======================== START TIME: Tue Jul 2 20:36:42 UTC 2024 python3 version = Python 3.10.14 ======================== 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 `huggingface-cli` if you want to set the git credential as well. Token is valid (permission: write). Your token has been saved to /admin/home/ferdinand_mom/.cache/huggingface/token Login successful Already on 'bench_cluster' M examples/config_tiny_llama.py M examples/config_tiny_llama.yaml M examples/train_tiny_llama.sh M src/nanotron/models/llama.py M src/nanotron/trainer.py Your branch is up to date with 'origin/bench_cluster'. Job status: RUNNING W0702 20:36:49.048000 139997183711040 torch/distributed/run.py:757] W0702 20:36:49.048000 139997183711040 torch/distributed/run.py:757] ***************************************** W0702 20:36:49.048000 139997183711040 torch/distributed/run.py:757] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. W0702 20:36:49.048000 139997183711040 torch/distributed/run.py:757] ***************************************** W0702 20:36:49.775000 139648004355904 torch/distributed/run.py:757] W0702 20:36:49.775000 139648004355904 torch/distributed/run.py:757] ***************************************** W0702 20:36:49.775000 139648004355904 torch/distributed/run.py:757] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. W0702 20:36:49.775000 139648004355904 torch/distributed/run.py:757] ***************************************** [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Config: [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Config(general=GeneralArgs(project='bench_cluster', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: run='%date_%jobid', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: seed=42, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: step=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: consumed_train_samples=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: benchmark_csv_path=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: ignore_sanity_checks=True), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: parallelism=ParallelismArgs(dp=2, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: pp=8, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tp=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: pp_engine=, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tp_mode=, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tp_linear_async_communication=False, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: expert_parallel_size=1), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: model=ModelArgs(model_config=LlamaConfig(bos_token_id=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: eos_token_id=2, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: hidden_act='silu', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: hidden_size=2048, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: initializer_range=0.02, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: intermediate_size=4096, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: is_llama_config=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: max_position_embeddings=4096, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_attention_heads=32, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_hidden_layers=24, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_key_value_heads=32, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: pad_token_id=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: pretraining_tp=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: rms_norm_eps=1e-05, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: rope_scaling=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: rope_theta=10000.0, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tie_word_embeddings=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: use_cache=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: vocab_size=50257), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: init_method=RandomInit(std=0.025), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: dtype=torch.bfloat16, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: make_vocab_size_divisible_by=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: ddp_bucket_cap_mb=25), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tokenizer=TokenizerArgs(tokenizer_name_or_path='openai-community/gpt2', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tokenizer_revision=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tokenizer_max_length=None), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: checkpoints=CheckpointsArgs(checkpoints_path=Path('/dev/null'), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: checkpoint_interval=100000, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: save_initial_state=False, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: resume_checkpoint_path=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: checkpoints_path_is_shared_file_system=False), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: logging=LoggingArgs(log_level='info', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: log_level_replica='info', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: iteration_step_info_interval=1), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tokens=TokensArgs(sequence_length=4096, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: train_steps=20, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: micro_batch_size=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: batch_accumulation_per_replica=512, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: val_check_interval=-1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: limit_val_batches=0, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: limit_test_batches=0), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: optimizer=OptimizerArgs(optimizer_factory=AdamWOptimizerArgs(adam_eps=1e-08, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: adam_beta1=0.9, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: adam_beta2=0.95, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: torch_adam_is_fused=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: name='adamW'), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: zero_stage=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: weight_decay=0.01, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: clip_grad=1.0, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: accumulate_grad_in_fp32=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: learning_rate_scheduler=LRSchedulerArgs(learning_rate=0.0001, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: lr_warmup_steps=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: lr_warmup_style='linear', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: lr_decay_style='linear', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: lr_decay_steps=19, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: lr_decay_starting_step=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: min_decay_lr=1e-05)), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: data_stages=[DatasetStageArgs(name='Training Stage', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: start_training_step=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: data=DataArgs(dataset=PretrainDatasetsArgs(hf_dataset_or_datasets='roneneldan/TinyStories', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: hf_dataset_splits='train', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: hf_dataset_config_name=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: dataset_processing_num_proc_per_process=64, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: dataset_overwrite_cache=False, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: text_column_name='text'), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: seed=42, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_loading_workers=32))], [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: profiler=ProfilerArgs(profiler_export_path=Path('/fsx/ferdinandmom/ferdinand-hf/bench_cluster/results/llama-1B/16_GPUS/dp-2_tp-1_pp-8_mbz-1')), [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: lighteval=None) [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Model Config: [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: LlamaConfig(bos_token_id=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: eos_token_id=2, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: hidden_act='silu', [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: hidden_size=2048, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: initializer_range=0.02, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: intermediate_size=4096, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: is_llama_config=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: max_position_embeddings=4096, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_attention_heads=32, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_hidden_layers=24, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: num_key_value_heads=32, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: pad_token_id=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: pretraining_tp=1, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: rms_norm_eps=1e-05, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: rope_scaling=None, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: rope_theta=10000.0, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: tie_word_embeddings=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: use_cache=True, [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: vocab_size=50257) [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Building model.. [default0]:07/02/2024 20:37:12 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Setting PP block ranks... [default6]:07/02/2024 20:37:26 [INFO|DP=0|PP=3|TP=0|ip-26-0-165-24]: Local number of parameters: 168M (320.03MiB) [default6]:07/02/2024 20:37:26 [INFO|DP=0|PP=3|TP=0|ip-26-0-165-24]: [After model building] Memory usage: 324.04MiB. Peak allocated: 326.07MiB Peak reserved: 336.00MiB [default6]:07/02/2024 20:37:26 [INFO|DP=0|PP=3|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Total number of parameters: 1.21G (2312.82MiB) [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Local number of parameters: 271M (516.35MiB) [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [After model building] Memory usage: 520.36MiB. Peak allocated: 522.39MiB Peak reserved: 534.00MiB [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Parametrizing model parameters using StandardParametrizator [default4]:07/02/2024 20:37:26 [INFO|DP=0|PP=2|TP=0|ip-26-0-165-24]: Local number of parameters: 126M (240.02MiB) [default4]:07/02/2024 20:37:26 [INFO|DP=0|PP=2|TP=0|ip-26-0-165-24]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB [default4]:07/02/2024 20:37:26 [INFO|DP=0|PP=2|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default2]:07/02/2024 20:37:26 [INFO|DP=0|PP=1|TP=0|ip-26-0-165-24]: Local number of parameters: 126M (240.02MiB) [default2]:07/02/2024 20:37:26 [INFO|DP=0|PP=1|TP=0|ip-26-0-165-24]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB [default2]:07/02/2024 20:37:26 [INFO|DP=0|PP=1|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default2]:07/02/2024 20:37:26 [INFO|DP=0|PP=5|TP=0|ip-26-0-170-160]: Local number of parameters: 126M (240.02MiB) [default2]:07/02/2024 20:37:26 [INFO|DP=0|PP=5|TP=0|ip-26-0-170-160]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB [default2]:07/02/2024 20:37:26 [INFO|DP=0|PP=5|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=4|TP=0|ip-26-0-170-160]: Local number of parameters: 126M (240.02MiB) [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=4|TP=0|ip-26-0-170-160]: [After model building] Memory usage: 243.03MiB. Peak allocated: 245.06MiB Peak reserved: 262.00MiB [default0]:07/02/2024 20:37:26 [INFO|DP=0|PP=4|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default6]:07/02/2024 20:37:26 [INFO|DP=0|PP=7|TP=0|ip-26-0-170-160]: Local number of parameters: 103M (196.32MiB) [default6]:07/02/2024 20:37:26 [INFO|DP=0|PP=7|TP=0|ip-26-0-170-160]: [After model building] Memory usage: 196.33MiB. Peak allocated: 196.34MiB Peak reserved: 200.00MiB [default6]:07/02/2024 20:37:26 [INFO|DP=0|PP=7|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default4]:07/02/2024 20:37:26 [INFO|DP=0|PP=6|TP=0|ip-26-0-170-160]: Local number of parameters: 168M (320.03MiB) [default4]:07/02/2024 20:37:26 [INFO|DP=0|PP=6|TP=0|ip-26-0-170-160]: [After model building] Memory usage: 324.04MiB. Peak allocated: 326.07MiB Peak reserved: 336.00MiB [default4]:07/02/2024 20:37:26 [INFO|DP=0|PP=6|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default1]:07/02/2024 20:37:26 [INFO|DP=1|PP=0|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default3]:07/02/2024 20:37:26 [INFO|DP=1|PP=1|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default7]:07/02/2024 20:37:26 [INFO|DP=1|PP=3|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default1]:07/02/2024 20:37:26 [INFO|DP=1|PP=4|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default3]:07/02/2024 20:37:26 [INFO|DP=1|PP=5|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default5]:07/02/2024 20:37:26 [INFO|DP=1|PP=2|TP=0|ip-26-0-165-24]: No checkpoint path provided. [default5]:07/02/2024 20:37:26 [INFO|DP=1|PP=6|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default7]:07/02/2024 20:37:26 [INFO|DP=1|PP=7|TP=0|ip-26-0-170-160]: No checkpoint path provided. [default0]:07/02/2024 20:37:29 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [Optimizer Building] Using LearningRateForSP as learning rate [default0]:07/02/2024 20:37:29 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [ZeRO sharding] Size of optimizer params per rank: [default0]:07/02/2024 20:37:29 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [ZeRO sharding] DP Rank 0 has 135M out of 271M (50.00%) params' optimizer states [default0]:07/02/2024 20:37:29 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [ZeRO sharding] DP Rank 1 has 135M out of 271M (50.00%) params' optimizer states [default0]:07/02/2024 20:37:30 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [Training Plan] Stage Training Stage has 19 remaining training steps and has consumed 0 samples [default0]:07/02/2024 20:37:30 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Using `datasets` library [default0]:07/02/2024 20:37:30 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Loading tokenizer from openai-community/gpt2 and transformers/hf_hub versions ('4.41.2', '0.23.4') [default0]:07/02/2024 20:37:30 [WARNING|DP=0|PP=0|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default0]:Repo card metadata block was not found. Setting CardData to empty. [default0]:07/02/2024 20:37:32 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [Training Plan] There are 1 training stages [default0]:07/02/2024 20:37:32 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [Stage Training Stage] start from step 1 [default0]:07/02/2024 20:37:32 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [default0]:07/02/2024 20:37:32 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: [Start training] datetime: 2024-07-02 20:37:32.785004 | mbs: 1 | grad_accum: 512 | global_batch_size: 1024 | sequence_length: 4096 | train_steps: 20 | start_iteration_step: 0 | consumed_train_samples: 0 [default0]:07/02/2024 20:37:32 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Resuming training from stage Training Stage, it has trained for 0 samples and has 19 remaining train steps [default0]:07/02/2024 20:37:32 [INFO|DP=0|PP=0|TP=0|ip-26-0-165-24]: Memory usage: 2069.40MiB. Peak allocated 2069.40MiB. Peak reserved: 2086.00MiB [default6]:07/02/2024 20:37:32 [WARNING|DP=0|PP=3|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default1]:07/02/2024 20:37:32 [WARNING|DP=1|PP=0|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default5]:Repo card metadata block was not found. Setting CardData to empty. [default4]:Repo card metadata block was not found. Setting CardData to empty. [default7]:Repo card metadata block was not found. Setting CardData to empty. [default4]:07/02/2024 20:37:32 [WARNING|DP=0|PP=2|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default7]:07/02/2024 20:37:32 [WARNING|DP=1|PP=3|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default2]:07/02/2024 20:37:32 [WARNING|DP=0|PP=1|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default2]:07/02/2024 20:37:32 [WARNING|DP=0|PP=5|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default3]:07/02/2024 20:37:32 [WARNING|DP=1|PP=5|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default1]:07/02/2024 20:37:32 [WARNING|DP=1|PP=4|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default0]:07/02/2024 20:37:32 [WARNING|DP=0|PP=4|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default6]:07/02/2024 20:37:32 [WARNING|DP=0|PP=7|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default6]:Repo card metadata block was not found. Setting CardData to empty. [default2]:Repo card metadata block was not found. Setting CardData to empty. [default1]:Repo card metadata block was not found. Setting CardData to empty. [default5]:07/02/2024 20:37:32 [WARNING|DP=1|PP=6|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default5]:07/02/2024 20:37:32 [WARNING|DP=1|PP=2|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default6]:Repo card metadata block was not found. Setting CardData to empty. [default1]:Repo card metadata block was not found. Setting CardData to empty. [default3]:Repo card metadata block was not found. Setting CardData to empty. [default5]:Repo card metadata block was not found. Setting CardData to empty. [default2]:Repo card metadata block was not found. Setting CardData to empty. [default0]:Repo card metadata block was not found. Setting CardData to empty. [default7]:Repo card metadata block was not found. Setting CardData to empty. [default7]:07/02/2024 20:37:32 [WARNING|DP=1|PP=7|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default4]:Repo card metadata block was not found. Setting CardData to empty. [default4]:07/02/2024 20:37:32 [WARNING|DP=0|PP=6|TP=0|ip-26-0-170-160]: Repo card metadata block was not found. Setting CardData to empty. [default3]:07/02/2024 20:37:33 [WARNING|DP=1|PP=1|TP=0|ip-26-0-165-24]: Repo card metadata block was not found. Setting CardData to empty. [default3]:Repo card metadata block was not found. Setting CardData to empty. [default0]:[rank0]: Traceback (most recent call last): [default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default0]:[rank0]: trainer.train(dataloader) [default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default0]:[rank0]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default0]:[rank0]: outputs = self.pipeline_engine.train_batch_iter( [default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 250, in train_batch_iter [default0]:[rank0]: micro_batch = next(batch) [default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 465, in [default0]:[rank0]: batch=(next(dataloader) for _ in range(self.n_micro_batches_per_batch)), [default0]:[rank0]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/dataloader.py", line 46, in sanity_check_dataloader [default0]:[rank0]: for batch in dataloader: [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 439, in __iter__ [default0]:[rank0]: return self._get_iterator() [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 387, in _get_iterator [default0]:[rank0]: return _MultiProcessingDataLoaderIter(self) [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1040, in __init__ [default0]:[rank0]: w.start() [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/process.py", line 121, in start [default0]:[rank0]: self._popen = self._Popen(self) [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 224, in _Popen [default0]:[rank0]: return _default_context.get_context().Process._Popen(process_obj) [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 281, in _Popen [default0]:[rank0]: return Popen(process_obj) [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 19, in __init__ [default0]:[rank0]: self._launch(process_obj) [default0]:[rank0]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 66, in _launch [default0]:[rank0]: self.pid = os.fork() [default0]:[rank0]: OSError: [Errno 12] Cannot allocate memory [default6]:[rank14]: Traceback (most recent call last): [default6]:[rank14]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default6]:[rank14]: trainer.train(dataloader) [default6]:[rank14]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default6]:[rank14]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default6]:[rank14]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default6]:[rank14]: outputs = self.pipeline_engine.train_batch_iter( [default6]:[rank14]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 276, in train_batch_iter [default6]:[rank14]: for micro_batch in batch: [default6]:[rank14]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 465, in [default6]:[rank14]: batch=(next(dataloader) for _ in range(self.n_micro_batches_per_batch)), [default6]:[rank14]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/dataloader.py", line 46, in sanity_check_dataloader [default6]:[rank14]: for batch in dataloader: [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 439, in __iter__ [default6]:[rank14]: return self._get_iterator() [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 387, in _get_iterator [default6]:[rank14]: return _MultiProcessingDataLoaderIter(self) [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1040, in __init__ [default6]:[rank14]: w.start() [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/process.py", line 121, in start [default6]:[rank14]: self._popen = self._Popen(self) [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 224, in _Popen [default6]:[rank14]: return _default_context.get_context().Process._Popen(process_obj) [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 281, in _Popen [default6]:[rank14]: return Popen(process_obj) [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 19, in __init__ [default6]:[rank14]: self._launch(process_obj) [default6]:[rank14]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 66, in _launch [default6]:[rank14]: self.pid = os.fork() [default6]:[rank14]: OSError: [Errno 12] Cannot allocate memory [default7]:[rank15]: Traceback (most recent call last): [default7]:[rank15]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default7]:[rank15]: trainer.train(dataloader) [default7]:[rank15]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default7]:[rank15]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default7]:[rank15]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default7]:[rank15]: outputs = self.pipeline_engine.train_batch_iter( [default7]:[rank15]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 276, in train_batch_iter [default7]:[rank15]: for micro_batch in batch: [default7]:[rank15]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 465, in [default7]:[rank15]: batch=(next(dataloader) for _ in range(self.n_micro_batches_per_batch)), [default7]:[rank15]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/dataloader.py", line 46, in sanity_check_dataloader [default7]:[rank15]: for batch in dataloader: [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 439, in __iter__ [default7]:[rank15]: return self._get_iterator() [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 387, in _get_iterator [default7]:[rank15]: return _MultiProcessingDataLoaderIter(self) [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1040, in __init__ [default7]:[rank15]: w.start() [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/process.py", line 121, in start [default7]:[rank15]: self._popen = self._Popen(self) [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 224, in _Popen [default7]:[rank15]: return _default_context.get_context().Process._Popen(process_obj) [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 281, in _Popen [default7]:[rank15]: return Popen(process_obj) [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 19, in __init__ [default7]:[rank15]: self._launch(process_obj) [default7]:[rank15]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 66, in _launch [default7]:[rank15]: self.pid = os.fork() [default7]:[rank15]: OSError: [Errno 12] Cannot allocate memory [default1]:[rank1]: Traceback (most recent call last): [default1]:[rank1]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default1]:[rank1]: trainer.train(dataloader) [default1]:[rank1]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default1]:[rank1]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default1]:[rank1]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default1]:[rank1]: outputs = self.pipeline_engine.train_batch_iter( [default1]:[rank1]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 250, in train_batch_iter [default1]:[rank1]: micro_batch = next(batch) [default1]:[rank1]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 465, in [default1]:[rank1]: batch=(next(dataloader) for _ in range(self.n_micro_batches_per_batch)), [default1]:[rank1]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/dataloader.py", line 46, in sanity_check_dataloader [default1]:[rank1]: for batch in dataloader: [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 439, in __iter__ [default1]:[rank1]: return self._get_iterator() [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 387, in _get_iterator [default1]:[rank1]: return _MultiProcessingDataLoaderIter(self) [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/utils/data/dataloader.py", line 1040, in __init__ [default1]:[rank1]: w.start() [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/process.py", line 121, in start [default1]:[rank1]: self._popen = self._Popen(self) [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 224, in _Popen [default1]:[rank1]: return _default_context.get_context().Process._Popen(process_obj) [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/context.py", line 281, in _Popen [default1]:[rank1]: return Popen(process_obj) [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 19, in __init__ [default1]:[rank1]: self._launch(process_obj) [default1]:[rank1]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/multiprocessing/popen_fork.py", line 66, in _launch [default1]:[rank1]: self.pid = os.fork() [default1]:[rank1]: OSError: [Errno 12] Cannot allocate memory [default1]:[rank1]:[E ProcessGroupNCCL.cpp:1316] [PG 0 Rank 1] Heartbeat monitor timed out! Process will be terminated after dumping debug info. workMetaList_.size()=0 [default1]:[rank1]:[E ProcessGroupNCCL.cpp:1153] [PG 0 Rank 1] ProcessGroupNCCL preparing to dump debug info. [default7]:[rank15]:[E ProcessGroupNCCL.cpp:1316] [PG 0 Rank 15] Heartbeat monitor timed out! Process will be terminated after dumping debug info. workMetaList_.size()=0 [default7]:[rank15]:[E ProcessGroupNCCL.cpp:1153] [PG 0 Rank 15] ProcessGroupNCCL preparing to dump debug info. [default6]:[rank14]:[E ProcessGroupNCCL.cpp:1316] [PG 0 Rank 14] Heartbeat monitor timed out! Process will be terminated after dumping debug info. workMetaList_.size()=0 [default6]:[rank14]:[E ProcessGroupNCCL.cpp:1153] [PG 0 Rank 14] ProcessGroupNCCL preparing to dump debug info. [default0]:[rank0]:[E ProcessGroupNCCL.cpp:1316] [PG 0 Rank 0] Heartbeat monitor timed out! Process will be terminated after dumping debug info. workMetaList_.size()=0 [default0]:[rank0]:[E ProcessGroupNCCL.cpp:1153] [PG 0 Rank 0] ProcessGroupNCCL preparing to dump debug info. [default1]:[rank1]:[F ProcessGroupNCCL.cpp:1169] [PG 0 Rank 1] [PG 0 Rank 1] ProcessGroupNCCL's watchdog got stuck for 600 seconds without making progress in monitoring enqueued collectives. This typically indicates a NCCL/CUDA API hang blocking the watchdog, and could be triggered by another thread holding the GIL inside a CUDA api, or other deadlock-prone behaviors.If you suspect the watchdog is not actually stuck and a longer timeout would help, you can either increase the timeout (TORCH_NCCL_HEARTBEAT_TIMEOUT_SEC) to a larger value or disable the heartbeat monitor (TORCH_NCCL_ENABLE_MONITORING=0).If either of aforementioned helps, feel free to file an issue to PyTorch about the short timeout or false positive abort; otherwise, please attempt to debug the hang. workMetaList_.size() = 0 [default7]:[rank15]:[F ProcessGroupNCCL.cpp:1169] [PG 0 Rank 15] [PG 0 Rank 15] ProcessGroupNCCL's watchdog got stuck for 600 seconds without making progress in monitoring enqueued collectives. This typically indicates a NCCL/CUDA API hang blocking the watchdog, and could be triggered by another thread holding the GIL inside a CUDA api, or other deadlock-prone behaviors.If you suspect the watchdog is not actually stuck and a longer timeout would help, you can either increase the timeout (TORCH_NCCL_HEARTBEAT_TIMEOUT_SEC) to a larger value or disable the heartbeat monitor (TORCH_NCCL_ENABLE_MONITORING=0).If either of aforementioned helps, feel free to file an issue to PyTorch about the short timeout or false positive abort; otherwise, please attempt to debug the hang. workMetaList_.size() = 0 [default6]:[rank14]:[F ProcessGroupNCCL.cpp:1169] [PG 0 Rank 14] [PG 0 Rank 14] ProcessGroupNCCL's watchdog got stuck for 600 seconds without making progress in monitoring enqueued collectives. This typically indicates a NCCL/CUDA API hang blocking the watchdog, and could be triggered by another thread holding the GIL inside a CUDA api, or other deadlock-prone behaviors.If you suspect the watchdog is not actually stuck and a longer timeout would help, you can either increase the timeout (TORCH_NCCL_HEARTBEAT_TIMEOUT_SEC) to a larger value or disable the heartbeat monitor (TORCH_NCCL_ENABLE_MONITORING=0).If either of aforementioned helps, feel free to file an issue to PyTorch about the short timeout or false positive abort; otherwise, please attempt to debug the hang. workMetaList_.size() = 0 [default0]:[rank0]:[F ProcessGroupNCCL.cpp:1169] [PG 0 Rank 0] [PG 0 Rank 0] ProcessGroupNCCL's watchdog got stuck for 600 seconds without making progress in monitoring enqueued collectives. This typically indicates a NCCL/CUDA API hang blocking the watchdog, and could be triggered by another thread holding the GIL inside a CUDA api, or other deadlock-prone behaviors.If you suspect the watchdog is not actually stuck and a longer timeout would help, you can either increase the timeout (TORCH_NCCL_HEARTBEAT_TIMEOUT_SEC) to a larger value or disable the heartbeat monitor (TORCH_NCCL_ENABLE_MONITORING=0).If either of aforementioned helps, feel free to file an issue to PyTorch about the short timeout or false positive abort; otherwise, please attempt to debug the hang. workMetaList_.size() = 0 [default5]:[rank5]: Traceback (most recent call last): [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default5]:[rank5]: trainer.train(dataloader) [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default5]:[rank5]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default5]:[rank5]: outputs = self.pipeline_engine.train_batch_iter( [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default5]:[rank5]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default7]:[rank7]: Traceback (most recent call last): [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default4]:[rank4]: Traceback (most recent call last): [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default4]:[rank4]: trainer.train(dataloader) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default4]:[rank4]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default4]:[rank4]: outputs = self.pipeline_engine.train_batch_iter( [default5]:[rank5]: output = model(**micro_batch) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank5]: return self._call_impl(*args, **kwargs) [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank5]: return forward_call(*args, **kwargs) [default4]:[rank4]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default5]:[rank5]: sharded_logits = self.model( [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default4]:[rank4]: output = model(**micro_batch) [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank4]: return self._call_impl(*args, **kwargs) [default5]:[rank5]: return self._call_impl(*args, **kwargs) [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank4]: return forward_call(*args, **kwargs) [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default5]:[rank5]: return forward_call(*args, **kwargs) [default4]:[rank4]: sharded_logits = self.model( [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default5]:[rank5]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank4]: return