[
https://issues.apache.org/jira/browse/FLINK-19700?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17216602#comment-17216602
]
Xintong Song commented on FLINK-19700:
--------------------------------------
Hi [~nicholasjiang],
I think what [~fly_in_gis] means is to replace the executor used by
{{KubeClient}}, rather than the main thread executor. If you look at
{{KubeClientFactory#fromConfiguration}}, currently a
{{KubeClientFactory::createThreadPoolForAsyncIO}} is used for creating
{{Fabric8FlinkKubeClient}}.
> Make Kubernetes Client in KubernetesResourceManagerDriver use io executor
> -------------------------------------------------------------------------
>
> Key: FLINK-19700
> URL: https://issues.apache.org/jira/browse/FLINK-19700
> Project: Flink
> Issue Type: Improvement
> Components: Deployment / Kubernetes
> Reporter: Yang Wang
> Priority: Major
> Fix For: 1.12.0
>
>
> Currently, the Kubernetes Client in {{KubernetesResourceManagerDriver}} is
> using a dedicated thread pool. After FLINK-19037 and FLINK-18722, we could
> get the io executor in {{KubernetesResourceManagerDriver}} and eliminate the
> redundant thread pool.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
