Depending on what parallelism you et for the job, only one node may be needed for the execution.
If you increase the parallelism, the second node will start working as well. Here is a bit of an illustration how tasks and slots work in Apache Flink https://ci.apache.org/projects/flink/flink-docs-master/concepts/concepts.html#workers-slots-resources On Thu, Jun 9, 2016 at 9:09 AM, Ismaël Mejía <[email protected]> wrote: > Probably [email protected] is a more appropriate mailing list to ask > this. > > > On Thu, Jun 9, 2016 at 1:34 AM, amir bahmanyari <[email protected]> > wrote: > >> Hi colleagues, >> Whats the default mechanism/algorithm for Flink Cluster LB? >> I have a 2-nodes Flink cluster receiving from a single instance Kafka. >> Both Flink instances are up. >> But only the one with the JobMgr is processing. >> The second node is idle. Any idea? >> Thanks for your time. >> Amir- >> > >