self._call_impl(*args, **kwargs) [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default5]:[rank5]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default4]:[rank4]: return forward_call(*args, **kwargs) [default7]:[rank7]: trainer.train(dataloader) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default7]:[rank7]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default7]:[rank7]: outputs = self.pipeline_engine.train_batch_iter( [default4]:[rank4]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default5]:[rank5]: return self._call_impl(*args, **kwargs) [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default7]:[rank7]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default4]:[rank4]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default5]:[rank5]: return forward_call(*args, **kwargs) [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default7]:[rank7]: output = model(**micro_batch) [default5]:[rank5]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default4]:[rank4]: return self._call_impl(*args, **kwargs) [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default7]:[rank7]: return self._call_impl(*args, **kwargs) [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank5]: pipeline_state.run_communication() [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank4]: return forward_call(*args, **kwargs) [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default7]:[rank7]: return forward_call(*args, **kwargs) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default5]:[rank5]: recv_activation_tensor = recv_activation() [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default4]:[rank4]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default7]:[rank7]: sharded_logits = self.model( [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank4]: pipeline_state.run_communication() [default5]:[rank5]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default7]:[rank7]: return self._call_impl(*args, **kwargs) [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank4]: recv_activation_tensor = recv_activation() [default7]:[rank7]: return forward_call(*args, **kwargs) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default5]:[rank5]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default4]:[rank4]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default5]:[rank5]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default4]:[rank4]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default7]:[rank7]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default5]:[rank5]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default5]:[rank5]: dist.recv( [default4]:[rank4]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default7]:[rank7]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default5]:[rank5]: return func(*args, **kwargs) [default4]:[rank4]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank4]: dist.recv( [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default7]:[rank7]: return self._call_impl(*args, **kwargs) [default4]:[rank4]: return func(*args, **kwargs) [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank5]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default4]:[rank4]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default5]:[rank5]: pg.recv([tensor], group_src_rank, tag).wait() [default7]:[rank7]: return forward_call(*args, **kwargs) [default4]:[rank4]: pg.recv([tensor], group_src_rank, tag).wait() [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default4]:[rank4]: torch.distributed.DistBackendError: [2] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '1:2', but store->get('1:2') got error: Connection reset by peer [default5]:[rank5]: torch.distributed.DistBackendError: [2] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '1:2', but store->get('1:2') got error: Connection reset by peer [default7]:[rank7]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default5]:[rank5]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default5]:[rank5]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f181fa72897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default7]:[rank7]: pipeline_state.run_communication() [default4]:[rank4]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default5]:[rank5]: frame #1: + 0x5b3a23e (0x7f185958f23e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f6d621e8897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default7]:[rank7]: recv_activation_tensor = recv_activation() [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default4]:[rank4]: frame #1: + 0x5b3a23e (0x7f6d9bd0523e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f1859589c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default5]:[rank5]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f1859589f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default4]:[rank4]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f6d9bcffc87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default5]:[rank5]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f185958afd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f6d9bcfff82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default5]:[rank5]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f185953f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f6d9bd00fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default4]:[rank4]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f6d9bcb5371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default4]:[rank4]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f6d9bcb5371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f185953f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f6d9bcb5371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: dist.recv( [default5]:[rank5]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f185953f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f6d9bcb5371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default5]:[rank5]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f185953f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f1820d4c189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default4]:[rank4]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f6d634c2189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default7]:[rank7]: return func(*args, **kwargs) [default4]:[rank4]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f6d634c9610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default7]:[rank7]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default5]:[rank5]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f1820d53610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default4]:[rank4]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f6d634e8978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default7]:[rank7]: pg.recv([tensor], group_src_rank, tag).wait() [default5]:[rank5]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f1820d72978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default4]:[rank4]: frame #12: + 0x5adc309 (0x7f6d9bca7309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: torch.distributed.DistBackendError: [3] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '2:3', but store->get('2:3') got error: Connection reset by peer [default5]:[rank5]: frame #12: + 0x5adc309 (0x7f1859531309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default4]:[rank4]: frame #13: + 0x5ae6f10 (0x7f6d9bcb1f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fbd9ee98897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default7]:[rank7]: frame #1: + 0x5b3a23e (0x7fbdd89b523e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7fbdd89afc87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7fbdd89aff82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7fbdd89b0fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fbdd8965371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fbdd8965371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fbdd8965371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fbdd8965371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #13: + 0x5ae6f10 (0x7f185953bf10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7fbda0172189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default7]:[rank7]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7fbda0179610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default7]:[rank7]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7fbda0198978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default7]:[rank7]: frame #12: + 0x5adc309 (0x7fbdd8957309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #14: + 0x5ae6fa5 (0x7f185953bfa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #13: + 0x5ae6f10 (0x7fbdd8961f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #15: + 0x5124446 (0x7f1858b79446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #16: + 0x1acf4b8 (0x7f18555244b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #14: + 0x5ae6fa5 (0x7f6d9bcb1fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #17: + 0x5aee004 (0x7f1859543004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #15: + 0x5124446 (0x7f6d9b2ef446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #14: + 0x5ae6fa5 (0x7fbdd8961fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #16: + 0x1acf4b8 (0x7f6d97c9a4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #18: + 0x5af36b5 (0x7f18595486b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #15: + 0x5124446 (0x7fbdd7f9f446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #19: + 0xd2631e (0x7f186c13231e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default4]:[rank4]: frame #17: + 0x5aee004 (0x7f6d9bcb9004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default7]:[rank7]: frame #16: + 0x1acf4b8 (0x7fbdd494a4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #20: + 0x47def4 (0x7f186b889ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default7]:[rank7]: frame #17: + 0x5aee004 (0x7fbdd8969004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #21: + 0x1445a6 (0x55ec6575d5a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #18: + 0x5af36b5 (0x7fbdd896e6b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank4]: frame #18: + 0x5af36b5 (0x7f6d9bcbe6b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank5]: frame #22: _PyObject_MakeTpCall + 0x26b (0x55ec65756a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #19: + 0xd2631e (0x7f6dae8a831e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default4]:[rank4]: frame #20: + 0x47def4 (0x7f6dadfffef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default7]:[rank7]: frame #19: + 0xd2631e (0x7fbdeb55831e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default4]:[rank4]: frame #21: + 0x1445a6 (0x556cf40145a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #23: + 0x150866 (0x55ec65769866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #20: + 0x47def4 (0x7fbdeacafef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default4]:[rank4]: frame #22: _PyObject_MakeTpCall + 0x26b (0x556cf400da6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #21: + 0x1445a6 (0x5576c2a735a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #22: _PyObject_MakeTpCall + 0x26b (0x5576c2a6ca6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x55ec65752142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #23: + 0x150866 (0x556cf4020866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #23: + 0x150866 (0x5576c2a7f866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x556cf4009142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #25: _PyFunction_Vectorcall + 0x6c (0x55ec6575da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #25: _PyFunction_Vectorcall + 0x6c (0x556cf4014a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #26: PyObject_Call + 0xbc (0x55ec65769f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #26: PyObject_Call + 0xbc (0x556cf4020f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x5576c2a68142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x55ec657502b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #25: _PyFunction_Vectorcall + 0x6c (0x5576c2a73a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #28: _PyFunction_Vectorcall + 0x6c (0x55ec6575da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x556cf40072b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #28: _PyFunction_Vectorcall + 0x6c (0x556cf4014a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x55ec6574e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #26: PyObject_Call + 0xbc (0x5576c2a7ff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x5576c2a662b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #30: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x556cf40058fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #28: _PyFunction_Vectorcall + 0x6c (0x5576c2a73a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x5576c2a648fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #30: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #30: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x5576c2a648fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x556cf40058fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x55ec6574e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: Traceback (most recent call last): [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default4]:[rank4]: frame #32: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x556cf40058fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #34: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #32: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x5576c2a648fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #32: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x55ec6574e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x556cf40058fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: trainer.train(dataloader) [default4]:[rank4]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x556cf400cf50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #34: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #34: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x5576c2a648fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x55ec6574e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x55ec65755f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #37: _PyObject_Call_Prepend + 0x69 (0x556cf401ec39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #38: + 0x211239 (0x556cf40e1239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default3]:[rank3]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default4]:[rank4]: frame #39: _PyObject_MakeTpCall + 0x26b (0x556cf400da6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x556cf40093e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #41: _PyFunction_Vectorcall + 0x6c (0x556cf4014a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #37: _PyObject_Call_Prepend + 0x69 (0x55ec65767c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x5576c2a6bf50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default4]:[rank4]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x556cf4004c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #43: _PyFunction_Vectorcall + 0x6c (0x556cf4014a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #37: _PyObject_Call_Prepend + 0x69 (0x5576c2a7dc39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #38: + 0x211239 (0x55ec6582a239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #39: _PyObject_MakeTpCall + 0x26b (0x55ec65756a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x556cf40058fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #45: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #38: + 0x211239 (0x5576c2b40239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #46: PyObject_Call + 0xbc (0x556cf4020f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x556cf40072b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: outputs = self.pipeline_engine.train_batch_iter( [default7]:[rank7]: frame #39: _PyObject_MakeTpCall + 0x26b (0x5576c2a6ca6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default4]:[rank4]: frame #48: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x5576c2a683e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x55ec657523e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #49: PyObject_Call + 0xbc (0x556cf4020f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #41: _PyFunction_Vectorcall + 0x6c (0x5576c2a73a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x5576c2a63c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #41: _PyFunction_Vectorcall + 0x6c (0x55ec6575da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x556cf40072b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x55ec6574dc5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #43: _PyFunction_Vectorcall + 0x6c (0x5576c2a73a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #51: _PyFunction_Vectorcall + 0x6c (0x556cf4014a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x556cf400d007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x5576c2a648fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default4]:[rank4]: frame #53: _PyObject_Call_Prepend + 0x69 (0x556cf401ec39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #43: _PyFunction_Vectorcall + 0x6c (0x55ec6575da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default4]:[rank4]: frame #54: + 0x211239 (0x556cf40e1239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #45: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x55ec6574e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #46: PyObject_Call + 0xbc (0x5576c2a7ff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #45: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #46: PyObject_Call + 0xbc (0x55ec65769f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: output = model(**micro_batch) [default4]:[rank4]: frame #55: PyObject_Call + 0x207 (0x556cf4021067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x55ec657502b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #48: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default3]:[rank3]: return self._call_impl(*args, **kwargs) [default4]:[rank4]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x556cf40072b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #57: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default7]:[rank7]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x5576c2a662b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x556cf40058fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #59: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #60: PyObject_Call + 0xbc (0x556cf4020f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x556cf40072b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: frame #62: + 0x150582 (0x556cf4020582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: return forward_call(*args, **kwargs) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default7]:[rank7]: frame #48: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #49: PyObject_Call + 0xbc (0x55ec65769f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: sharded_logits = self.model( [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default3]:[rank3]: return self._call_impl(*args, **kwargs) [default5]:[rank5]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x55ec657502b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default3]:[rank3]: return forward_call(*args, **kwargs) [default5]:[rank5]: frame #51: _PyFunction_Vectorcall + 0x6c (0x55ec6575da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #49: PyObject_Call + 0xbc (0x5576c2a7ff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x5576c2a662b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x55ec65756007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #53: _PyObject_Call_Prepend + 0x69 (0x55ec65767c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #51: _PyFunction_Vectorcall + 0x6c (0x5576c2a73a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x5576c2a6c007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #54: + 0x211239 (0x55ec6582a239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #55: PyObject_Call + 0x207 (0x55ec6576a067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x55ec657502b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default3]:[rank3]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default7]:[rank7]: frame #53: _PyObject_Call_Prepend + 0x69 (0x5576c2a7dc39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #54: + 0x211239 (0x5576c2b40239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default4]:[rank4]: frame #63: PyObject_Call + 0xbc (0x556cf4020f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank4]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default3]:[rank3]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank5]: frame #57: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #55: PyObject_Call + 0x207 (0x5576c2a80067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: return self._call_impl(*args, **kwargs) [default7]:[rank7]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x5576c2a662b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x55ec6574e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #57: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank5]: frame #59: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: return forward_call(*args, **kwargs) [default5]:[rank5]: frame #60: PyObject_Call + 0xbc (0x55ec65769f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default7]:[rank7]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x5576c2a648fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #59: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: frame #60: PyObject_Call + 0xbc (0x5576c2a7ff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default5]:[rank5]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x55ec657502b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default5]:[rank5]: frame #62: + 0x150582 (0x55ec65769582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: pipeline_state.run_communication() [default7]:[rank7]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x5576c2a662b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank5]: frame #63: PyObject_Call + 0xbc (0x55ec65769f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default7]:[rank7]: frame #62: + 0x150582 (0x5576c2a7f582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: recv_activation_tensor = recv_activation() [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default7]:[rank7]: frame #63: PyObject_Call + 0xbc (0x5576c2a7ff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default7]:[rank7]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default3]:[rank3]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default3]:[rank3]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default3]:[rank3]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default5]:[rank5]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default3]:[rank3]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default3]:[rank3]: dist.recv( [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default3]:[rank3]: return func(*args, **kwargs) [default3]:[rank3]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default3]:[rank3]: pg.recv([tensor], group_src_rank, tag).wait() [default3]:[rank3]: torch.distributed.DistBackendError: [1] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0:1', but store->get('0:1') got error: Connection reset by peer [default3]:[rank3]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default3]:[rank3]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f10e8b04897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default3]:[rank3]: frame #1: + 0x5b3a23e (0x7f112262123e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f112261bc87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f112261bf82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f112261cfd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f11225d1371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f11225d1371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f11225d1371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f11225d1371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f10e9dde189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default3]:[rank3]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f10e9de5610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default3]:[rank3]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f10e9e04978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default3]:[rank3]: frame #12: + 0x5adc309 (0x7f11225c3309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #13: + 0x5ae6f10 (0x7f11225cdf10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #14: + 0x5ae6fa5 (0x7f11225cdfa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #15: + 0x5124446 (0x7f1121c0b446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #16: + 0x1acf4b8 (0x7f111e5b64b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #17: + 0x5aee004 (0x7f11225d5004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #18: + 0x5af36b5 (0x7f11225da6b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank3]: frame #19: + 0xd2631e (0x7f11351c431e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default3]:[rank3]: frame #20: + 0x47def4 (0x7f113491bef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default3]:[rank3]: frame #21: + 0x1445a6 (0x55dbc2bc05a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #22: _PyObject_MakeTpCall + 0x26b (0x55dbc2bb9a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #23: + 0x150866 (0x55dbc2bcc866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x55dbc2bb5142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #25: _PyFunction_Vectorcall + 0x6c (0x55dbc2bc0a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #26: PyObject_Call + 0xbc (0x55dbc2bccf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x55dbc2bb32b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #28: _PyFunction_Vectorcall + 0x6c (0x55dbc2bc0a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x55dbc2bb18fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #30: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x55dbc2bb18fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #32: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x55dbc2bb18fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #34: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x55dbc2bb18fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x55dbc2bb8f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #37: _PyObject_Call_Prepend + 0x69 (0x55dbc2bcac39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #38: + 0x211239 (0x55dbc2c8d239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #39: _PyObject_MakeTpCall + 0x26b (0x55dbc2bb9a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x55dbc2bb53e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #41: _PyFunction_Vectorcall + 0x6c (0x55dbc2bc0a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x55dbc2bb0c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #43: _PyFunction_Vectorcall + 0x6c (0x55dbc2bc0a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x55dbc2bb18fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #45: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #46: PyObject_Call + 0xbc (0x55dbc2bccf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x55dbc2bb32b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #48: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #49: PyObject_Call + 0xbc (0x55dbc2bccf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x55dbc2bb32b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #51: _PyFunction_Vectorcall + 0x6c (0x55dbc2bc0a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x55dbc2bb9007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #53: _PyObject_Call_Prepend + 0x69 (0x55dbc2bcac39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #54: + 0x211239 (0x55dbc2c8d239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #55: PyObject_Call + 0x207 (0x55dbc2bcd067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x55dbc2bb32b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #57: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x55dbc2bb18fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #59: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #60: PyObject_Call + 0xbc (0x55dbc2bccf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x55dbc2bb32b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #62: + 0x150582 (0x55dbc2bcc582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: frame #63: PyObject_Call + 0xbc (0x55dbc2bccf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank3]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default0]:[rank8]: Traceback (most recent call last): [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default0]:[rank8]: trainer.train(dataloader) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default0]:[rank8]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default0]:[rank8]: outputs = self.pipeline_engine.train_batch_iter( [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default0]:[rank8]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default0]:[rank8]: output = model(**micro_batch) [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default0]:[rank8]: return self._call_impl(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default0]:[rank8]: return forward_call(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default0]:[rank8]: sharded_logits = self.model( [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default0]:[rank8]: return self._call_impl(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default0]:[rank8]: return forward_call(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default0]:[rank8]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default0]:[rank8]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default0]:[rank8]: return self._call_impl(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default0]:[rank8]: return forward_call(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default0]:[rank8]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default0]:[rank8]: pipeline_state.run_communication() [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default0]:[rank8]: recv_activation_tensor = recv_activation() [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default0]:[rank8]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default0]:[rank8]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default0]:[rank8]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default0]:[rank8]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default0]:[rank8]: dist.recv( [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default0]:[rank8]: return func(*args, **kwargs) [default0]:[rank8]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default0]:[rank8]: pg.recv([tensor], group_src_rank, tag).wait() [default0]:[rank8]: torch.distributed.DistBackendError: [4] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '3:4', but store->get('3:4') got error: Connection reset by peer [default0]:[rank8]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default0]:[rank8]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f21f6b82897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default0]:[rank8]: frame #1: + 0x5b3a23e (0x7f223069f23e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f2230699c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f2230699f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f223069afd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f223064f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f223064f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f223064f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f223064f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f21f7e5c189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default0]:[rank8]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f21f7e63610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default0]:[rank8]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f21f7e82978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default0]:[rank8]: frame #12: + 0x5adc309 (0x7f2230641309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #13: + 0x5ae6f10 (0x7f223064bf10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #14: + 0x5ae6fa5 (0x7f223064bfa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #15: + 0x5124446 (0x7f222fc89446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #16: + 0x1acf4b8 (0x7f222c6344b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #17: + 0x5aee004 (0x7f2230653004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #18: + 0x5af36b5 (0x7f22306586b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default0]:[rank8]: frame #19: + 0xd2631e (0x7f224324231e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default0]:[rank8]: