📝 5 新 Issues
✅ 6 关闭
✨ 2 新 PRs
🎉 8 合并
🔥 核心要点
- 🎉 PR #3054 优化 getNullCounts() 返回 int[] 替代 Long[]，提升性能
- 🎉 PR #3032 添加 Flink filter pushdown 集成和文档支持
- 🎉 PR #3053 修复跨 poll 边界的过期 fetch 检测问题
- 🎉 PR #3050 引入 netty.server.max-request-size 配置限制请求大小
- ✨ PR #3067 支持 S3 委派的默认 AWS 凭证链，解决 IRSA/IAM 角色使用问题
⚠️ 重点关注
- ⚠️ Issue #3063 不稳定测试 RebalanceManagerITCase.testBuildClusterModel
- ⚠️ Issue #2992 不稳定测试 RemoteLogScannerITCase.testScanFromRemoteAndProject
💬 Issue/PR 动态
Issue 讨论
Issue #2669 [docs] Add Real-Time User Profile quickstart tutorial
- @platinumhamburg: 指出 Java Faker 的 numberBetween 参数问题，生成的值范围与预期不符
- @Prajwal-banakar: 感谢评论，已修复问题并请求再次审阅
- @polyzos: 建议 rebase 到 master 分支以修复文档构建失败问题
Issue #3022 [Blog] realtime profiles blog post
- @platinumhamburg: 建议统一使用 flink-roaringbitmap 仓库作为未来发布的标准基线
- @polytos: 说明当前工作量较大，等 Fluss 直接引入功能后再统一标准
Issue #2836 [FLUSS-2686] Add COS filesystem support
- @XuQianJin-Stars: 已更新 PR，请求审阅者帮忙 review
- @leekeiabstraction: 指出配置项名称不一致，并询问 region 配置缺失问题
Issue #3026 [client] Replace Netty PooledByteBufAllocator with bump-pointer ChunkedAllocationManager for Arrow memory allocation.
- @loserwang1024: 解释选择当前方案的原因：避免 OOM 同时减少内存分配开销
Issue #3067 [filesystem] Support default AWS credential chain for S3 delegation
- @fresh-borzoni: 请求审阅者关注这个针对 IRSA 支持的修复方案
- @affo: 感谢贡献，将在明天早上进行完整审阅
PR Review
PR #3053 Fix stale fetch detection across poll boundaries
- @Copilot: 建议警告日志记录两个 offset 值避免诊断时误导
PR #2836 [FLUSS-2686] Add COS filesystem support
- @leekeiabstraction: 指出 fs.cosn.endpoint 与 PR 描述不一致，询问正确的配置名称
- @leekeiabstraction: 询问为何不在此 PR 中实现 STS 生成 token 功能
PR #3042 [spark] Support batch union read for lake-enabled primary key tables
- @fresh-borzoni: 指出 scanner 在构造函数中订阅 log 会导致 PartitionNotExistException
📝 新建 Issue/PR
Issues
- #3068 [client] Add validation for client.scanner.log.max-poll-records and client.connect-timeout config options @Prajwal-banakar
- #3066 [filesystem] S3 credential delegation requires static keys, blocking IRSA/IAM role usage @fresh-borzoni
- #3064 [Feature] Support secondary index on `user_id` / (`user_id`, `order_date`) for detail tables used by Flink SQL @c8679724
- #3063 [test] Unstable test RebalanceManagerITCase.testBuildClusterModel @app/github-actions
- #3061 [flink] CI compilation failure in FlinkTableSourceFilterPushDownTest after #3054 @platinumhamburg
Pull Requests
- #3067 [filesystem] Support default AWS credential chain for S3 delegation @fresh-borzoni
- #3059 [metrics] Fix pendingRecords metric to return actual row count for primary key tables when lake tiering hasn't started @beryllw
✅ 关闭 Issue/PR
已关闭 Issues
- #3061 [flink] CI compilation failure in FlinkTableSourceFilterPushDownTest after #3054 @platinumhamburg
- #3052 [client] LogFetchCollector uses immutable fetchOffset for staleness check, causing cross-poll CompletedFetch to be drained @platinumhamburg
- #3049 Introduce `netty.server.max-request-size` to limit the request size @loserwang1024 version: 0.9.1
- #3031 [flink] Add Flink filter pushdown integration and documentation @platinumhamburg
- #3021 Optimize DefaultLogRecordBatchStatistics.getNullCounts() to return int[] instead of Long[] @platinumhamburg
- #2729 Use default chunk size as 4 MB than than 16Mb in arrow's rolling policy. @loserwang1024
已合并 PRs
- #3065 [release-0.9] chore: rustfs assume role fixes and docusaurus fix @fresh-borzoni
- #3060 chore: fix broken main with Long[] -> int[] issue @fresh-borzoni
- #3058 chore: fix docs for RustFS AssumeRole @fresh-borzoni
- #3054 [common] Optimize getNullCounts() to return int[] instead of Long[] @platinumhamburg
- #3053 Fix stale fetch detection across poll boundaries @platinumhamburg
- #3050 [server][netty] Introduce netty.server.max-request-size to limit the request size @loserwang1024
- #3032 [flink] Add Flink filter pushdown integration and documentation @platinumhamburg
- #2732 [common] A custom rounding policy that reduces Arrow's chunk size fm 16MB to 4MB which same as netty arena. @loserwang1024