Thanks Wei-Chiu for the supplement. Yeah, I forget this part. On Sat, 7 Jun 2025 at 01:27, Wei-Chiu Chuang <weic...@apache.org> wrote:
> On Fri, Jun 6, 2025 at 1:04 AM Sammi Chen <sammic...@apache.org> wrote: > > > Attenders: Hao, Wei-Chiu, Peter, Pony, Chung en, Sammi > > > > Hao: > > - Has a plan to upgrade the cluster to Java 21/17. > > - Has a plan to support multi-AZ aware ozone clusters. > > - Found one root cause of a follower SCM failed to start. A > addContainer() > > is called for a closed pipeline. > > - long lived RpcClient in s3g suffers memory due to no upper size > > limitation of used ElasticByteBufferPool. > > Sammi: - HDDS-10239-container-reconciliation merge vote is going on in > the > > community. > > Wei-Chiu: > > - snapshot scalability enhancement, to support 10k snapshots or even > more. > > - working on Ozone document improvements, adding new documents, and > > updating old documents to have up-to-date content, under the help of AI > > tools. > > Pony: > > - Object TTL new PR "HDDS-12783. Implement S3 Gateway api for > > LifeCycleConfiguration"(https://github.com/apache/ozone/pull/8557) > > - Datanode startup time optimization. It took 150s to read 10k container > > yaml files during datanode restart. A proposal is save all in memory > > container infos into one file as container yaml content cache during > > Datanode stop, and read this file when Datanode starts. > > Datanode container rocksdb compaction issue was discussed. > We discussed whether it's possible to adopt > https://issues.apache.org/jira/browse/HDDS-12819 for DN > and that other fixed commits under > https://issues.apache.org/jira/browse/HDDS-12189 could be useful too. > > > Peter: Working on POC of "HDDS-12712. Design Document of "Aggressive DB > > Compaction with Minimal Degradation dependencies" > > Chung En: Need help on review "HDDS-5984. S3 Event Notification design > doc" > > (https://github.com/apache/ozone/pull/8449) > > >