frame #20: + 0x47def4 (0x7f2242999ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default0]:[rank8]: frame #21: + 0x1445a6 (0x556bba3965a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #22: _PyObject_MakeTpCall + 0x26b (0x556bba38fa6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #23: + 0x150866 (0x556bba3a2866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x556bba38b142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #25: _PyFunction_Vectorcall + 0x6c (0x556bba396a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #26: PyObject_Call + 0xbc (0x556bba3a2f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x556bba3892b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #28: _PyFunction_Vectorcall + 0x6c (0x556bba396a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x556bba3878fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #30: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x556bba3878fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #32: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x556bba3878fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #34: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x556bba3878fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x556bba38ef50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #37: _PyObject_Call_Prepend + 0x69 (0x556bba3a0c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #38: + 0x211239 (0x556bba463239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #39: _PyObject_MakeTpCall + 0x26b (0x556bba38fa6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x556bba38b3e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #41: _PyFunction_Vectorcall + 0x6c (0x556bba396a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x556bba386c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #43: _PyFunction_Vectorcall + 0x6c (0x556bba396a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x556bba3878fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #45: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #46: PyObject_Call + 0xbc (0x556bba3a2f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x556bba3892b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #48: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #49: PyObject_Call + 0xbc (0x556bba3a2f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x556bba3892b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #51: _PyFunction_Vectorcall + 0x6c (0x556bba396a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x556bba38f007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #53: _PyObject_Call_Prepend + 0x69 (0x556bba3a0c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #54: + 0x211239 (0x556bba463239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #55: PyObject_Call + 0x207 (0x556bba3a3067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x556bba3892b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #57: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x556bba3878fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #59: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #60: PyObject_Call + 0xbc (0x556bba3a2f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x556bba3892b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #62: + 0x150582 (0x556bba3a2582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: frame #63: PyObject_Call + 0xbc (0x556bba3a2f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default0]:[rank8]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default3]:[rank11]: Traceback (most recent call last): [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default3]:[rank11]: trainer.train(dataloader) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default3]:[rank11]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default3]:[rank11]: outputs = self.pipeline_engine.train_batch_iter( [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default3]:[rank11]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default3]:[rank11]: output = model(**micro_batch) [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default3]:[rank11]: return self._call_impl(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default3]:[rank11]: return forward_call(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default3]:[rank11]: sharded_logits = self.model( [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default3]:[rank11]: return self._call_impl(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default3]:[rank11]: return forward_call(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default3]:[rank11]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default3]:[rank11]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default3]:[rank11]: return self._call_impl(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default3]:[rank11]: return forward_call(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default3]:[rank11]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default3]:[rank11]: pipeline_state.run_communication() [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default3]:[rank11]: recv_activation_tensor = recv_activation() [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default3]:[rank11]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default3]:[rank11]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default3]:[rank11]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default3]:[rank11]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default3]:[rank11]: dist.recv( [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default3]:[rank11]: return func(*args, **kwargs) [default3]:[rank11]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default3]:[rank11]: pg.recv([tensor], group_src_rank, tag).wait() [default3]:[rank11]: torch.distributed.DistBackendError: [5] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '4:5', but store->get('4:5') got error: Connection reset by peer [default3]:[rank11]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default3]:[rank11]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f67a2e52897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default3]:[rank11]: frame #1: + 0x5b3a23e (0x7f67dc96f23e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f67dc969c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f67dc969f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f67dc96afd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f67dc91f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f67dc91f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f67dc91f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f67dc91f371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f67a412c189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default3]:[rank11]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f67a4133610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default3]:[rank11]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f67a4152978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default3]:[rank11]: frame #12: + 0x5adc309 (0x7f67dc911309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #13: + 0x5ae6f10 (0x7f67dc91bf10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #14: + 0x5ae6fa5 (0x7f67dc91bfa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #15: + 0x5124446 (0x7f67dbf59446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #16: + 0x1acf4b8 (0x7f67d89044b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #17: + 0x5aee004 (0x7f67dc923004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #18: + 0x5af36b5 (0x7f67dc9286b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default3]:[rank11]: frame #19: + 0xd2631e (0x7f67ef51231e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default3]:[rank11]: frame #20: + 0x47def4 (0x7f67eec69ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default3]:[rank11]: frame #21: + 0x1445a6 (0x5562fabe35a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #22: _PyObject_MakeTpCall + 0x26b (0x5562fabdca6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #23: + 0x150866 (0x5562fabef866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x5562fabd8142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #25: _PyFunction_Vectorcall + 0x6c (0x5562fabe3a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #26: PyObject_Call + 0xbc (0x5562fabeff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x5562fabd62b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #28: _PyFunction_Vectorcall + 0x6c (0x5562fabe3a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x5562fabd48fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #30: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x5562fabd48fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #32: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x5562fabd48fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #34: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x5562fabd48fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x5562fabdbf50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #37: _PyObject_Call_Prepend + 0x69 (0x5562fabedc39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #38: + 0x211239 (0x5562facb0239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #39: _PyObject_MakeTpCall + 0x26b (0x5562fabdca6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x5562fabd83e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #41: _PyFunction_Vectorcall + 0x6c (0x5562fabe3a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x5562fabd3c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #43: _PyFunction_Vectorcall + 0x6c (0x5562fabe3a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x5562fabd48fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #45: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #46: PyObject_Call + 0xbc (0x5562fabeff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x5562fabd62b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #48: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #49: PyObject_Call + 0xbc (0x5562fabeff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x5562fabd62b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #51: _PyFunction_Vectorcall + 0x6c (0x5562fabe3a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x5562fabdc007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #53: _PyObject_Call_Prepend + 0x69 (0x5562fabedc39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #54: + 0x211239 (0x5562facb0239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #55: PyObject_Call + 0x207 (0x5562fabf0067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x5562fabd62b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #57: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x5562fabd48fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #59: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #60: PyObject_Call + 0xbc (0x5562fabeff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x5562fabd62b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #62: + 0x150582 (0x5562fabef582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: frame #63: PyObject_Call + 0xbc (0x5562fabeff1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default3]:[rank11]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default6]:[rank6]: Traceback (most recent call last): [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default6]:[rank6]: trainer.train(dataloader) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default6]:[rank6]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default6]:[rank6]: outputs = self.pipeline_engine.train_batch_iter( [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default6]:[rank6]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default6]:[rank6]: output = model(**micro_batch) [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default6]:[rank6]: return self._call_impl(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default6]:[rank6]: return forward_call(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default6]:[rank6]: sharded_logits = self.model( [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default6]:[rank6]: return self._call_impl(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default6]:[rank6]: return forward_call(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default6]:[rank6]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default6]:[rank6]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default6]:[rank6]: return self._call_impl(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default6]:[rank6]: return forward_call(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default6]:[rank6]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default6]:[rank6]: pipeline_state.run_communication() [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default6]:[rank6]: recv_activation_tensor = recv_activation() [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default6]:[rank6]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default6]:[rank6]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default6]:[rank6]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default6]:[rank6]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default6]:[rank6]: dist.recv( [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default6]:[rank6]: return func(*args, **kwargs) [default6]:[rank6]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default6]:[rank6]: pg.recv([tensor], group_src_rank, tag).wait() [default6]:[rank6]: torch.distributed.DistBackendError: [3] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '2:3', but store->get('2:3') got error: Connection reset by peer [default6]:[rank6]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default6]:[rank6]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f51b021c897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default6]:[rank6]: frame #1: + 0x5b3a23e (0x7f51e9d3923e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f51e9d33c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f51e9d33f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f51e9d34fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f51e9ce9371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f51e9ce9371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f51e9ce9371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f51e9ce9371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f51b14f6189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default6]:[rank6]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f51b14fd610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default6]:[rank6]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f51b151c978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default6]:[rank6]: frame #12: + 0x5adc309 (0x7f51e9cdb309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #13: + 0x5ae6f10 (0x7f51e9ce5f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #14: + 0x5ae6fa5 (0x7f51e9ce5fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #15: + 0x5124446 (0x7f51e9323446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #16: + 0x1acf4b8 (0x7f51e5cce4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #17: + 0x5aee004 (0x7f51e9ced004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #18: + 0x5af36b5 (0x7f51e9cf26b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default6]:[rank6]: frame #19: + 0xd2631e (0x7f51fc8dc31e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default6]:[rank6]: frame #20: + 0x47def4 (0x7f51fc033ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default6]:[rank6]: frame #21: + 0x1445a6 (0x56299b6f45a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #22: _PyObject_MakeTpCall + 0x26b (0x56299b6eda6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #23: + 0x150866 (0x56299b700866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x56299b6e9142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #25: _PyFunction_Vectorcall + 0x6c (0x56299b6f4a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #26: PyObject_Call + 0xbc (0x56299b700f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x56299b6e72b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: Traceback (most recent call last): [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default6]:[rank6]: frame #28: _PyFunction_Vectorcall + 0x6c (0x56299b6f4a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x56299b6e58fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #30: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x56299b6e58fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #32: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x56299b6e58fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #34: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: trainer.train(dataloader) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default6]:[rank6]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x56299b6e58fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default6]:[rank6]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x56299b6ecf50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #37: _PyObject_Call_Prepend + 0x69 (0x56299b6fec39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #38: + 0x211239 (0x56299b7c1239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: outputs = self.pipeline_engine.train_batch_iter( [default6]:[rank6]: frame #39: _PyObject_MakeTpCall + 0x26b (0x56299b6eda6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x56299b6e93e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #41: _PyFunction_Vectorcall + 0x6c (0x56299b6f4a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x56299b6e4c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #43: _PyFunction_Vectorcall + 0x6c (0x56299b6f4a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default2]:[rank2]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default2]:[rank2]: output = model(**micro_batch) [default6]:[rank6]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x56299b6e58fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default2]:[rank2]: return self._call_impl(*args, **kwargs) [default6]:[rank6]: frame #45: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #46: PyObject_Call + 0xbc (0x56299b700f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x56299b6e72b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #48: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #49: PyObject_Call + 0xbc (0x56299b700f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x56299b6e72b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #51: _PyFunction_Vectorcall + 0x6c (0x56299b6f4a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x56299b6ed007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #53: _PyObject_Call_Prepend + 0x69 (0x56299b6fec39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #54: + 0x211239 (0x56299b7c1239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #55: PyObject_Call + 0x207 (0x56299b701067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x56299b6e72b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #57: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x56299b6e58fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default2]:[rank2]: return forward_call(*args, **kwargs) [default6]:[rank6]: frame #59: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #60: PyObject_Call + 0xbc (0x56299b700f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default2]:[rank2]: sharded_logits = self.model( [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default6]:[rank6]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x56299b6e72b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: return self._call_impl(*args, **kwargs) [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default2]:[rank2]: return forward_call(*args, **kwargs) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default2]:[rank2]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default2]:[rank2]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default6]:[rank6]: frame #62: + 0x150582 (0x56299b700582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: frame #63: PyObject_Call + 0xbc (0x56299b700f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default6]:[rank6]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default2]:[rank2]: return self._call_impl(*args, **kwargs) [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default2]:[rank2]: return forward_call(*args, **kwargs) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default2]:[rank2]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default2]:[rank2]: pipeline_state.run_communication() [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default2]:[rank2]: recv_activation_tensor = recv_activation() [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default2]:[rank2]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default2]:[rank2]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default2]:[rank2]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default2]:[rank2]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default2]:[rank2]: dist.recv( [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default2]:[rank2]: return func(*args, **kwargs) [default2]:[rank2]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default2]:[rank2]: pg.recv([tensor], group_src_rank, tag).wait() [default2]:[rank2]: torch.distributed.DistBackendError: [1] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '0:1', but store->get('0:1') got error: Connection reset by peer [default2]:[rank2]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default2]:[rank2]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f7b2bb6c897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default2]:[rank2]: frame #1: + 0x5b3a23e (0x7f7b6568923e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f7b65683c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f7b65683f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f7b65684fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f7b65639371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f7b65639371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f7b65639371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f7b65639371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f7b2ce46189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default2]:[rank2]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f7b2ce4d610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default2]:[rank2]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f7b2ce6c978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default2]:[rank2]: frame #12: + 0x5adc309 (0x7f7b6562b309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #13: + 0x5ae6f10 (0x7f7b65635f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #14: + 0x5ae6fa5 (0x7f7b65635fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #15: + 0x5124446 (0x7f7b64c73446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #16: + 0x1acf4b8 (0x7f7b6161e4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #17: + 0x5aee004 (0x7f7b6563d004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #18: + 0x5af36b5 (0x7f7b656426b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank2]: frame #19: + 0xd2631e (0x7f7b7822c31e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default2]:[rank2]: frame #20: + 0x47def4 (0x7f7b77983ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default2]:[rank2]: frame #21: + 0x1445a6 (0x5611b494b5a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #22: _PyObject_MakeTpCall + 0x26b (0x5611b4944a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #23: + 0x150866 (0x5611b4957866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x5611b4940142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #25: _PyFunction_Vectorcall + 0x6c (0x5611b494ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #26: PyObject_Call + 0xbc (0x5611b4957f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x5611b493e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #28: _PyFunction_Vectorcall + 0x6c (0x5611b494ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x5611b493c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #30: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x5611b493c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #32: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x5611b493c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #34: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x5611b493c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x5611b4943f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #37: _PyObject_Call_Prepend + 0x69 (0x5611b4955c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #38: + 0x211239 (0x5611b4a18239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #39: _PyObject_MakeTpCall + 0x26b (0x5611b4944a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x5611b49403e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #41: _PyFunction_Vectorcall + 0x6c (0x5611b494ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x5611b493bc5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #43: _PyFunction_Vectorcall + 0x6c (0x5611b494ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x5611b493c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #45: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #46: PyObject_Call + 0xbc (0x5611b4957f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x5611b493e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #48: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #49: PyObject_Call + 0xbc (0x5611b4957f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x5611b493e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #51: _PyFunction_Vectorcall + 0x6c (0x5611b494ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x5611b4944007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #53: _PyObject_Call_Prepend + 0x69 (0x5611b4955c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #54: + 0x211239 (0x5611b4a18239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #55: PyObject_Call + 0x207 (0x5611b4958067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x5611b493e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #57: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x5611b493c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #59: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #60: PyObject_Call + 0xbc (0x5611b4957f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x5611b493e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #62: + 0x150582 (0x5611b4957582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: frame #63: PyObject_Call + 0xbc (0x5611b4957f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank2]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default1]:[rank9]: Traceback (most recent call last): [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default1]:[rank9]: trainer.train(dataloader) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default1]:[rank9]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default1]:[rank9]: outputs = self.pipeline_engine.train_batch_iter( [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default1]:[rank9]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default1]:[rank9]: output = model(**micro_batch) [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default1]:[rank9]: return self._call_impl(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default1]:[rank9]: return forward_call(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default1]:[rank9]: sharded_logits = self.model( [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default1]:[rank9]: return self._call_impl(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default1]:[rank9]: return forward_call(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default1]:[rank9]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default1]:[rank9]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default1]:[rank9]: return self._call_impl(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default1]:[rank9]: return forward_call(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default1]:[rank9]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default1]:[rank9]: pipeline_state.run_communication() [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default1]:[rank9]: recv_activation_tensor = recv_activation() [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default1]:[rank9]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default1]:[rank9]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default1]:[rank9]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default1]:[rank9]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default1]:[rank9]: dist.recv( [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default1]:[rank9]: return func(*args, **kwargs) [default1]:[rank9]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default1]:[rank9]: pg.recv([tensor], group_src_rank, tag).wait() [default1]:[rank9]: torch.distributed.DistBackendError: [4] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '3:4', but store->get('3:4') got error: Connection reset by peer [default1]:[rank9]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default1]:[rank9]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f5de5759897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default1]:[rank9]: frame #1: + 0x5b3a23e (0x7f5e1f27623e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f5e1f270c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f5e1f270f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f5e1f271fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f5e1f226371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f5e1f226371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f5e1f226371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f5e1f226371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f5de6a33189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default1]:[rank9]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f5de6a3a610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default1]:[rank9]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f5de6a59978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default1]:[rank9]: frame #12: + 0x5adc309 (0x7f5e1f218309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #13: + 0x5ae6f10 (0x7f5e1f222f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #14: + 0x5ae6fa5 (0x7f5e1f222fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #15: + 0x5124446 (0x7f5e1e860446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #16: + 0x1acf4b8 (0x7f5e1b20b4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #17: + 0x5aee004 (0x7f5e1f22a004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #18: + 0x5af36b5 (0x7f5e1f22f6b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default1]:[rank9]: frame #19: + 0xd2631e (0x7f5e31e1931e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default1]:[rank9]: frame #20: + 0x47def4 (0x7f5e31570ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default1]:[rank9]: frame #21: + 0x1445a6 (0x5630b396b5a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #22: _PyObject_MakeTpCall + 0x26b (0x5630b3964a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #23: + 0x150866 (0x5630b3977866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x5630b3960142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #25: _PyFunction_Vectorcall + 0x6c (0x5630b396ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #26: PyObject_Call + 0xbc (0x5630b3977f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x5630b395e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #28: _PyFunction_Vectorcall + 0x6c (0x5630b396ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x5630b395c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #30: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x5630b395c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #32: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x5630b395c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #34: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x5630b395c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x5630b3963f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #37: _PyObject_Call_Prepend + 0x69 (0x5630b3975c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #38: + 0x211239 (0x5630b3a38239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #39: _PyObject_MakeTpCall + 0x26b (0x5630b3964a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x5630b39603e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #41: _PyFunction_Vectorcall + 0x6c (0x5630b396ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x5630b395bc5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #43: _PyFunction_Vectorcall + 0x6c (0x5630b396ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x5630b395c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #45: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #46: