Actually few pipelines are failed and there is some GC issue. right now manually restart the cluster so need to automate the restart cluster at 8 PM daily. Main thing is as per data coming continuously we cant stop and then start the cluster as we have 30 mins window time