Asaf, would you like to send a PR ? Enrico
Il giorno lun 23 mag 2022 alle ore 13:38 PengHui Li <peng...@apache.org> ha scritto: > > I support enabling it in 2.11.0 > > Without the `Acknowledgement in Batch Level index` > Users will see a weird behavior `the acked messages dispatched to the > consumer again` > Many new users thought it was a bug. > > For the default configuration, we should be more friendly to new users, > because they have too much pulsar experience. For pulsar veterans, they can > disable it to reduce memory consumption, but they know what the expected > behavior is > after diable it. > > And regarding memory consumption, we can optimize in later versions. > > Thanks, > Penghui > > > On Mon, May 23, 2022 at 5:54 PM Enrico Olivelli <eolive...@gmail.com> wrote: > > > Asaf, > > I don't have numbers to say something about the runtime impact. > > > > It is a important feature and I am +1 to enabling it by default > > > > You can send a PR in the meantime. > > I am not sure about the impact on the tests, maybe we will have to > > take a look carefully in order to not introduce new flaky tests > > > > Enrico > > > > Il giorno dom 22 mag 2022 alle ore 18:11 Asaf Mesika > > <asaf.mes...@gmail.com> ha scritto: > > > > > > Resurfacing this as it seems (to me) an important correction to be made. > > > > > > Would love your opinions on this. > > > > > > Thanks! > > > > > > Asaf > > > > > > On Wed, May 11, 2022 at 1:16 PM Asaf Mesika <asaf.mes...@gmail.com> > > wrote: > > > > > > > Hello all, > > > > > > > > I have a question regarding a feature introduced in 2.6.0 called > > Negative > > > > Acknowledgement in Batch Level index, described in PIP-54 > > > > < > > https://github.com/apache/pulsar/wiki/PIP-54:-Support-acknowledgment-at-batch-index-level > > > > > > > . > > > > > > > > In short: Before this feature, if you would write in batches (in > > producer) > > > > to Pulsar (which makes sense most would for performance reasons), then > > a > > > > consumer sending a negative acknowledgment for 1 message in the batch > > (say > > > > we have 500 in a single batch), would cause the consumer to get the all > > > > messages in the batch (500) redelivered as if they were not > > acknowledge. > > > > > > > > PIP-54 fixes that by keeping the ack per message in a batch using a > > bitmap > > > > index. The caveats are more memory consumption since the broker keeps > > those > > > > bitmaps in-memory for any inflight batch. > > > > > > > > With PIP-54 Pulsar IMO becomes "complete" in that it acts the way you > > > > would expect it to be, in normal circumstances (as opposed to > > disconnects, > > > > machine crashing, etc). > > > > > > > > This feature was introduced in 2.6.0, roughly 2 years ago, and is off > > by > > > > default. > > > > > > > > I was wondering what the developers community thoughts on turning it > > on by > > > > default? Has anyone experienced any performance degradation to it? > > Have you > > > > turned it on in your clusters? > > > > > > > > Thanks a lot for your time. > > > > > > > > Asaf Mesika > > > > > >
