=== Avatar Training (fire-and-forget) === Script: train_pi0fast.py Started: 2026-03-15 03:41:24 UTC >>> [0/4] Raising file descriptor limit... ulimit -n = 65536 >>> [1/4] Setting environment... HF_TOKEN set: 37 chars WANDB_API_KEY set: 0 chars >>> [2/4] Logging into HuggingFace... The token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `hf`CLI if you want to set the git credential as well. Token is valid (permission: write). The token `avatar` has been saved to /root/.cache/huggingface/stored_tokens Your token has been saved to /root/.cache/huggingface/token Login successful. Note: Environment variable`HF_TOKEN` is set and is the current active token independently from the token you've just configured. ⚠️ Warning: 'huggingface-cli login' is deprecated. Use 'hf auth login' instead. >>> [3/4] Pre-downloading pretrained model... Downloading lerobot/pi0fast-base (default)... Fetching 6 files: 0%| | 0/6 [00:00>> [4/4] Launching accelerate with 8 GPUs... Script: train_pi0fast.py Train args: --dataset alpha_bot_right_combined --steps 400000 --batch-size 8 --lr 5e-5 --warmup-steps 1000 --save-freq 30000 --log-freq 100 --chunk-size 10 --num-workers 8 --push-to-hub --wandb The following values were not passed to `accelerate launch` and had defaults used instead: `--num_machines` was set to a value of `1` `--mixed_precision` was set to a value of `'no'` `--dynamo_backend` was set to a value of `'no'` To avoid this warning pass in values for each of the problematic parameters or run `accelerate config`. ============================================================ π₀-FAST Training ============================================================ Dataset: avatarrobotics/alpha-bot-right-combined Pretrained: lerobot/pi0fast-base Model repo: avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 Steps: 400000 Batch/GPU: 8 (eff: 64) Chunk size: 10 LR: 5e-05 Save freq: 30000 GPUs: 8 ============================================================ [Avatar] W&B key invalid (0 chars), skipping [Avatar] W&B key invalid (0 chars), skipping [Avatar] W&B key invalid (0 chars), skipping [Avatar] W&B key invalid (0 chars), skipping [Avatar] W&B key invalid (0 chars), skipping [Avatar] W&B key invalid (0 chars), skipping [Avatar] W&B key invalid (0 chars), skipping [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) [Avatar] Auto-tagged dataset with version: v3.0 [Avatar] W&B key invalid (0 chars), skipping lerobot_train \ --dataset.repo_id=avatarrobotics/alpha-bot-right-combined \ --policy.type=pi0_fast \ --policy.pretrained_path=lerobot/pi0fast-base \ --policy.repo_id=avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 \ --policy.dtype=bfloat16 \ --policy.gradient_checkpointing=true \ --policy.chunk_size=10 \ --policy.n_action_steps=10 \ --policy.max_action_tokens=256 \ --policy.optimizer_lr=5e-05 \ --policy.optimizer_weight_decay=0.01 \ --policy.scheduler_warmup_steps=1000 \ --policy.scheduler_decay_steps=400000 \ --policy.scheduler_decay_lr=1e-6 \ --batch_size=8 \ --steps=400000 \ --save_freq=30000 \ --log_freq=100 \ --save_checkpoint=true \ --seed=1000 \ --num_workers=8 \ --output_dir=outputs/train/pi0fast \ --tolerance_s=2.0 \ --policy.push_to_hub=true \ --policy.private=true \ [Avatar] cudnn.benchmark=True, tf32=True [Avatar] Checkpoint push enabled → avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341 (with per-step tags) [Avatar] Dataset task text patched (multi-task from tasks.json) [Avatar] Dataset timestamp tolerance set to 2.0s [Avatar] Installed corrupt-frame retry handler [Avatar] Patched resize_with_pad_torch -> center_crop_resize (no padding waste) WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. WARNING:lerobot.configs.policies:Device 'None' is not available. Switching to 'cuda'. INFO 2026-03-15 03:42:00 ot_train.py:197 {'batch_size': 8, 'checkpoint_path': None, 'dataset': {'episodes': None, 'image_transforms': {'enable': False, 'max_num_transforms': 3, 'random_order': False, 'tfs': {'affine': {'kwargs': {'degrees': [-5.0, 5.0], 'translate': [0.05, 0.05]}, 'type': 'RandomAffine', 'weight': 1.0}, 'brightness': {'kwargs': {'brightness': [0.8, 1.2]}, 'type': 'ColorJitter', 'weight': 1.0}, 'contrast': {'kwargs': {'contrast': [0.8, 1.2]}, 'type': 'ColorJitter', 'weight': 1.0}, 'hue': {'kwargs': {'hue': [-0.05, 0.05]}, 'type': 'ColorJitter', 'weight': 1.0}, 'saturation': {'kwargs': {'saturation': [0.5, 1.5]}, 'type': 'ColorJitter', 'weight': 1.0}, 'sharpness': {'kwargs': {'sharpness': [0.5, 1.5]}, 'type': 'SharpnessJitter', 'weight': 1.0}}}, 'repo_id': 'avatarrobotics/alpha-bot-right-combined', 'revision': None, 'root': None, 'streaming': False, 'use_imagenet_stats': True, 'video_backend': 'torchcodec'}, 'env': None, 'eval': {'batch_size': 50, 'n_episodes': 50, 'use_async_envs': False}, 'eval_freq': 20000, 'job_name': 'pi0_fast', 'log_freq': 100, 'num_workers': 8, 'optimizer': {'betas': [0.9, 0.95], 'eps': 1e-08, 'grad_clip_norm': 1.0, 'lr': 5e-05, 'type': 'adamw', 'weight_decay': 0.01}, 'output_dir': 'outputs/train/pi0fast', 'peft': None, 'policy': {'action_expert_variant': 'gemma_300m', 'action_tokenizer_name': 'physical-intelligence/fast', 'chunk_size': 10, 'compile_mode': 'max-autotune', 'compile_model': False, 'device': 'cuda', 'dtype': 'bfloat16', 'empty_cameras': 0, 'fast_skip_tokens': 128, 'gradient_checkpointing': True, 'image_resolution': [224, 224], 'input_features': {}, 'license': None, 'max_action_dim': 32, 