I remember that thread..
but I think in most cases primary / backup makes more sense... But I don't mind which term we choose TBH... IMO we should just stick to primary / backup, but if somewhere specifically leader / follower makes more sense. .why not? I would leave it at the discression of the person implementing the change. When you get your hands on it makes more sense. @JB If you send a Pull Request and want an extra pair of eyes to make sure on the changes.. let me know on this thread and i will help reviewing it. On Tue, Nov 10, 2020 at 1:27 PM Christopher Shannon <[email protected]> wrote: > > There was already another thread on this topic along with a Jira: > http://activemq.2283324.n4.nabble.com/DISCUSS-Draft-proposal-for-terminology-change-td4758351.html > https://issues.apache.org/jira/browse/AMQ-7514 > > New terms were already somewhat decided in that thread as primary/backup > doesn't make sense in all cases. It depends on what the application is > (leader/follower, etc) > > On Tue, Nov 10, 2020 at 12:05 PM Jean-Baptiste Onofre <[email protected]> > wrote: > > > Hi, > > > > I agree with the terms (I think we have kind of consensus). > > > > I will start the change on ActiveMQ side (as I’m working on new releases > > and updates). > > > > Regards > > JB > > > > > Le 10 nov. 2020 à 17:26, Clebert Suconic <[email protected]> a > > écrit : > > > > > > What about this... lets propose the following changes: > > > > > > - master should become primary (we could refer to it as primary server > > in docs) > > > - slave should become backup (same way, we could refer to it as backup > > > server in docs) > > > - whitelist: allowlist > > > - blacklist: denylist > > > > > > TBH: master and slave are the most used words among the list, on both > > > activemq and artemis codebase. > > > > > > > > > I'm working with my company (Red Hat) to allow time from someone on > > > our team to work on this, and I believe we can set up someone > > > dedicated to it early 2021 on the ActiveMQ Artemis codebase. > > > > > > We still need volunteers to do it on the ActiveMQ codebase.... > > > > > > > > > In regard to the list of names, I am not particularly strongly > > > opinionated with the terms.. but if someone is, please suggest a > > > different term to the list. > > > > > > > > > > > > > > > > > > > > > > > > On Mon, Nov 9, 2020 at 3:38 PM Rich Bowen <[email protected]> wrote: > > >> > > >> > > >> > > >> On 2020/11/05 17:34:25, Clebert Suconic <[email protected]> > > wrote: > > >>> *My* particular issue around this was not knowing what to do with > > >>> configuration parameters and APIs. > > >>> > > >>> If we simply remove those, older clients, older configs would not > > work any > > >>> longer. > > >>> > > >>> Is deprecation here a valid approach? Is there consensus around it ? > > >> > > >> Yes, we definitely recommend that you have a published deprecation > > plan, so that there's sufficient warning, and you don't break existing > > installations. Exactly what that timing is, is going to vary a great deal > > from one project to another, and only you and your users can figure that > > out. > > > > > > > > > > > > -- > > > Clebert Suconic > > > > -- Clebert Suconic
