ump45nose commented on PR #2643: URL: https://github.com/apache/datafusion-ballista/pull/2643#issuecomment-6544009868
@andygrove the failure is `killing_every_executor_terminates_the_job::case_1_aqe_off` in `ballista-chaos --test ha`, and this PR didn't cause it. The cluster never started: executor-1's Flight server couldn't bind port 35773 (`AddrInUse`) because executor-0's gRPC server had already taken it. `free_ports()` in `chaos-testing/src/cluster.rs` is called once per executor, so two executors can be handed the same ephemeral port. This PR only changes `query_stage_scheduler.rs`. On the PR head (4dd8be6) the full `cargo test -p ballista-chaos --test ha` passes locally (22/22). Could you re-run the failed job? I haven't rebased, because current main fails to compile `ballista-scheduler` tests (missing `ExecutionPlan` import in `execution_graph.rs`). (AI-assisted reply.) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
