[rabbitmq-discuss] Fwd: cluster & HA (not Mirrored Queue Behaviour)

Michael Klishin mklishin at gopivotal.com
Tue Jul 30 14:12:53 BST 2013


Adding back rabbitmq-discuss

> De: 池凱琳/Kalin Chih <kalinchih at gmail.com>
> 
> Hi MK,
> 
> Many thanks for your prompt reply.
> 
> I does not use the rabbitmq logstash plugin. We just use the logstash to send message to rabbitmq via layer 4 switch.
> 
> So if I want to distribute/share the messages from producers, is this plan OK?
>  
> 
> - cluster 1
> - node 1: disc
> - node 2: disc
> - cluster 2
> - node 3: disc
> - node 4: disc
> 
> - layer 4 switch binds all nodes.
> - producers send messages to layer 4 switch
> - consumers poll messages to layer 4 switch

This setup is OK. Performance characteristics are really workload-specific so it's a good
idea to start with all disk nodes and then see if you really need to use RAM nodes.

You've two clusters listed. Are you going to use shovel or federation to connect them?
Or how it's going to work?
--
MK

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20130730/83b150e8/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 495 bytes
Desc: Message signed with OpenPGP using GPGMail
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20130730/83b150e8/attachment.pgp>


More information about the rabbitmq-discuss mailing list