nitpick: AMQP 1.0 also supports 'named producers'. Thats actually all the protocol supported until an additional capability was documented for doing 'anonymous producer' stuff as e.g needed by some of JMS 1.1. Though its worth noting all producers on the 'simplified api' added in JMS 2.0 are effectively anonymous.
On Fri, 3 Nov 2023 at 15:48, Justin Bertram <jbert...@apache.org> wrote: > > The broker does its best to track producers, but most protocols don't even > have the concept of a "named" producer (i.e. a producer that's registered > on the server to a particular endpoint). Most protocols only support > "anonymous" producers which can send to any endpoint at any time which > means in most circumstances the broker doesn't actually know when a client > is going to send a message (if at all) and when a client does send a > message the broker doesn't know where it's going to go until it actually > arrives. > > The only API I'm aware of that has named producers is JMS and even JMS > supports anonymous producers. Furthermore, of the three protocols used by > JMS clients (i.e. core, OpenWire, and AMQP) only OpenWire supports named > producers. > > In short, there's no real way for the broker to know when there's a > producer "on" an address. > > The situation for consumers is, of course, the complete opposite. The > broker must know about consumers so it can dispatch messages to them. > > Ultimately if an address is removed while a slow producer is still sending > messages to it then it will be re-created automatically if the > configuration supports that. > > Hope that helps! > > > Justin > > On Fri, Nov 3, 2023 at 9:17 AM Dondorp, Erwin <erwin.dond...@ns.nl.invalid> > wrote: > > > Classification: Public > > > > Hello, > > > > Today I noticed that an address was auto-removed while it still had a > > producer active. > > The producer has a very low production rate, its interval was above the > > value of auto-delete-addresses-delay, so the timing constraints were still > > ok. > > Nothing was broken and I can easily increase auto-delete-addresses-delay. > > > > In the source-code, in function "reapAddresses", it looks to me as if > > Artemis does not try to prevent such address-removal with a producer > > present. > > > > Therefore, the question remains: > > Should an address be auto-removed when there still is a producer active? > > > > thx! > > Erwin > > > > ________________________________ > > > > Deze e-mail, inclusief eventuele bijlagen, is uitsluitend bestemd voor > > (gebruik door) de geadresseerde. De e-mail kan persoonlijke of > > vertrouwelijke informatie bevatten. Openbaarmaking, vermenigvuldiging, > > verspreiding en/of verstrekking van (de inhoud van) deze e-mail (en > > eventuele bijlagen) aan derden is uitdrukkelijk niet toegestaan. Indien u > > niet de bedoelde geadresseerde bent, wordt u vriendelijk verzocht degene > > die de e-mail verzond hiervan direct op de hoogte te brengen en de e-mail > > (en eventuele bijlagen) te vernietigen. > > > > Informatie vennootschap<http://www.ns.nl/emaildisclaimer> > >
