You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@flink.apache.org by Akshay Iyangar <ai...@godaddy.com> on 2020/06/01 19:46:27 UTC

Native kubernetes integration for standalone job clusters.

Hi

So we actually run Flink in Kubernetes as a standalone Flink job for each Flink pipeline currently.

We wanted to take advantage of Flink HA using K8 but looks like it only supports Flink session  clusters currently for version 1.10.

Any ideas when will it have support for standalone job clusters?

Thanks
Akshay I

Re: Native kubernetes integration for standalone job clusters.

Posted by Yang Wang <da...@gmail.com>.

Hi Akshay,

Currently, zookeeper HA service could be used for both session cluster and
job cluster
when deploying Flink on K8s. If you mean to using the native K8s HA(i.g.
native leader
election and configmap to store meta)[1], i think it could not be supported
now.


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

Best,
Yang

Akshay Iyangar <ai...@godaddy.com> 于2020年6月2日周二 上午3:46写道：

>
>
> Hi
>
>
>
> So we actually run Flink in Kubernetes as a standalone Flink job for each
> Flink pipeline currently.
>
>
>
> We wanted to take advantage of Flink HA using K8 but looks like it only
> supports Flink session  clusters currently for version 1.10.
>
>
>
> Any ideas when will it have support for standalone job clusters?
>
>
>
> Thanks
>
> Akshay I
>
>
>