+1 Thanks Kamal for the KIP. Discussed offline with Kamal on the proposal and the changes. It improves the performance of consumers' remote reads in read_committed mode. It brings minor changes in a backward compatible manner.
~Satish. On Mon, 4 Nov 2024 at 19:33, Divij Vaidya <divijvaidy...@gmail.com> wrote: > +1 > > (I have participated in the review and I think we have a good solution in > the KIP) > > -- > Divij Vaidya > > > > On Mon, Nov 4, 2024 at 5:41 AM Kamal Chandraprakash < > kamal.chandraprak...@gmail.com> wrote: > > > Hi all, > > > > Bumping up the voting thread. > > > > On Mon, Sep 30, 2024 at 3:27 PM Francois Visconte > > <francois.visco...@datadoghq.com.invalid> wrote: > > > > > Could we vote on this? This is causing a bunch of tiered storage read > > > issues as many consumers default to READ_COMMITTED (eg. librdkafka) > > > > > > Thanks, > > > F. > > > > > > On Mon, Sep 16, 2024 at 7:20 AM Kamal Chandraprakash < > > > kamal.chandraprak...@gmail.com> wrote: > > > > > > > Bumping this thread for vote. PTAL. > > > > > > > > On Mon, Sep 9, 2024 at 2:01 PM Kamal Chandraprakash < > > > > kamal.chandraprak...@gmail.com> wrote: > > > > > > > > > Hi all, > > > > > > > > > > I'd like to open voting for KIP-1058. This KIP improves the > consumer > > > > > reading from remote storage when READ_COMMITTED isolation level is > > > > enabled. > > > > > PTAL. > > > > > > > > > > KIP-1058 > > > > > < > > > > > > > > > > https://cwiki.apache.org/confluence/display/KAFKA/KIP-1058%3A+Txn+consumer+exerts+pressure+on+remote+storage+when+collecting+aborted+transactions > > > > > > > > > > > > > > > Thanks, > > > > > Kamal > > > > > > > > > > > > > > >
