Thanks, Ryanne. Can you add information about this way forward to the KIP? Also it would be good to clarify that this work needs to get done before removing MM1.
best, Colin On Thu, May 20, 2021, at 16:00, Ryanne Dolan wrote: > Hey y'all, reviving this thread because it seems we have a way forward > w.r.t. IdentityReplicationPolicy aka LegacyReplicationPolicy, which I > believe is the only missing feature in MM2 that we need to deprecate MM1. > > If there are no objections over the next couple of days I'll consider this > adopted. Thanks! > > Ryanne > > On Fri, Apr 2, 2021 at 10:48 AM Colin McCabe <cmcc...@apache.org> wrote: > > > Hi Ryanne, > > > > Thanks for the response. It would be good to have a PR for KIP-382, I > > agree. > > > > Perhaps one possible compromise for KIP-712 would be to make the changes > > in MM2 first, and then backport them to MM1. I think it's important that > > when we have a deprecated way of doing something and a non-deprecated way, > > the non-deprecated way is the recommended way. If we are onboarding more > > users to the deprecated code path (for example, because there's major > > features missing in the new code path), we're doing something wrong. > > > > best, > > Colin > > > > > > On Thu, Apr 1, 2021, at 15:48, Ryanne Dolan wrote: > > > Colin, the only feature gap I'm aware of is that users must provide their > > > own ReplicationPolicy in order to replicate topics without renaming them. > > > This is straightforward, and such ReplicationPolicy implementations are > > > easy to find. We could provide one OOTB, and indeed KIP-382 proposes we > > do > > > so, but the PR is missing. I'm happy to pick that up, no problem. > > > > > > wrt KIP-712, the changes are immediately applicable to MM2, at least as > > it > > > is currently written. I have no dog in the fight wrt whether the changes > > > also land in MM1, but, assuming both KIPs land concurrently in 3.0, I > > don't > > > see why the two KIPs would be in conflict. Obvs, this KIP marks MM1 as > > > deprecated going forward, but I don't think that precludes a concurrent > > > improvement. > > > > > > If KIP-712 were being proposed after 3.0, I'd agree with you. > > > > > > I think the reality is that MM1 has been sort of unofficially deprecated > > > for a long time, so people are understandably disinterested in landing > > new > > > features there. But let's have that debate in the KIP-712 thread. I > > believe > > > we'd be having the same discussion there with or without KIP-720 passing. > > > > > > Ryanne > > > > > > On Thu, Apr 1, 2021, 2:07 PM Colin McCabe <cmcc...@apache.org> wrote: > > > > > > > Thanks for bringing this up, Ismael. I agree that we need to figure > > this > > > > out before we accept this KIP. > > > > > > > > If MM1 is deprecated, then that means we are telling users they need to > > > > migrate away from it as soon as they can. I think that rules out > > adding > > > > big new features to MM1, unless those features relate towards > > migrating to > > > > MM2. So we need to figure out if that's really what we want to do, or > > if > > > > we want to keep MM1 around for a while. This is certainly relevant to > > the > > > > discussion in the KIP-712 thread -- right now, these KIPs contradict > > each > > > > other. > > > > > > > > It's also important that MM2 reaches feature parity with MM1 before > > > > deprecating MM1. Or if we can't reach feature parity, we should > > explain > > > > why the unsupported features are not needed going forward. Do we have > > a > > > > list of all the gaps? > > > > > > > > best, > > > > Colin > > > > > > > > > > > > On Thu, Apr 1, 2021, at 09:44, Ismael Juma wrote: > > > > > OK. :) Maybe something like: > > > > > > > > > > "We believe MirrorMaker 2 is an improvement over the original > > MirrorMaker > > > > > when it comes to reliability and functionality for the majority of > > use > > > > > cases. We intend to focus on MirrorMaker 2 for future development and > > > > hence > > > > > we propose deprecating MirrorMaker 2 for future removal." > > > > > > > > > > Is this accurate? How does it sound? > > > > > > > > > > Ismael > > > > > > > > > > On Thu, Apr 1, 2021 at 9:10 AM Ryanne Dolan <ryannedo...