Hi Janus, Thanks for working on the 1.4.0 release!
For HDDS-9784, it probably is not a blocker since it is about the performance of setting various values of a conf. Users could change the conf as they please. There are trade-offs between system behaviors. It probably won't have a perfect conf satisfying all systems. (Or, if it exists, why not remove the conf?) Tsz-Wo On Tue, Nov 28, 2023 at 12:00 AM Janus Chow <yiyang0...@gmail.com> wrote: > Hi Attila and devs, > > Thank you for the updates. > > We also found another issue related to Datanodes' write performance > downgration, as mentioned in HDDS-9784 > <https://issues.apache.org/jira/browse/HDDS-9784>. > > Let's postpone the branch cut until these 3 tickets resolved. > > > > To fix tickets: > > - https://issues.apache.org/jira/browse/HDDS-9784 > - https://issues.apache.org/jira/browse/HDDS-9709 > - https://issues.apache.org/jira/browse/HDDS-9762 > > > > Warm Regards. > > -- > Yiyang Zhou > > > Attila Doroszlai <adorosz...@apache.org> 于2023年11月28日周二 02:33写道： > > > Hi Yiyang and devs, > > > > In today's US community sync two other bugs targeted at 1.4.0 were > > marked as blocker: > > > > * HDDS-9709. Intermittent NO_REPLICA_FOUND due to OM pipeline cache bug > > * HDDS-9762. Hadoop s3a protocol does not work with FSO buckets > > > > The first one should be quick: it already has a patch available, only > > needs review. > > > > The second one still needs to be investigated, any contribution welcome. > > > > We also discussed the need to publish a 1.3.1 patch release with the > > fix for HDDS-8647, to avoid hitting this bug during upgrade to 1.4.0 > > (HDDS-8711). > > > > -Attila > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: dev-unsubscr...@ozone.apache.org > > For additional commands, e-mail: dev-h...@ozone.apache.org > > > > >
