qpid-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sergey_Zhemzhit...@troika.ru
Subject Re: QPID Reliability
Date Fri, 04 Feb 2011 09:32:21 GMT
Hi Alan,

First of all thanks for helping to understand the meaning of these queues.
How long the broker will keep those queues? Because if they exist until 
the broker restarts, it will be difficult to find a concrete queue because 
of amount of _@amq.failover_xxxxxxxx.


Best Regards,
Sergey Zhemzhitsky




Alan Conway <aconway@redhat.com> 
03.02.2011 22:08
Please respond to
users@qpid.apache.org


To
users@qpid.apache.org
cc
Sergey_Zhemzhitsky@troika.ru
Subject
Re: QPID Reliability






On 02/03/2011 08:30 AM, Sergey_Zhemzhitsky@troika.ru wrote:
> Jonathan, thanks for help.
>
> Those are queues, not exchanges.
> There are a lot of queues like that
> _@amq.failover_54cb1666-c325-472b-9b83-1bececd2f0a8 when I'm executing
> qpid-stat -q.
> It seems that they are created automatically?
>
> When client starts for the first time there are no  such queues. They 
are
> created automatically when broker's journal overflow happens. When the
> journal is back to normal I can still see thouse queues.
>
>

These queues are created automatically by clients. They are bound to the 
amq.failover exchange to receive updates on cluster membership, so that 
clients 
have an up-to-date picture of the brokers they can use for failover. They 
are 
very low-traffic (one message each time a broker joins or leaves the 
cluster) 
and are not stored persistently. They should not have any noticeable 
impact on 
performance of your application.

---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project:      http://qpid.apache.org
Use/Interact: mailto:users-subscribe@qpid.apache.org



Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message