'max_action_tokens': 256, 'max_decoding_steps': 256, 'max_state_dim': 32, 'n_action_steps': 10, 'n_obs_steps': 1, 'normalization_mapping': {'ACTION': , 'STATE': , 'VISUAL': }, 'optimizer_betas': [0.9, 0.95], 'optimizer_eps': 1e-08, 'optimizer_grad_clip_norm': 1.0, 'optimizer_lr': 5e-05, 'optimizer_weight_decay': 0.01, 'output_features': {}, 'paligemma_variant': 'gemma_2b', 'pretrained_path': 'lerobot/pi0fast-base', 'private': True, 'push_to_hub': True, 'repo_id': 'avatarrobotics/pi0fast-alpha-bot-right-b64-20260315-0341', 'rtc_config': None, 'scheduler_decay_lr': 1e-06, 'scheduler_decay_steps': 400000, 'scheduler_warmup_steps': 1000, 'tags': None, 'temperature': 0.0, 'text_tokenizer_name': 'google/paligemma-3b-pt-224', 'tokenizer_max_length': 200, 'type': 'pi0_fast', 'use_amp': False, 'use_kv_cache': True, 'use_peft': False, 'validate_action_token_prefix': True}, 'rabc_epsilon': 1e-06, 'rabc_head_mode': 'sparse', 'rabc_kappa': 0.01, 'rabc_progress_path': None, 'rename_map': {}, 'resume': False, 'save_checkpoint': True, 'save_freq': 30000, 'scheduler': {'decay_lr': 1e-06, 'num_decay_steps': 400000, 'num_warmup_steps': 1000, 'peak_lr': 5e-05, 'type': 'cosine_decay_with_warmup'}, 'seed': 1000, 'steps': 400000, 'tolerance_s': 2.0, 'use_policy_training_preset': True, 'use_rabc': False, 'wandb': {'disable_artifact': False, 'enable': False, 'entity': None, 'mode': None, 'notes': None, 'project': 'lerobot', 'run_id': None}} INFO 2026-03-15 03:42:00 ot_train.py:205 Logs will be saved locally. INFO 2026-03-15 03:42:00 ot_train.py:217 Creating dataset Fetching 6 files: 0%| | 0/6 [00:00 [rank6]: main() [rank6]: File "/tmp/train_pi0fast.py", line 414, in main [rank6]: lerobot_main() [rank6]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank6]: train() [rank6]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank6]: response = fn(cfg, *args, **kwargs) [rank6]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank6]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank6]: accelerator.wait_for_everyone() [rank6]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank6]: wait_for_everyone() [rank6]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank6]: PartialState().wait_for_everyone() [rank6]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank6]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank6]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank6]: return func(*args, **kwargs) [rank6]: ^^^^^^^^^^^^^^^^^^^^^ [rank6]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank6]: work = group.barrier(opts=opts) [rank6]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank6]: torch.distributed.DistBackendError: [6] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank6]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank6]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x72732a7785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank6]: frame #1: + 0x5b2100e (0x727313a4600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #2: + 0x13638f5 (0x72730f2888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x727313acafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x727313acc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x727313a7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x727313a7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x727313a7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x7272d51c0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank6]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x7272d51d0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank6]: frame #10: + 0x11ecf83 (0x7272d51d5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank6]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x7272d51d677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank6]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x7272d51e63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank6]: frame #13: + 0x5b4363f (0x727313a6863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #14: + 0x5b52cfe (0x727313a77cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #15: + 0x52125b0 (0x7273131375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #16: + 0x5b61a02 (0x727313a86a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #17: + 0x5b6293d (0x727313a8793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank6]: frame #18: + 0xc8c143 (0x7273229da143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank6]: frame #19: + 0x38a1ac (0x7273220d81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank6]: frame #20: /usr/bin/python() [0x55563b] [rank6]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank6]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank6]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank6]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank6]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank6]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank6]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank6]: frame #28: /usr/bin/python() [0x613f24] [rank6]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank6]: frame #30: /usr/bin/python() [0x63439b] [rank6]: frame #31: /usr/bin/python() [0x6305f4] [rank6]: frame #32: /usr/bin/python() [0x644c55] [rank6]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank6]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank6]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank6]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank6]: frame #37: + 0x29d90 (0x72732b406d90 in /lib/x86_64-linux-gnu/libc.so.6) [rank6]: frame #38: __libc_start_main + 0x80 (0x72732b406e40 in /lib/x86_64-linux-gnu/libc.so.6) [rank6]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank6]: . This may indicate a possible application crash on rank 0 or a network set up issue. [rank7]:[W315 03:51:59.067207215 socket.cpp:460] [c10d] waitForInput: poll for socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60130, remote=[::ffff:127.0.0.1]:29500) returned 0, likely a timeout [rank7]:[W315 03:51:59.084291765 socket.cpp:485] [c10d] waitForInput: socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60130, remote=[::ffff:127.0.0.1]:29500) timed out after 600000ms [rank7]: Traceback (most recent call last): [rank7]: File "/tmp/train_pi0fast.py", line 418, in [rank7]: main() [rank7]: File "/tmp/train_pi0fast.py", line 414, in main [rank7]: lerobot_main() [rank7]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank7]: train() [rank7]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank7]: response = fn(cfg, *args, **kwargs) [rank7]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank7]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank7]: accelerator.wait_for_everyone() [rank7]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank7]: wait_for_everyone() [rank7]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank7]: PartialState().wait_for_everyone() [rank7]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank7]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank7]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank7]: return func(*args, **kwargs) [rank7]: ^^^^^^^^^^^^^^^^^^^^^ [rank7]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank7]: work = group.barrier(opts=opts) [rank7]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank7]: torch.distributed.DistBackendError: [7] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank7]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank7]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x7d78acb785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank7]: frame #1: + 0x5b2100e (0x7d7895e4600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #2: + 0x13638f5 (0x7d78916888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x7d7895ecafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x7d7895ecc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d7895e7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d7895e7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d7895e7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x7d78575c0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank7]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x7d78575d0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank7]: frame #10: + 0x11ecf83 (0x7d78575d5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank7]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x7d78575d677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank7]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x7d78575e63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank7]: frame #13: + 0x5b4363f (0x7d7895e6863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #14: + 0x5b52cfe (0x7d7895e77cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #15: + 0x52125b0 (0x7d78955375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #16: + 0x5b61a02 (0x7d7895e86a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #17: + 0x5b6293d (0x7d7895e8793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank7]: frame #18: + 0xc8c143 (0x7d78a4dda143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank7]: frame #19: + 0x38a1ac (0x7d78a44d81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank7]: frame #20: /usr/bin/python() [0x55563b] [rank7]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank7]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank7]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank7]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank7]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank7]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank7]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank7]: frame #28: /usr/bin/python() [0x613f24] [rank7]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank7]: frame #30: /usr/bin/python() [0x63439b] [rank7]: frame #31: /usr/bin/python() [0x6305f4] [rank7]: frame #32: /usr/bin/python() [0x644c55] [rank7]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank7]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank7]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank7]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank7]: frame #37: + 0x29d90 (0x7d78ad8bbd90 in /lib/x86_64-linux-gnu/libc.so.6) [rank7]: frame #38: __libc_start_main + 0x80 (0x7d78ad8bbe40 in /lib/x86_64-linux-gnu/libc.so.6) [rank7]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank7]: . This may indicate a possible application crash on rank 0 or a network set up issue. [rank1]:[W315 03:51:59.150093269 socket.cpp:460] [c10d] waitForInput: poll for socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60148, remote=[::ffff:127.0.0.1]:29500) returned 0, likely a timeout [rank1]:[W315 03:51:59.166666390 socket.cpp:485] [c10d] waitForInput: socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60148, remote=[::ffff:127.0.0.1]:29500) timed out after 600000ms [rank1]: Traceback (most recent call last): [rank1]: File "/tmp/train_pi0fast.py", line 418, in [rank1]: main() [rank1]: File "/tmp/train_pi0fast.py", line 414, in main [rank1]: lerobot_main() [rank1]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank1]: train() [rank1]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank1]: response = fn(cfg, *args, **kwargs) [rank1]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank1]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank1]: accelerator.wait_for_everyone() [rank1]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank1]: wait_for_everyone() [rank1]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank1]: PartialState().wait_for_everyone() [rank1]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank1]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank1]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank1]: return func(*args, **kwargs) [rank1]: ^^^^^^^^^^^^^^^^^^^^^ [rank1]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank1]: work = group.barrier(opts=opts) [rank1]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank1]: torch.distributed.DistBackendError: [1] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank1]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank1]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x784e2bd785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank1]: frame #1: + 0x5b2100e (0x784e1544600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #2: + 0x13638f5 (0x784e10c888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x784e154cafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x784e154cc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x784e1547a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x784e1547a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x784e1547a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x784dd6bc0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank1]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x784dd6bd0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank1]: frame #10: + 0x11ecf83 (0x784dd6bd5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank1]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x784dd6bd677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank1]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x784dd6be63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank1]: frame #13: + 0x5b4363f (0x784e1546863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #14: + 0x5b52cfe (0x784e15477cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #15: + 0x52125b0 (0x784e14b375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #16: + 0x5b61a02 (0x784e15486a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #17: + 0x5b6293d (0x784e1548793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank1]: frame #18: + 0xc8c143 (0x784e243da143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank1]: frame #19: + 0x38a1ac (0x784e23ad81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank1]: frame #20: /usr/bin/python() [0x55563b] [rank1]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank1]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank1]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank1]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank1]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank1]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank1]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank1]: frame #28: /usr/bin/python() [0x613f24] [rank1]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank1]: frame #30: /usr/bin/python() [0x63439b] [rank1]: frame #31: /usr/bin/python() [0x6305f4] [rank1]: frame #32: /usr/bin/python() [0x644c55] [rank1]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank1]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank1]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank1]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank1]: frame #37: + 0x29d90 (0x784e2ccf0d90 in /lib/x86_64-linux-gnu/libc.so.6) [rank1]: frame #38: __libc_start_main + 0x80 (0x784e2ccf0e40 in /lib/x86_64-linux-gnu/libc.so.6) [rank1]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank1]: . This may indicate a possible application crash on rank 0 or a network set up issue. [rank3]:[W315 03:52:00.409050362 socket.cpp:460] [c10d] waitForInput: poll for socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60174, remote=[::ffff:127.0.0.1]:29500) returned 0, likely a timeout [rank4]:[W315 03:52:00.421098591 socket.cpp:460] [c10d] waitForInput: poll for socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60160, remote=[::ffff:127.0.0.1]:29500) returned 0, likely a timeout [rank3]:[W315 03:52:00.425639603 socket.cpp:485] [c10d] waitForInput: socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60174, remote=[::ffff:127.0.0.1]:29500) timed out after 600000ms [rank4]:[W315 03:52:00.437778883 socket.cpp:485] [c10d] waitForInput: socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60160, remote=[::ffff:127.0.0.1]:29500) timed out after 600000ms [rank3]: Traceback (most recent call last): [rank3]: File "/tmp/train_pi0fast.py", line 418, in [rank3]: main() [rank3]: File "/tmp/train_pi0fast.py", line 414, in main [rank3]: lerobot_main() [rank3]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank3]: train() [rank3]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank3]: response = fn(cfg, *args, **kwargs) [rank3]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank3]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank3]: accelerator.wait_for_everyone() [rank3]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank3]: wait_for_everyone() [rank3]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank3]: PartialState().wait_for_everyone() [rank3]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank3]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank3]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank3]: return func(*args, **kwargs) [rank3]: ^^^^^^^^^^^^^^^^^^^^^ [rank3]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank3]: work = group.barrier(opts=opts) [rank3]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank3]: torch.distributed.DistBackendError: [3] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank3]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank3]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x7196ab3785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank3]: frame #1: + 0x5b2100e (0x71969464600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #2: + 0x13638f5 (0x71968fe888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x7196946cafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x7196946cc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x71969467a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x71969467a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x71969467a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x719655dc0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank3]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x719655dd0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank3]: frame #10: + 0x11ecf83 (0x719655dd5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank3]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x719655dd677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank3]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x719655de63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank3]: frame #13: + 0x5b4363f (0x71969466863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #14: + 0x5b52cfe (0x719694677cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #15: + 0x52125b0 (0x719693d375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #16: + 0x5b61a02 (0x719694686a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #17: + 0x5b6293d (0x71969468793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank3]: frame #18: + 0xc8c143 (0x7196a35da143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank3]: frame #19: + 0x38a1ac (0x7196a2cd81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank3]: frame #20: /usr/bin/python() [0x55563b] [rank3]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank3]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank3]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank3]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank3]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank3]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank3]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank3]: frame #28: /usr/bin/python() [0x613f24] [rank3]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank3]: frame #30: /usr/bin/python() [0x63439b] [rank3]: frame #31: /usr/bin/python() [0x6305f4] [rank3]: frame #32: /usr/bin/python() [0x644c55] [rank3]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank3]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank3]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank3]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank3]: frame #37: + 0x29d90 (0x7196ac01ad90 in /lib/x86_64-linux-gnu/libc.so.6) [rank3]: frame #38: __libc_start_main + 0x80 (0x7196ac01ae40 in /lib/x86_64-linux-gnu/libc.so.6) [rank3]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank3]: . This may indicate a possible application crash on rank 0 or a network set up issue. [rank2]:[W315 03:52:00.474163488 socket.cpp:460] [c10d] waitForInput: poll for socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60170, remote=[::ffff:127.0.0.1]:29500) returned 0, likely a timeout [rank4]: Traceback (most recent call last): [rank4]: File "/tmp/train_pi0fast.py", line 418, in [rank4]: main() [rank4]: File "/tmp/train_pi0fast.py", line 414, in main [rank4]: lerobot_main() [rank4]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank4]: train() [rank4]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank4]: response = fn(cfg, *args, **kwargs) [rank4]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank4]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank4]: accelerator.wait_for_everyone() [rank4]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank4]: wait_for_everyone() [rank4]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank4]: PartialState().wait_for_everyone() [rank4]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank4]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank4]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank4]: return func(*args, **kwargs) [rank4]: ^^^^^^^^^^^^^^^^^^^^^ [rank4]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank4]: work = group.barrier(opts=opts) [rank4]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank4]: torch.distributed.DistBackendError: [4] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank4]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank4]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x7d38a1d785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank4]: frame #1: + 0x5b2100e (0x7d388b44600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #2: + 0x13638f5 (0x7d3886c888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x7d388b4cafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x7d388b4cc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d388b47a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d388b47a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d388b47a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x7d384cbc0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank4]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x7d384cbd0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank4]: frame #10: + 0x11ecf83 (0x7d384cbd5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank4]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x7d384cbd677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank4]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x7d384cbe63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank4]: frame #13: + 0x5b4363f (0x7d388b46863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #14: + 0x5b52cfe (0x7d388b477cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #15: + 0x52125b0 (0x7d388ab375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #16: + 0x5b61a02 (0x7d388b486a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #17: + 0x5b6293d (0x7d388b48793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank4]: frame #18: + 0xc8c143 (0x7d389a3da143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank4]: frame #19: + 0x38a1ac (0x7d3899ad81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank4]: frame #20: /usr/bin/python() [0x55563b] [rank4]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank4]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank4]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank4]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank4]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank4]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank4]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank4]: frame #28: /usr/bin/python() [0x613f24] [rank4]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank4]: frame #30: /usr/bin/python() [0x63439b] [rank4]: frame #31: /usr/bin/python() [0x6305f4] [rank4]: frame #32: /usr/bin/python() [0x644c55] [rank4]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank4]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank4]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank4]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank4]: frame #37: + 0x29d90 (0x7d38a2d74d90 in /lib/x86_64-linux-gnu/libc.so.6) [rank4]: frame #38: __libc_start_main + 0x80 (0x7d38a2d74e40 in /lib/x86_64-linux-gnu/libc.so.6) [rank4]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank4]: . This may indicate a possible application crash on rank 0 or a network set up issue. [rank5]:[W315 03:52:00.488353259 socket.cpp:460] [c10d] waitForInput: poll for socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60182, remote=[::ffff:127.0.0.1]:29500) returned 0, likely a timeout [rank2]:[W315 03:52:00.491152437 socket.cpp:485] [c10d] waitForInput: socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60170, remote=[::ffff:127.0.0.1]:29500) timed out after 600000ms [rank5]:[W315 03:52:00.504851676 socket.cpp:485] [c10d] waitForInput: socket SocketImpl(fd=44, addr=[::ffff:127.0.0.1]:60182, remote=[::ffff:127.0.0.1]:29500) timed out after 600000ms [rank2]: Traceback (most recent call last): [rank2]: File "/tmp/train_pi0fast.py", line 418, in [rank2]: main() [rank2]: File "/tmp/train_pi0fast.py", line 414, in main [rank2]: lerobot_main() [rank2]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank2]: train() [rank2]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank2]: response = fn(cfg, *args, **kwargs) [rank2]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank2]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank2]: accelerator.wait_for_everyone() [rank2]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank2]: wait_for_everyone() [rank2]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank2]: PartialState().wait_for_everyone() [rank2]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank2]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank2]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank2]: return func(*args, **kwargs) [rank2]: ^^^^^^^^^^^^^^^^^^^^^ [rank2]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank2]: work = group.barrier(opts=opts) [rank2]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank2]: torch.distributed.DistBackendError: [2] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank2]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank2]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x7d55dd3785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank2]: frame #1: + 0x5b2100e (0x7d55c6a4600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #2: + 0x13638f5 (0x7d55c22888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x7d55c6acafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x7d55c6acc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d55c6a7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d55c6a7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x7d55c6a7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x7d55881c0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank2]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x7d55881d0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank2]: frame #10: + 0x11ecf83 (0x7d55881d5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank2]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x7d55881d677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank2]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x7d55881e63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank2]: frame #13: + 0x5b4363f (0x7d55c6a6863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #14: + 0x5b52cfe (0x7d55c6a77cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #15: + 0x52125b0 (0x7d55c61375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #16: + 0x5b61a02 (0x7d55c6a86a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #17: + 0x5b6293d (0x7d55c6a8793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank2]: frame #18: + 0xc8c143 (0x7d55d59da143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank2]: frame #19: + 0x38a1ac (0x7d55d50d81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank2]: frame #20: /usr/bin/python() [0x55563b] [rank2]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank2]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank2]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank2]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank2]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank2]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank2]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank2]: frame #28: /usr/bin/python() [0x613f24] [rank2]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank2]: frame #30: /usr/bin/python() [0x63439b] [rank2]: frame #31: /usr/bin/python() [0x6305f4] [rank2]: frame #32: /usr/bin/python() [0x644c55] [rank2]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank2]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank2]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank2]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank2]: frame #37: + 0x29d90 (0x7d55de3a0d90 in /lib/x86_64-linux-gnu/libc.so.6) [rank2]: frame #38: __libc_start_main + 0x80 (0x7d55de3a0e40 in /lib/x86_64-linux-gnu/libc.so.6) [rank2]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank2]: . This may indicate a possible application crash on rank 0 or a network set up issue. [rank5]: Traceback (most recent call last): [rank5]: File "/tmp/train_pi0fast.py", line 418, in [rank5]: main() [rank5]: File "/tmp/train_pi0fast.py", line 414, in main [rank5]: lerobot_main() [rank5]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 547, in main [rank5]: train() [rank5]: File "/usr/local/lib/python3.11/dist-packages/lerobot/configs/parser.py", line 233, in wrapper_inner [rank5]: response = fn(cfg, *args, **kwargs) [rank5]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank5]: File "/usr/local/lib/python3.11/dist-packages/lerobot/scripts/lerobot_train.py", line 220, in train [rank5]: accelerator.wait_for_everyone() [rank5]: File "/usr/local/lib/python3.11/dist-packages/accelerate/accelerator.py", line 3256, in wait_for_everyone [rank5]: wait_for_everyone() [rank5]: File "/usr/local/lib/python3.11/dist-packages/accelerate/utils/other.py", line 320, in wait_for_everyone [rank5]: PartialState().wait_for_everyone() [rank5]: File "/usr/local/lib/python3.11/dist-packages/accelerate/state.py", line 412, in wait_for_everyone [rank5]: torch.distributed.barrier(device_ids=[self.local_process_index]) [rank5]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/c10d_logger.py", line 81, in wrapper [rank5]: return func(*args, **kwargs) [rank5]: ^^^^^^^^^^^^^^^^^^^^^ [rank5]: File "/usr/local/lib/python3.11/dist-packages/torch/distributed/distributed_c10d.py", line 4635, in barrier [rank5]: work = group.barrier(opts=opts) [rank5]: ^^^^^^^^^^^^^^^^^^^^^^^^ [rank5]: torch.distributed.DistBackendError: [5] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0', but store->get('0') got error: wait timeout after 600000ms, keys: /default_pg/0//cuda//0 [rank5]: Exception raised from doWait at /pytorch/torch/csrc/distributed/c10d/TCPStore.cpp:589 (most recent call first): [rank5]: frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string, std::allocator >) + 0x98 (0x734bb09785e8 in /usr/local/lib/python3.11/dist-packages/torch/lib/libc10.so) [rank5]: frame #1: + 0x5b2100e (0x734b99c4600e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #2: + 0x13638f5 (0x734b954888f5 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #3: c10d::TCPStore::doGet(std::__cxx11::basic_string, std::allocator > const&) + 0x33 (0x734b99ccafa3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #4: c10d::TCPStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0xab (0x734b99ccc08b in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #5: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x734b99c7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #6: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x734b99c7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #7: c10d::PrefixStore::get(std::__cxx11::basic_string, std::allocator > const&) + 0x2f (0x734b99c7a94f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #8: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::__cxx11::basic_string, std::allocator > const&, int) + 0x16d (0x734b5b3c0e7d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank5]: frame #9: c10d::ProcessGroupNCCL::initNCCLComm(std::__cxx11::basic_string, std::allocator > const&, c10::Device&, c10d::OpType, int, bool) + 0x16bf (0x734b5b3d0f6f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank5]: frame #10: + 0x11ecf83 (0x734b5b3d5f83 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank5]: frame #11: c10d::ProcessGroupNCCL::allreduce_impl(at::Tensor&, char const*, c10d::AllreduceOptions const&) + 0xee (0x734b5b3d677e in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank5]: frame #12: c10d::ProcessGroupNCCL::barrier(c10d::BarrierOptions const&) + 0x683 (0x734b5b3e63a3 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cuda.so) [rank5]: frame #13: + 0x5b4363f (0x734b99c6863f in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #14: + 0x5b52cfe (0x734b99c77cfe in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #15: + 0x52125b0 (0x734b993375b0 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #16: + 0x5b61a02 (0x734b99c86a02 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #17: + 0x5b6293d (0x734b99c8793d in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_cpu.so) [rank5]: frame #18: + 0xc8c143 (0x734ba8bda143 in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank5]: frame #19: + 0x38a1ac (0x734ba82d81ac in /usr/local/lib/python3.11/dist-packages/torch/lib/libtorch_python.so) [rank5]: frame #20: /usr/bin/python() [0x55563b] [rank5]: frame #21: _PyObject_MakeTpCall + 0x27c (0x52f68c in /usr/bin/python) [rank5]: frame #22: _PyEval_EvalFrameDefault + 0x6bd (0x53d81d in /usr/bin/python) [rank5]: frame #23: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank5]: frame #24: PyObject_Call + 0xf4 (0x570764 in /usr/bin/python) [rank5]: frame #25: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank5]: frame #26: _PyFunction_Vectorcall + 0x173 (0x566263 in /usr/bin/python) [rank5]: frame #27: _PyEval_EvalFrameDefault + 0x4929 (0x541a89 in /usr/bin/python) [rank5]: frame #28: /usr/bin/python() [0x613f24] [rank5]: frame #29: PyEval_EvalCode + 0x97 (0x613587 in /usr/bin/python) [rank5]: frame #30: /usr/bin/python() [0x63439b] [rank5]: frame #31: /usr/bin/python() [0x6305f4] [rank5]: frame #32: /usr/bin/python() [0x644c55] [rank5]: frame #33: _PyRun_SimpleFileObject + 0x194 (0x6442c4 in /usr/bin/python) [rank5]: frame #34: _PyRun_AnyFileObject + 0x47 (0x6440d7 in /usr/bin/python) [rank5]: frame #35: Py_RunMain + 0x2c9 (0x63ec29 in /usr/bin/python) [rank5]: frame #36: Py_BytesMain + 0x2d (0x60486d in /usr/bin/python) [rank5]: frame #37: + 0x29d90 (0x734bb166cd90 in /lib/x86_64-linux-gnu/libc.so.6) [rank5]: frame #38: __libc_start_main + 0x80 (0x734bb166ce40 in /lib/x86_64-linux-gnu/libc.so.6) [rank5]: frame #39: _start + 0x25 (0x6046f5 in /usr/bin/python) [rank5]: . This may indicate a possible application crash on rank 0 or a network set up issue. W0315 03:52:00.741000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2470 closing signal SIGTERM W0315 03:52:00.743000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2471 closing signal SIGTERM W0315 03:52:00.745000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2472 closing signal SIGTERM W0315 03:52:00.745000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2473 closing signal SIGTERM W0315 03:52:00.746000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2474 closing signal SIGTERM W0315 03:52:00.746000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2475 closing signal SIGTERM W0315 03:52:00.747000 2402 torch/distributed/elastic/multiprocessing/api.py:900] Sending process 2477 closing signal SIGTERM E0315 03:52:01.569000 2402 torch/distributed/elastic/multiprocessing/api.py:874] failed (exitcode: 1) local_rank: 6 (pid: 2476) of binary: /usr/bin/python Traceback (most recent call last): File "/usr/local/bin/accelerate", line 8, in sys.exit(main()) ^^^^^^ File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/accelerate_cli.py", line 50, in main args.func(args) File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/launch.py", line 1396, in launch_command multi_gpu_launcher(args) File "/usr/local/lib/python3.11/dist-packages/accelerate/commands/launch.py", line 1023, in multi_gpu_launcher distrib_run.run(args) File "/usr/local/lib/python3.11/dist-packages/torch/distributed/run.py", line 883, in run elastic_launch( File "/usr/local/lib/python3.11/dist-packages/torch/distributed/launcher/api.py", line 139, in __call__ return launch_agent(self._config, self._entrypoint, list(args)) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/dist-packages/torch/distributed/launcher/api.py", line 270, in launch_agent raise ChildFailedError( torch.distributed.elastic.multiprocessing.errors.ChildFailedError: ============================================================ /tmp/train_pi0fast.py FAILED ------------------------------------------------------------ Failures: ------------------------------------------------------------ Root Cause (first observed failure): [0]: time : 2026-03-15_03:52:00 host : 4947f6b9293d rank : 6 (local_rank: 6) exitcode : 1 (pid: 2476) error_file: traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html ============================================================ >>> Training finished with exit code 1 Ended: 2026-03-15 03:52:02 UTC >>> Saving training log to HuggingFace...