ProcessBundleProgressRequest/ProcessBundleProgressResponse are usable for this purpose, and are recommended for gauging bundle progress anyway.
https://github.com/apache/beam/blob/master/model/fn-execution/src/main/proto/org/apache/beam/model/fn_execution/v1/beam_fn_api.proto#L154 Same for the WorkerStatus request which should provide information on the worker. This has the advantage of being out of the ControlService path. https://github.com/apache/beam/blob/master/model/fn-execution/src/main/proto/org/apache/beam/model/fn_execution/v1/beam_fn_api.proto#L1278 I think a heartbeat was considered, but honestly we are either sending bundles/ and rpcs anyway, or would want regular updates on the worker level health from the status message, so I don't think there's value in another RPC. Grpc possibly has something built in as well, if only for connection monitoring. On Fri, Oct 4, 2024, 10:46 AM Joey Tran <joey.t...@schrodinger.com> wrote: > Hey all, > > I have a python runner I've written and I'm debugging the case when a SDK > worker crashes. Currently my driver/ runner starts the SDK worker and then > pushes data / instructions through the data and control connections, but > this just silently waits forever if the worker has actually crashed for > whatever reason. > > Is there a way to poll SDK workers to see if they're even alive? > > Cheers, > Joey >
