Getting stuck after confirm joining federated training session

[luckystarufo]

Hello there,

I am trying to run the code on three different machines within a network. The three machines can communicate properly but the training never starts.

After digging a little bit, I find the the code stucks at the run() function within federated-xgboost/dmlc-core/tracker/dmlc_tracker/rpc.py:

Here's what I saw on the server side:

And here are the ones on the two clients side:

Any suggestions? Thanks!