flink on k8s 如果jobmanager 所在pod重启后job失败如何处理

    请教一个问题。在使用k8s 部署的flink 集群，如果jobmanger 重启后，1）job所在的jar包会清除，jobmanager 找不到这个job的jar 包，2）正在运行的job也会取消，重启后的jobmanager 如何找到之前运行的job

---

[hidden email]

[Yun Tang]

Hi

Please use English to ask questions in user mailing list. I have added this thread to user-zh mailing list, if you would like to reply this thread again, please remove user mailing list in senders.

When talking about the question how to handle job manager failure in k8s, you could consider jobmanager high availability[1] and you could refer to [2] for plans of HighAvailabilityService based on native k8s APIs.

[1] https://ci.apache.org/projects/flink/flink-docs-stable/ops/jobmanager_high_availability.html

[2] https://issues.apache.org/jira/browse/FLINK-12884

Best

Yun Tang

---

From: [hidden email] <[hidden email]>
Sent: Friday, September 4, 2020 11:04
To: user <[hidden email]>
Subject: flink on k8s 如果jobmanager 所在pod重启后job失败如何处理

 

    请教一个问题。在使用k8s 部署的flink 集群，如果jobmanger 重启后，1）job所在的jar包会清除，jobmanager 找不到这个job的jar 包，2）正在运行的job也会取消，重启后的jobmanager 如何找到之前运行的job

---

[hidden email]