michaelpearce-gain commented on issue #2850: ARTEMIS-2504 implement retroactive addresses URL: https://github.com/apache/activemq-artemis/pull/2850#issuecomment-536613138 @jbertram re: "My original plan was to have the ring queues for the retroactive address under the main address itself, but I didn't want those queues to be subject to the address-settings of that address. Separating the ring queues onto their own address means better configuration flexibility." The issue here are 1) you have to message copy, not just make new references - expensive if you have a large retro active to get through, also will cause paging faults to occur as copies occur, even if consumer isnt at that message. 2) its going to be very messy as an end user to make address settings for the internal's, especially as typically (as an end user, i probably just want the same settings for bit like paging etc) 3) address settings do not update the queue (in general they are defaults, overrides and actuals are always set at queue level), as such once queue is created its not going to be possible to adjust Also having it at the address allows us to, and the way i was seing this working so then you can set this being better configurable and working nicely is to have these defined as sepecial queues under the address (eg. in broker.xml) <address name="MyAddress"> <multicast> <queue name="MyQueue" group-buckets="64" group-rebalance="true"> <filter string="ServiceId = 1" /> <durable>true</durable> </queue> <retro-active-queue group-buckets="0" > <message-count>1024</message-count> </queue> </multicast> </address> This way then it can expose any queue settings, easily.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