@gmail.com> > > > > wrote: > > > > > > > > > > > Ah, do you mind wording it for me, Ismael? Or do you mean I should > > just > > > > > > remove the "MM1 is still useful" part? > > > > > > > > > > > > Ryanne > > > > > > > > > > > > On Thu, Apr 1, 2021, 11:01 AM Ismael Juma <ism...@juma.me.uk> > > wrote: > > > > > > > > > > > > > Can we please add proper motivation? I'm -1 with the current > > > > motivation > > > > > > > even though I'm in favor of the change. > > > > > > > > > > > > > > On Thu, Apr 1, 2021, 8:46 AM Ryanne Dolan <ryannedo...@gmail.com > > > > > > > wrote: > > > > > > > > > > > > > > > Hey y'all, looks like we've got the requisite votes for this to > > > > pass, > > > > > > and > > > > > > > > the various concerns wrt KIP-712 are now being discussed on > > that > > > > > > thread. > > > > > > > So > > > > > > > > I'm going to go ahead and close the vote here. > > > > > > > > > > > > > > > > Thanks for the votes! > > > > > > > > > > > > > > > > Ryanne > > > > > > > > > > > > > > > > On Fri, Mar 26, 2021, 11:26 PM Ismael Juma <ism...@juma.me.uk> > > > > wrote: > > > > > > > > > > > > > > > > > It does mean more than that. We don't remove or replace > > things in > > > > > > > Apache > > > > > > > > > Kafka without good reasons (since it's typically costly for > > > > users). > > > > > > And > > > > > > > > > once something is scheduled for removal, it's typically in > > > > > > maintenance > > > > > > > > mode > > > > > > > > > and only bug fixes are expected. > > > > > > > > > > > > > > > > > > Ismael > > > > > > > > > > > > > > > > > > On Fri, Mar 26, 2021, 8:28 PM Ryanne Dolan < > > > > ryannedo...@gmail.com> > > > > > > > > wrote: > > > > > > > > > > > > > > > > > > > Ismael, "deprecated" implies something is scheduled to be > > > > removed > > > > > > or > > > > > > > > > > replaced, but I don't think it implies anything more than > > that. > > > > > > > KIP-720 > > > > > > > > > is > > > > > > > > > > proposing to deprecate MM1 so it can eventually be removed. > > > > That's > > > > > > > all > > > > > > > > > this > > > > > > > > > > particular KIP is proposing. > > > > > > > > > > > > > > > > > > > > Ryanne > > > > > > > > > > > > > > > > > > > > On Fri, Mar 26, 2021, 7:24 PM Ismael Juma < > > ism...@juma.me.uk> > > > > > > wrote: > > > > > > > > > > > > > > > > > > > > > Thanks Tom, this is a good elaboration on what I meant. > > > > Also, if > > > > > > > it's > > > > > > > > > > > deprecated, then we should definitely not be adding > > features. > > > > > > I'm a > > > > > > > > > > puzzled > > > > > > > > > > > that we are saying that MM1 is useful, deserves > > additional > > > > > > > > development > > > > > > > > > > and > > > > > > > > > > > should be deprecated - all at the same time. > > > > > > > > > > > > > > > > > > > > > > Ismael > > > > > > > > > > > > > > > > > > > > > > On Fri, Mar 26, 2021, 9:20 AM Tom Bentley < > > > > tbent...@redhat.com> > > > > > > > > wrote: > > > > > > > > > > > > > > > > > > > > > > > Hi Ryanne, > > > > > > > > > > > > > > > > > > > > > > > > With respect, there's a difference between "we still > > use it > > > > > > > because > > > > > > > > > we > > > > > > > > > > > > can't be bothered to switch to MM2, or just haven't > > yet" > > > > and > > > > > > > "it's > > > > > > > > > > > > important for xyz because MM2 doesn't serve our use > > case > > > > > > > properly". > > > > > > > > > > While > > > > > > > > > > > > the former is not a good reason to argue against > > > > deprecation, > > > > > > the > > > > > > > > > > latter > > > > > > > > > > > > might be, depending on the details. It isn't completely > > > > clear > > > > > > to > > > > > > > me > > > > > > > > > > > whether > > > > > > > > > > > > you're asserting that MM2 covers all the same use > > cases as > > > > MM1. > > > > > > > On > > > > > > > > > the > > > > > > > > > > > one > > > > > > > > > > > > hand you don't want to make claims, but on the other > > you're > > > > > > > saying > > > > > > > > we > > > > > > > > > > > have > > > > > > > > > > > > MM2 now. An assertion that MM2 addressed all the MM1 > > use > > > > cases > > > > > > > > would > > > > > > > > > be > > > > > > > > > > > > Ismael's explanation about why MM1 is no longer > > useful, I > > > > > > think. > > > > > > > > OTOH > > > > > > > > > > the > > > > > > > > > > > > KIP says it is still useful. So personally I'm confused > > > > about > > > > > > > which > > > > > > > > > the > > > > > > > > > > > > situation is. > > > > > > > > > > > > > > > > > > > > > > > > Are we deprecating something which for some users MM2 > > > > cannot > > > > > > > > replace? > > > > > > > > > > If > > > > > > > > > > > > so, I think the KIP should explain clearly why we're > > > > > > > intentionally > > > > > > > > > > doing > > > > > > > > > > > > that. > > > > > > > > > > > > > > > > > > > > > > > > Kind regards, > > > > > > > > > > > > > > > > > > > > > > > > Tom > > > > > > > > > > > > > > > > > > > > > > > > On Fri, Mar 26, 2021 at 3:22 PM Ryanne Dolan < > > > > > > > > ryannedo...@gmail.com> > > > > > > > > > > > > wrote: > > > > > > > > > > > > > > > > > > > > > > > > > Ismael, I think it is very difficult in general to > > argue > > > > for > > > > > > > > > > > deprecation > > > > > > > > > > > > -- > > > > > > > > > > > > > someone will always say "we still use it" or "it's > > > > important > > > > > > > for > > > > > > > > > xyz" > > > > > > > > > > > -- > > > > > > > > > > > > so > > > > > > > > > > > > > I don't want to make claims that prompt such > > responses. > > > > The > > > > > > > > > > motivation > > > > > > > > > > > > for > > > > > > > > > > > > > deprecating MM1 is that we now have MM2, and there > > isn't > > > > much > > > > > > > > else > > > > > > > > > to > > > > > > > > > > > > say, > > > > > > > > > > > > > IMO. > > > > > > > > > > > > > > > > > > > > > > > > > > Ryanne > > > > > > > > > > > > > > > > > > > > > > > > > > On Mon, Mar 22, 2021 at 8:04 AM Ismael Juma < > > > > > > ism...@juma.me.uk > > > > > > > > > > > > > > > > > > wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > I am in favor of this change, but the KIP doesn't > > > > include > > > > > > > > proper > > > > > > > > > > > > > > motivation. It says "While the original MirrorMaker > > > > remains > > > > > > > > > useful, > > > > > > > > > > > we > > > > > > > > > > > > > want > > > > > > > > > > > > > > to take advantage of the upcoming 3.0 major > > release to > > > > > > > > officially > > > > > > > > > > > > > deprecate > > > > > > > > > > > > > > this legacy code". I would hope we would explain > > why > > > > it's > > > > > > no > > > > > > > > > longer > > > > > > > > > > > > > useful > > > > > > > > > > > > > > instead. > > > > > > > > > > > > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > > > > > > > Ismael > > > > > > > > > > > > > > > > > > > > > > > > > > > > On Sat, Mar 20, 2021 at 10:41 AM Ryanne Dolan < > > > > > > > > > > ryannedo...@gmail.com > > > > > > > > > > > > > > > > > > > > > > > > > > wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Hey y'all, I'm starting the vote on KIP-720, > > which > > > > > > proposes > > > > > > > > to > > > > > > > > > > > > > deprecate > > > > > > > > > > > > > > > the original MirrorMaker in the upcoming 3.0 > > major > > > > > > release. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > https://cwiki.apache.org/confluence/display/KAFKA/KIP-720%3A+Deprecate+MirrorMaker+v1 > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > Thanks! > > > > > > > > > > > > > > > Ryanne > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >
