RuntimeError: Address already in use

I installed pytorch and slowfast. It works well when i run the training procedure on a single GPU, but when i run it on multiple GPUs I got this error.

-- Process 0 terminated with the following error:
Traceback (most recent call last):
File "/home/alfasly/anaconda3/envs/pyt/lib/python3.7/site-packages/torch/multiprocessing/spawn.py", line 19, in _wrap
fn(i, *args)
File "/home/alfasly/slowfast/slowfast/utils/multiprocessing.py", line 57, in run
raise e
File "/home/alfasly/slowfast/slowfast/utils/multiprocessing.py", line 54, in run
rank=rank,
File "/home/alfasly/anaconda3/envs/pyt/lib/python3.7/site-packages/torch/distributed/distributed_c10d.py", line 423, in init_process_group
store, rank, world_size = next(rendezvous_iterator)
File "/home/alfasly/anaconda3/envs/pyt/lib/python3.7/site-packages/torch/distributed/rendezvous.py", line 133, in _tcp_rendezvous_handler
store = TCPStore(result.hostname, result.port, world_size, start_daemon, timeout)
RuntimeError: Address already in use

anyone faced same problem?
-----------------
Ubuntu 18.04
python: 3.7
anaconda
torch: 1.7.0+cu101
cuda: 101
torchvision: 0.8.1