[rabbitmq-discuss] what would cause queue size to immediately go to 0?
shane at digitalsanctum.com
Fri Nov 5 16:07:50 GMT 2010
I was using 1.0.0.M1 and now using a nightly snapshot (20101104).
Stepping through the code (ChannelN) I saw that noack is true with code
unchanged. I set container.setChannelTransacted(true); and now noack is
false and I see the desired behavior.
I should mention that I'm using the AmqpOutboundEndpoint in the Spring
Integration sandbox to publish messages from a JdbcInboundAdapter. Should I
be aware of any other changes that would effect this setup?
On Fri, Nov 5, 2010 at 11:51 AM, Dave Syer <david_syer at hotmail.com> wrote:
> On Nov 5, 3:22 pm, Rob Harrop <r... at rabbitmq.com> wrote:
> > Are you using one of the MessageListenerContainer implementations? Also,
> are you using transactions?
> Also are you using 1.0.0.M1 or a more recent snapshot (because there
> is only one flag to set in recent versions - transactional=true/false,
> which we translate into noAck if you request non-transactional
> usage)? It's less moving parts and the semantics for common messaging
> use cases are simpler. NoAck is still the default though, so you need
> to be explicit, and from the code you posted that is not the case.
> I would recommend using a snapshot if you can stand the potential
> instability because we have added full support for rollback since
> rabbitmq-discuss mailing list
> rabbitmq-discuss at lists.rabbitmq.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the rabbitmq-discuss