Kunfu Panda

@krasinski we are using [Manual Mode of External Backend on k8](https://github.com/h2oai/sparkling-water/issues/2751) with some changes on the stateful set as mentioned in this [post](https://github.com/h2oai/sparkling-water/issues/2748#issuecomment-1077619126). Note: we are using preemptive nodes for...

@mn-mikke Sorry my bad. The value is h2o-service.sparkling-water.svc.cluster.local only. And i have set the SW property to spark.ext.h2o.cloud.representative="h2o-service.sparkling-water.svc.cluster.local".

@mn-mikke I tried on the non-preemptive nodes and facing following issue. which is fixed according to this [jira ](https://h2oai.atlassian.net/browse/SW-1799)ticket Current spark version: 3.1.2 H2O Version: 3.36.0.3-1-3.1 ``` 22/05/25 16:23:10 INFO...

@krasinski the following is my spark config ``` conf["spark.driver.cores"]=2 conf["spark.driver.memory"]="2g" conf["spark.executor.cores"]=4 conf["spark.executor.memory"]="12g" conf["spark.executor.instances"]=1

@mn-mikke But currently, when I try with two request parallelly, one of the request have been paused at this state. And getting the following error ```2022-05-23T09:23:26.783742943Z 22/05/23 09:23:26 INFO H2OContext:...

so @mn-mikke there is no way to save the model using separate thread currently. Correct me if i am wrong or do we have any alternate options?