Re: TTL issue with large RocksDB keyed state

Hi, > 1. After multiple full checkpoints and a NATIVE savepoint the size was > unchanged. I'm wondering if RocksDb compaction is because we never update > key values? The state is nearly fully composed of keys' space. Do keys not > get freed using RocksDb compaction filter for TTL? Regarding

Re: 【求助】关于 Flink ML 迭代中使用keyBy算子报错

您好！ 看起来这个问题与 FLINK-35066[1] 有关，该问题描述了在 IterationBody 内实现自定义的RichCoProcessFunction 或 CoFlatMapFunction 算子时遇到的拆包问题，可以追溯到这个[2]邮件列表中的问题报告。看起来这个问题也同样影响您使用的 RichCoMapFunction 算子。 该问题已被此 Pull Request[3] 解决，并已合入 master 主分支。按照文档[4]尝试在本地编译 2.4-SNAPSHOT 快照版本并执行您的代码，看起来能够正常工作。 鉴于这是一个 Flink ML 2.3

RE: Implementing Multiple sink

Yes. But the filter us usually a very light weight operation. From: Mingliang Liu Sent: Monday, June 3, 2024 7:16 PM To: Colletta, Edward Cc: mejri houssem ; user@flink.apache.org Subject: Re: Implementing Multiple sink NOTICE: This email is from an external sender - do not click on links or

Re: Implementing Multiple sink

Colletta, I think that way, the upstream stream `streamWithMultipleConditions` will get processed twice, instead of once? Thanks, On Mon, Jun 3, 2024 at 10:28 AM Colletta, Edward wrote: > I usually just reuse the stream, sending it to through different filters > and adding different sinks to

Re: Implementing Multiple sink

Thank you very much Mingliang and Colletta for the suggestions. I will try them out. To the others , I am still open to additional suggestions as well. Le lun. 3 juin 2024 à 18:28, Colletta, Edward a écrit : > I usually just reuse the stream, sending it to through different filters > and

RE: Implementing Multiple sink

I usually just reuse the stream, sending it to through different filters and adding different sinks to the filtered streams. Something like streamWithMultipleConditions.filter(FilterForCondition1) .addSink(SinkforCondtiton1);

Re: Implementing Multiple sink

Hi Mejri, Have you checked side outputs? https://nightlies.apache.org/flink/flink-docs-release-1.19/docs/dev/datastream/side_output/ On Mon, Jun 3, 2024 at 8:28 AM mejri houssem wrote: > Hello community, > > We have a use case in our Flink job that requires the implementation of > multiple

Implementing Multiple sink

Hello community, We have a use case in our Flink job that requires the implementation of multiple sinks. I need to filter messages based on certain conditions (information in the message) to determine which sink to dispatch them to. To clarify, I would like to implement logic in the operator

State leak in tumbling windows

Dear Flink users, I spotted the ever growing checkpoint size in my Flink application which uses tumble windows. I found such a ticket: https://issues.apache.org/jira/browse/FLINK-33192, but no comments. Can somebody confirm the issue? BR, Adam.

Re: Flink Kubernetes Operator Pod Disruption Budget

Hey Jeremy! This sounds like a good / fairly simple extension to add. Since this would result in a larger extension of the current FlinkDeployment CRD, it would be good to cover it in a small FLIP. Cheers, Gyula On Wed, May 22, 2024 at 10:20 PM Jeremy Alvis via user < user@flink.apache.org>