list, so some folks will be
surprised, but it is better that they go down the shared storage road
from the start.

On 8 November 2012 14:02, Hiram Chirino <hiram@hiramchirino.com> wrote:
> How do you guys feel about dropping pure master/slave support from 5.8?
>  Pure master/slave adds lots of complexity to the broker implementation yet
> I've never been able to recommend anyone use it in production due to
> it's limitations.  M/S based on shared storage is fast, and most
> importantly very reliable.  So I think we clean house a remove this
> 'feature'.
>
>
> --
>
> **
>
> *Hiram Chirino*
>
> *Engineering | Red Hat, Inc.*
>
> *hchirino@redhat.com <hchirino@redhat.com> | fusesource.com | redhat.com*
>
> *skype: hiramchirino | twitter: @hiramchirino<http://twitter.com/hiramchirino>
> *
>
> *blog: Hiram Chirino's Bit Mojo <http://hiramchirino.com/blog/>*



-- 
http://redhat.com
http://blog.garytully.com

Mime
View raw message