[server] Oversized lake metadata causes Coordinator disconnections from ZooKeeper and ISR update failures
维护者通常 1 天内回复
@hanliu0830 已经在做这个了。
开始于 2026年9月15日。
评估
这个 Issue 还没有评估数据。
描述
Search before asking
- I searched in the issues and found nothing similar.
Fluss version
0.9.0 (latest release)
Please describe the bug 🐞
Environment: Paimon lake tiering with deletion vectors (DV) enabled on the affected table; Fluss on Kubernetes; a three-node ZooKeeper 3.8.4 ensemble on VMs. The Coordinator runs on Linux amd64 with Java 17.0.7.
Observed trigger: During production lake tiering, historical lake snapshot metadata kept accumulating under /<cluster>/tabletservers/tables/<table-id>/laketable. The node held 5,091 snapshot entries (1,048,318 bytes). The next complete setData request was 1,048,596 bytes, exceeding the ZooKeeper server's jute.maxbuffer=1048575.
Expected: Lake snapshot commits complete successfully and ISR updates remain stable.
Actual: ZooKeeper repeatedly closed the Coordinator connection. The Coordinator reported SUSPENDED and lost leadership; Tablets reported NOT_COORDINATOR_LEADER_EXCEPTION when updating ISR. All three ZooKeeper servers logged the following error (address redacted):
java.io.IOException: Len error. A message from <coordinator-address> with advertised length of 1048596 is either a malformed message or too large to process (length is greater than jute.maxbuffer=1048575)
After ZooKeeper connectivity recovered, Tablets continued reporting OPERATION_NOT_ATTEMPTED_EXCEPTION during ISR updates. These failures stopped after the Coordinator was restarted.
Solution
We recovered the cluster with the following temporary workaround:
- Set
-Djute.maxbuffer=104857600on all three ZooKeeper servers and restart them one at a time. - Restart the Coordinator, as ISR failures persisted after ZooKeeper connectivity recovered.
- Restart the lake tiering job.
ISR updates and lake snapshot commits resumed. No TabletServer restart was needed. Increasing the buffer limit restored service but did not resolve the metadata accumulation.
Related documentation issues:
- #4336
- #4338
Are you willing to submit a PR?
- I'm willing to submit a PR!
- 主要语言
- Java
- 星标
- 2.2k
- 派生
- 656
- 平均合并
- 1 天 21 小时
- 30 天内合并 PR
- 127
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 没有贡献指南
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
apache/fluss 的其他 Issue
-
[lake/iceberg] Reserved write-mode properties can be set and reset through their `iceberg.` key可能已有人在做 @aphenix-r7 于 1 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 82/100
维护者通常 1 天内回复
-
[common] Decimal#hashCode is inconsistent with Decimal#equals可能已有人在做 @firasbouzazi 于 13 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 78/100
维护者通常 1 天内回复
-
[security] PlainSaslServerProvider conflicts with Hadoop SASL provider causing authentication failure可能已有人在做 @xinzhang-hello 于 19 天前认领。 未关闭
难度 1/5 1 小时以内 新手友好度 88/100
维护者通常 1 天内回复
-
[server] KV snapshot scheduling stops permanently after a transient snapshot-init failure可能已有人在做 @zhang-arvin 于 19 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 88/100
维护者通常 1 天内回复
-
[docs] Document advertised-resource and ACL principal ignore-case options可能已有人在做 @EchoLee5 于 31 天前认领。 未关闭
难度 2/5 1-3 小时 新手友好度 84/100
维护者通常 1 天内回复
相似的 Issue
-
难度 1/5 1 小时以内 新手友好度 90/100
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 82/100
apache/rocketmq-dashboard#6896 ·
维护者通常 3 天内回复
-
bug
难度 2/5 1-3 小时 新手友好度 72/100
OpenIntegrationEngine/engine#496 ·
维护者通常 1 天内回复
-
enhancement good first issue
难度 2/5 1-3 小时 新手友好度 76/100
SasanLabs/VulnerableApp#836 ·
维护者通常 1 天内回复
-
enhancement frontend
难度 2/5 1-3 小时 新手友好度 83/100
Albertinesilva/backend-engineering-journey-java-springboot#46 ·