PyObject_Call + 0xbc (0x5630b3977f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x5630b395e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #48: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #49: PyObject_Call + 0xbc (0x5630b3977f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x5630b395e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #51: _PyFunction_Vectorcall + 0x6c (0x5630b396ba2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x5630b3964007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #53: _PyObject_Call_Prepend + 0x69 (0x5630b3975c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #54: + 0x211239 (0x5630b3a38239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #55: PyObject_Call + 0x207 (0x5630b3978067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x5630b395e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #57: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x5630b395c8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #59: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #60: PyObject_Call + 0xbc (0x5630b3977f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x5630b395e2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #62: + 0x150582 (0x5630b3977582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: frame #63: PyObject_Call + 0xbc (0x5630b3977f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default1]:[rank9]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default2]:[rank10]: Traceback (most recent call last): [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default2]:[rank10]: trainer.train(dataloader) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default2]:[rank10]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default2]:[rank10]: outputs = self.pipeline_engine.train_batch_iter( [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default2]:[rank10]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default2]:[rank10]: output = model(**micro_batch) [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default2]:[rank10]: return self._call_impl(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default2]:[rank10]: return forward_call(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default2]:[rank10]: sharded_logits = self.model( [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default2]:[rank10]: return self._call_impl(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default2]:[rank10]: return forward_call(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default2]:[rank10]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default2]:[rank10]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default2]:[rank10]: return self._call_impl(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default2]:[rank10]: return forward_call(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default2]:[rank10]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default2]:[rank10]: pipeline_state.run_communication() [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default2]:[rank10]: recv_activation_tensor = recv_activation() [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default2]:[rank10]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default2]:[rank10]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default2]:[rank10]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default2]:[rank10]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default2]:[rank10]: dist.recv( [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default2]:[rank10]: return func(*args, **kwargs) [default2]:[rank10]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default2]:[rank10]: pg.recv([tensor], group_src_rank, tag).wait() [default2]:[rank10]: torch.distributed.DistBackendError: [5] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '4:5', but store->get('4:5') got error: Connection reset by peer [default2]:[rank10]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default2]:[rank10]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7f9bf13da897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default2]:[rank10]: frame #1: + 0x5b3a23e (0x7f9c2aef723e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7f9c2aef1c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7f9c2aef1f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7f9c2aef2fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f9c2aea7371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f9c2aea7371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f9c2aea7371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7f9c2aea7371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7f9bf26b4189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default2]:[rank10]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7f9bf26bb610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default2]:[rank10]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7f9bf26da978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default2]:[rank10]: frame #12: + 0x5adc309 (0x7f9c2ae99309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #13: + 0x5ae6f10 (0x7f9c2aea3f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #14: + 0x5ae6fa5 (0x7f9c2aea3fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #15: + 0x5124446 (0x7f9c2a4e1446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #16: + 0x1acf4b8 (0x7f9c26e8c4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #17: + 0x5aee004 (0x7f9c2aeab004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #18: + 0x5af36b5 (0x7f9c2aeb06b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default2]:[rank10]: frame #19: + 0xd2631e (0x7f9c3da9a31e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default2]:[rank10]: frame #20: + 0x47def4 (0x7f9c3d1f1ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default2]:[rank10]: frame #21: + 0x1445a6 (0x55eaf4f205a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #22: _PyObject_MakeTpCall + 0x26b (0x55eaf4f19a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #23: + 0x150866 (0x55eaf4f2c866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x55eaf4f15142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #25: _PyFunction_Vectorcall + 0x6c (0x55eaf4f20a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #26: PyObject_Call + 0xbc (0x55eaf4f2cf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x55eaf4f132b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #28: _PyFunction_Vectorcall + 0x6c (0x55eaf4f20a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x55eaf4f118fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #30: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x55eaf4f118fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #32: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x55eaf4f118fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #34: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x55eaf4f118fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x55eaf4f18f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #37: _PyObject_Call_Prepend + 0x69 (0x55eaf4f2ac39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #38: + 0x211239 (0x55eaf4fed239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #39: _PyObject_MakeTpCall + 0x26b (0x55eaf4f19a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x55eaf4f153e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #41: _PyFunction_Vectorcall + 0x6c (0x55eaf4f20a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x55eaf4f10c5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #43: _PyFunction_Vectorcall + 0x6c (0x55eaf4f20a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x55eaf4f118fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #45: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #46: PyObject_Call + 0xbc (0x55eaf4f2cf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x55eaf4f132b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #48: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #49: PyObject_Call + 0xbc (0x55eaf4f2cf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x55eaf4f132b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #51: _PyFunction_Vectorcall + 0x6c (0x55eaf4f20a2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x55eaf4f19007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #53: _PyObject_Call_Prepend + 0x69 (0x55eaf4f2ac39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #54: + 0x211239 (0x55eaf4fed239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #55: PyObject_Call + 0x207 (0x55eaf4f2d067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x55eaf4f132b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #57: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x55eaf4f118fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #59: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #60: PyObject_Call + 0xbc (0x55eaf4f2cf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x55eaf4f132b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #62: + 0x150582 (0x55eaf4f2c582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: frame #63: PyObject_Call + 0xbc (0x55eaf4f2cf1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default2]:[rank10]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default4]:[rank12]: Traceback (most recent call last): [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default4]:[rank12]: trainer.train(dataloader) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default4]:[rank12]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default4]:[rank12]: outputs = self.pipeline_engine.train_batch_iter( [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default4]:[rank12]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default4]:[rank12]: output = model(**micro_batch) [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank12]: return self._call_impl(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank12]: return forward_call(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default4]:[rank12]: sharded_logits = self.model( [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank12]: return self._call_impl(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank12]: return forward_call(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default4]:[rank12]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default4]:[rank12]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default4]:[rank12]: return self._call_impl(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default4]:[rank12]: return forward_call(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default4]:[rank12]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default4]:[rank12]: pipeline_state.run_communication() [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default4]:[rank12]: recv_activation_tensor = recv_activation() [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default4]:[rank12]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default4]:[rank12]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default4]:[rank12]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default4]:[rank12]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default4]:[rank12]: dist.recv( [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default4]:[rank12]: return func(*args, **kwargs) [default4]:[rank12]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default4]:[rank12]: pg.recv([tensor], group_src_rank, tag).wait() [default4]:[rank12]: torch.distributed.DistBackendError: [6] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '5:6', but store->get('5:6') got error: Connection reset by peer [default4]:[rank12]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default4]:[rank12]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7ff21188c897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default4]:[rank12]: frame #1: + 0x5b3a23e (0x7ff24b3a923e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7ff24b3a3c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7ff24b3a3f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7ff24b3a4fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7ff24b359371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7ff24b359371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7ff24b359371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7ff24b359371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7ff212b66189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default4]:[rank12]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7ff212b6d610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default4]:[rank12]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7ff212b8c978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default4]:[rank12]: frame #12: + 0x5adc309 (0x7ff24b34b309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #13: + 0x5ae6f10 (0x7ff24b355f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #14: + 0x5ae6fa5 (0x7ff24b355fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #15: + 0x5124446 (0x7ff24a993446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #16: + 0x1acf4b8 (0x7ff24733e4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #17: + 0x5aee004 (0x7ff24b35d004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #18: + 0x5af36b5 (0x7ff24b3626b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default4]:[rank12]: frame #19: + 0xd2631e (0x7ff25df4c31e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default4]:[rank12]: frame #20: + 0x47def4 (0x7ff25d6a3ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default4]:[rank12]: frame #21: + 0x1445a6 (0x5654cee5c5a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #22: _PyObject_MakeTpCall + 0x26b (0x5654cee55a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #23: + 0x150866 (0x5654cee68866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x5654cee51142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #25: _PyFunction_Vectorcall + 0x6c (0x5654cee5ca2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #26: PyObject_Call + 0xbc (0x5654cee68f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x5654cee4f2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #28: _PyFunction_Vectorcall + 0x6c (0x5654cee5ca2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x5654cee4d8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #30: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x5654cee4d8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #32: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x5654cee4d8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #34: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x5654cee4d8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x5654cee54f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #37: _PyObject_Call_Prepend + 0x69 (0x5654cee66c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #38: + 0x211239 (0x5654cef29239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #39: _PyObject_MakeTpCall + 0x26b (0x5654cee55a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x5654cee513e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #41: _PyFunction_Vectorcall + 0x6c (0x5654cee5ca2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x5654cee4cc5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #43: _PyFunction_Vectorcall + 0x6c (0x5654cee5ca2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x5654cee4d8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #45: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #46: PyObject_Call + 0xbc (0x5654cee68f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x5654cee4f2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #48: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #49: PyObject_Call + 0xbc (0x5654cee68f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x5654cee4f2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #51: _PyFunction_Vectorcall + 0x6c (0x5654cee5ca2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x5654cee55007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #53: _PyObject_Call_Prepend + 0x69 (0x5654cee66c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #54: + 0x211239 (0x5654cef29239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #55: PyObject_Call + 0x207 (0x5654cee69067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x5654cee4f2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #57: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x5654cee4d8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #59: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #60: PyObject_Call + 0xbc (0x5654cee68f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x5654cee4f2b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #62: + 0x150582 (0x5654cee68582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: frame #63: PyObject_Call + 0xbc (0x5654cee68f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default4]:[rank12]: . This may indicate a possible application crash on rank 0 or a network set up issue. [default5]:[rank13]: Traceback (most recent call last): [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py", line 237, in [default5]:[rank13]: trainer.train(dataloader) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 429, in train [default5]:[rank13]: outputs, loss_avg = self.training_step(dataloader=self.current_dataloader) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/trainer.py", line 462, in training_step [default5]:[rank13]: outputs = self.pipeline_engine.train_batch_iter( [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 252, in train_batch_iter [default5]:[rank13]: output = self.forward(context=context, state=state, micro_batch=micro_batch, model=model) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/engine.py", line 44, in forward [default5]:[rank13]: output = model(**micro_batch) [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank13]: return self._call_impl(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank13]: return forward_call(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 891, in forward [default5]:[rank13]: sharded_logits = self.model( [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank13]: return self._call_impl(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank13]: return forward_call(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 764, in forward [default5]:[rank13]: return self.forward_with_hidden_states(input_ids=input_ids, input_mask=input_mask)[0] [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/models/llama.py", line 780, in forward_with_hidden_states [default5]:[rank13]: hidden_encoder_states = encoder_block(**hidden_encoder_states) [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl [default5]:[rank13]: return self._call_impl(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl [default5]:[rank13]: return forward_call(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/block.py", line 126, in forward [default5]:[rank13]: new_kwargs[name] = recv_from_pipeline_state_buffer( [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/functional.py", line 117, in recv_from_pipeline_state_buffer [default5]:[rank13]: pipeline_state.run_communication() [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 150, in run_communication [default5]:[rank13]: recv_activation_tensor = recv_activation() [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/state.py", line 31, in __call__ [default5]:[rank13]: return self.p2p.recv_tensors(num_tensors=1, from_rank=self.from_rank)[0] [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 353, in recv_tensors [default5]:[rank13]: buffers, futures = self.irecv_tensors(num_tensors=num_tensors, from_rank=from_rank, tag=tag) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 326, in irecv_tensors [default5]:[rank13]: meta = self._recv_meta(from_rank=from_rank, tag=tag) [default5]:[rank13]: File "/fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/src/nanotron/parallel/pipeline_parallel/p2p.py", line 246, in _recv_meta [default5]:[rank13]: dist.recv( [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 75, in wrapper [default5]:[rank13]: return func(*args, **kwargs) [default5]:[rank13]: File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 1932, in recv [default5]:[rank13]: pg.recv([tensor], group_src_rank, tag).wait() [default5]:[rank13]: torch.distributed.DistBackendError: [6] is setting up NCCL communicator and retrieving ncclUniqueId from [0] via c10d key-value store by key '5:6', but store->get('5:6') got error: Connection reset by peer [default5]:[rank13]: Exception raised from recvBytes at ../torch/csrc/distributed/c10d/Utils.hpp:672 (most recent call first): [default5]:[rank13]: frame #0: c10::Error::Error(c10::SourceLocation, std::string) + 0x57 (0x7fe9ac0cc897 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libc10.so) [default5]:[rank13]: frame #1: + 0x5b3a23e (0x7fe9e5be923e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #2: c10d::TCPStore::doWait(c10::ArrayRef, std::chrono::duration >) + 0x2c7 (0x7fe9e5be3c87 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #3: c10d::TCPStore::doGet(std::string const&) + 0x32 (0x7fe9e5be3f82 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #4: c10d::TCPStore::get(std::string const&) + 0xa1 (0x7fe9e5be4fd1 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #5: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fe9e5b99371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #6: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fe9e5b99371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #7: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fe9e5b99371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #8: c10d::PrefixStore::get(std::string const&) + 0x31 (0x7fe9e5b99371 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #9: c10d::ProcessGroupNCCL::broadcastUniqueNCCLID(ncclUniqueId*, bool, std::string const&, int) + 0xa9 (0x7fe9ad3a6189 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default5]:[rank13]: frame #10: c10d::ProcessGroupNCCL::getNCCLComm(std::string const&, c10::Device&, c10d::OpType, int, bool) + 0xc50 (0x7fe9ad3ad610 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default5]:[rank13]: frame #11: c10d::ProcessGroupNCCL::recv(std::vector >&, int, int) + 0x5f8 (0x7fe9ad3cc978 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cuda.so) [default5]:[rank13]: frame #12: + 0x5adc309 (0x7fe9e5b8b309 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #13: + 0x5ae6f10 (0x7fe9e5b95f10 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #14: + 0x5ae6fa5 (0x7fe9e5b95fa5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #15: + 0x5124446 (0x7fe9e51d3446 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #16: + 0x1acf4b8 (0x7fe9e1b7e4b8 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #17: + 0x5aee004 (0x7fe9e5b9d004 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #18: + 0x5af36b5 (0x7fe9e5ba26b5 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_cpu.so) [default5]:[rank13]: frame #19: + 0xd2631e (0x7fe9f878c31e in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default5]:[rank13]: frame #20: + 0x47def4 (0x7fe9f7ee3ef4 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/lib/libtorch_python.so) [default5]:[rank13]: frame #21: + 0x1445a6 (0x55f177a1d5a6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #22: _PyObject_MakeTpCall + 0x26b (0x55f177a16a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #23: + 0x150866 (0x55f177a29866 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #24: _PyEval_EvalFrameDefault + 0x4c12 (0x55f177a12142 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #25: _PyFunction_Vectorcall + 0x6c (0x55f177a1da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #26: PyObject_Call + 0xbc (0x55f177a29f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #27: _PyEval_EvalFrameDefault + 0x2d83 (0x55f177a102b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #28: _PyFunction_Vectorcall + 0x6c (0x55f177a1da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #29: _PyEval_EvalFrameDefault + 0x13ca (0x55f177a0e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #30: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #31: _PyEval_EvalFrameDefault + 0x13ca (0x55f177a0e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #32: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #33: _PyEval_EvalFrameDefault + 0x13ca (0x55f177a0e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #34: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #35: _PyEval_EvalFrameDefault + 0x13ca (0x55f177a0e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #36: _PyObject_FastCallDictTstate + 0xd0 (0x55f177a15f50 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #37: _PyObject_Call_Prepend + 0x69 (0x55f177a27c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #38: + 0x211239 (0x55f177aea239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #39: _PyObject_MakeTpCall + 0x26b (0x55f177a16a6b in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #40: _PyEval_EvalFrameDefault + 0x4eb6 (0x55f177a123e6 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #41: _PyFunction_Vectorcall + 0x6c (0x55f177a1da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #42: _PyEval_EvalFrameDefault + 0x72c (0x55f177a0dc5c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #43: _PyFunction_Vectorcall + 0x6c (0x55f177a1da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #44: _PyEval_EvalFrameDefault + 0x13ca (0x55f177a0e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #45: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #46: PyObject_Call + 0xbc (0x55f177a29f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #47: _PyEval_EvalFrameDefault + 0x2d83 (0x55f177a102b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #48: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #49: PyObject_Call + 0xbc (0x55f177a29f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #50: _PyEval_EvalFrameDefault + 0x2d83 (0x55f177a102b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #51: _PyFunction_Vectorcall + 0x6c (0x55f177a1da2c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #52: _PyObject_FastCallDictTstate + 0x187 (0x55f177a16007 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #53: _PyObject_Call_Prepend + 0x69 (0x55f177a27c39 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #54: + 0x211239 (0x55f177aea239 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #55: PyObject_Call + 0x207 (0x55f177a2a067 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #56: _PyEval_EvalFrameDefault + 0x2d83 (0x55f177a102b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #57: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #58: _PyEval_EvalFrameDefault + 0x13ca (0x55f177a0e8fa in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #59: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #60: PyObject_Call + 0xbc (0x55f177a29f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #61: _PyEval_EvalFrameDefault + 0x2d83 (0x55f177a102b3 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #62: + 0x150582 (0x55f177a29582 in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: frame #63: PyObject_Call + 0xbc (0x55f177a29f1c in /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10) [default5]:[rank13]: . This may indicate a possible application crash on rank 0 or a network set up issue. W0702 21:07:06.875000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 829693 closing signal SIGTERM W0702 21:07:06.876000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 829694 closing signal SIGTERM W0702 21:07:06.876000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 829695 closing signal SIGTERM W0702 21:07:06.876000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 829696 closing signal SIGTERM W0702 21:07:06.878000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 829697 closing signal SIGTERM W0702 21:07:06.880000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 852498 closing signal SIGTERM W0702 21:07:06.879000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 829698 closing signal SIGTERM W0702 21:07:06.881000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 852499 closing signal SIGTERM W0702 21:07:06.881000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 852500 closing signal SIGTERM W0702 21:07:06.882000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 852501 closing signal SIGTERM W0702 21:07:06.882000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 852502 closing signal SIGTERM W0702 21:07:06.882000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:851] Sending process 852503 closing signal SIGTERM E0702 21:07:12.175000 139997183711040 torch/distributed/elastic/multiprocessing/api.py:826] failed (exitcode: -6) local_rank: 0 (pid: 829691) of binary: /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10 Traceback (most recent call last): File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/torchrun", line 8, in sys.exit(main()) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 347, in wrapper return f(*args, **kwargs) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 879, in main run(args) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 870, in run elastic_launch( File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 132, in __call__ return launch_agent(self._config, self._entrypoint, list(args)) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 263, in launch_agent raise ChildFailedError( torch.distributed.elastic.multiprocessing.errors.ChildFailedError: ============================================================ /fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py FAILED ------------------------------------------------------------ Failures: [1]: time : 2024-07-02_21:07:06 host : ip-26-0-165-24.ec2.internal rank : 1 (local_rank: 1) exitcode : -6 (pid: 829692) error_file: traceback : Signal 6 (SIGABRT) received by PID 829692 ------------------------------------------------------------ Root Cause (first observed failure): [0]: time : 2024-07-02_21:07:06 host : ip-26-0-165-24.ec2.internal rank : 0 (local_rank: 0) exitcode : -6 (pid: 829691) error_file: traceback : Signal 6 (SIGABRT) received by PID 829691 ============================================================ E0702 21:07:12.668000 139648004355904 torch/distributed/elastic/multiprocessing/api.py:826] failed (exitcode: -6) local_rank: 6 (pid: 852504) of binary: /fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/python3.10 W0702 21:07:12.686000 139648004355904 torch/distributed/elastic/rendezvous/dynamic_rendezvous.py:1203] The node 'ip-26-0-170-160.ec2.internal_852424_0' has failed to shutdown the rendezvous 'none' due to an error of type RendezvousConnectionError. W0702 21:07:12.718000 139648004355904 torch/distributed/elastic/rendezvous/dynamic_rendezvous.py:1203] The node 'ip-26-0-170-160.ec2.internal_852424_0' has failed to shutdown the rendezvous 'none' due to an error of type RendezvousConnectionError. W0702 21:07:12.729000 139648004355904 torch/distributed/elastic/rendezvous/dynamic_rendezvous.py:1203] The node 'ip-26-0-170-160.ec2.internal_852424_0' has failed to shutdown the rendezvous 'none' due to an error of type RendezvousConnectionError. Traceback (most recent call last): File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/bin/torchrun", line 8, in sys.exit(main()) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 347, in wrapper return f(*args, **kwargs) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 879, in main run(args) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/run.py", line 870, in run elastic_launch( File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 132, in __call__ return launch_agent(self._config, self._entrypoint, list(args)) File "/fsx/ferdinandmom/miniforge3/envs/env-bench-cluster/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 263, in launch_agent raise ChildFailedError( torch.distributed.elastic.multiprocessing.errors.ChildFailedError: ============================================================ /fsx/ferdinandmom/ferdinand-hf/bench_cluster/nanotron/run_train.py FAILED ------------------------------------------------------------ Failures: [1]: time : 2024-07-02_21:07:06 host : ip-26-0-170-160.ec2.internal rank : 15 (local_rank: 7) exitcode : -6 (pid: 852505) error_file: traceback : Signal 6 (SIGABRT) received by PID 852505 ------------------------------------------------------------ Root Cause (first observed failure): [0]: time : 2024-07-02_21:07:06 host : ip-26-0-170-160.ec2.internal rank : 14 (local_rank: 6) exitcode : -6 (pid: 852504) error_file: traceback : Signal 6 (SIGABRT) received by PID 852504 ============================================================ srun: error: ip-26-0-165-24: task 0: Exited with exit code 1 srun: error: ip-26-0-170-160: task 1: Exited with exit code 1 Consider using `hf_transfer` for faster uploads. This solution comes with some limitations. See https://huggingface.co/docs/huggingface_hub/hf_transfer for more details.