Apache Flink 中文用户邮件列表
Search
everywhere
only in this topic
Advanced Search
FLINK WEEKLY 2019/47
Classic
List
Threaded
♦
♦
Locked
1 message
tison
Reply
|
Threaded
Open this post in threaded view
♦
♦
|
FLINK WEEKLY 2019/47
FLINK WEEKLY 2019/47 <
https://zhuanlan.zhihu.com/p/93587534
>用户问题
-
关于flink1.7.2checkpoint本地文件存储,设置文件个数的问题
<
https://lists.apache.org/x/thread.html/87b6d90774401a79422cc10460de5d5414391a94e33c5e79225df3f4@%3Cuser-zh.flink.apache.org%3E
>:
文件个数是 FLINK 内部逻辑,不可配置也无需配置
-
Cron style for checkpoint
<
https://lists.apache.org/x/thread.html/801e9d0946a9f8f7061f7a85cbb6734ea7f26dbd642d5a2da01bf1de@%3Cuser.flink.apache.org%3E
>:
社区成员想要一个类似于 CRON 任务的后台定期 trigger checkpoint 的机制,邮件列表讨论了其合理性和现有方案解决实际需求的一些方式
-
Metrics for Task States
<
https://lists.apache.org/x/thread.html/ac96420bd7071c640be47b110f6b4aff6dbbc8f48b028db4e113f4dd@%3Cuser.flink.apache.org%3E
>:
社区成员在构建 FLINK 作业管理平台时对 TM 的监控指标提出了新的需求,想要获得一个 JOB 不同 TASK 处于什么状态的指标
-
Uid and name for Flink sources and sinks
<
https://lists.apache.org/x/thread.html/b5ae277791f14c270409d026d58ed9e963209c84b0adeb851aaef3f3@%3Cuser.flink.apache.org%3E
>
-
Does apache flink support stream input from Postgresql?
<
https://lists.apache.org/x/thread.html/395221cb7250f311b340de83caa3b7d5ff742b3301184b77407eef19@%3Cuser.flink.apache.org%3E
>:
可以使用 JDBC connector
-
how to setup a ha flink cluster on k8s?
<
https://lists.apache.org/x/thread.html/bebeea66d61a7ca42e8767b64a705940acf89218714fbb04fd6e07d5@%3Cuser.flink.apache.org%3E
>
-
Flink configuration at runtime
<
https://lists.apache.org/x/thread.html/cfc868bf5aec107d51991b6cde4cf893917749583ad0fdce2d17e330@%3Cuser.flink.apache.org%3E
>:
有些 JOB 级别的配置在最近的改动中能够在运行时提交前动态确定。但是大多数的配置还是直接从 flink-conf.yaml 中读取,不可改变
-
Dynamically creating new Task Managers in YARN
<
https://lists.apache.org/x/thread.html/dc95db62ca77bf87066103be0d2b169b6ed0742fc39377d694d65155@%3Cuser.flink.apache.org%3E
>:
FLIP-6 之后的 FLINK on YARN 已经是动态申请资源了
开发讨论
-
[DISCUSS] Support configure remote flink jar
<
https://lists.apache.org/x/thread.html/d875df9ab2c112309d564ae03dcb82a741cab03aadea8feba5fedaa8@%3Cuser.flink.apache.org%3E
>:
我发起了关于在 YARN 上部署作业时运行时依赖例如 FLINK 框架 JAR
从远端直接拉取的讨论。这一讨论可以延伸到任意远端依赖上,主要想解决的问题是目前 FLINK
在部署作业时总是要部署大量的用户依赖,这些依赖可能有很多共同的内容,却在每次部署是都需要重新上传和本地化。讨论着重在利用 YARN
的本地化机制来优化这一开销,K8S 上有类似的初始化机制可以优化
-
FLINK 1.8.3 即将发布
-
FLINK shaded 9.0 已经发布
-
FLINK pulsar connector 工作
<
https://docs.google.com/document/d/1LMnABtXn-wQedsmWv8hopvx-B-jbdr8-jHbIiDhdsoE/edit
>由
Yijie Shen 稳步推进中
-
[DISCUSS] Remove old WebUI
<
http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/DISCUSS-Remove-old-WebUI-td35218.html
>:
1.9 中引进的新版 WebUI 稳定运行中,Chesnay 提出移除旧版的 WebUI
-
[DISCUSS] Primary keys in Table API
<
http://apache-flink-mailing-list-archive.1008284.n3.nabble.com/DISCUSS-Primary-keys-in-Table-API-td35138.html
>:
Dawid 的 FLIP-87 旨在向 TableAPI 中引入主键的概念以助于运行时优化
社区发展
-
[ANNOUNCE] Launch of flink-packages.org: A website to foster the Flink
Ecosystem
<
https://lists.apache.org/x/thread.html/ba04b6f8a790da2475ab89a74ac89cda3f90f43633d3a7319b92c2d0@%3Cuser.flink.apache.org%3E
>:
FLINK 也有了自己的生态系统网站 <
https://flink-packages.org/
>啦~
Free forum by Nabble
Edit this page