[Bug] Can't run distributed training on multiple nodes with slurm using the oumi distributed wrapper.

### What happened? When trying to run a training using oumi distrbuted torchrun wrapper with slurm, I wait an endless time with nothing happening (probably a rendez-vous problem). I fixed the rendez-vous endpoint but still not working. ### Steps to reproduce the bug Steps to reproduce : 1. salloc 2 or more nodes using slurm. 2. srun one task per node using the torchrun wrapper (oumi distributed). 3. Wait ### System Info ```shell ____ _ _ __ __ _____ / __ \| | | | \/ |_ _| | | | | | | | \ / | | | | | | | | | | |\/| | | | | |__| | |__| | | | |_| |_ \____/ \____/|_| |_|_____| ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Oumi environment information: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ┌────────────────┬──────────────────────────────────────────────┐ │ Oumi version │ 0.4.2 │ │ Python version │ 3.10.15 │ │ Platform │ Linux-6.8.0-84-generic-x86_64-with-glibc2.39 │ └────────────────┴──────────────────────────────────────────────┘ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Installed dependencies: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ┏━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓ ┃ PACKAGE ┃ VERSION ┃ ┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩ │ accelerate │ 1.11.0 │ │ aiohttp │ 3.13.2 │ │ bitsandbytes │ <not installed> │ │ datasets │ 3.2.0 │ │ diffusers │ <not installed> │ │ einops │ <not installed> │ │ jsonlines │ 4.0.0 │ │ liger-kernel │ 0.6.3 │ │ llama-cpp-python │ <not installed> │ │ lm-eval │ 0.4.9.1 │ │ mlflow │ 3.6.0 │ │ numpy │ 2.2.6 │ │ nvidia-ml-py │ <not installed> │ │ omegaconf │ 2.4.0.dev3 │ │ open_clip_torch │ <not installed> │ │ pandas │ 2.3.3 │ │ peft │ 0.17.1 │ │ pexpect │ 4.9.0 │ │ pillow │ 11.3.0 │ │ pydantic │ 2.11.10 │ │ responses │ 0.25.8 │ │ sglang │ <not installed> │ │ skypilot │ 0.10.5 │ │ tensorboard │ 2.20.0 │ │ timm │ <not installed> │ │ torch │ 2.8.0 │ │ torchdata │ 0.9.0 │ │ torchvision │ 0.23.0 │ │ tqdm │ 4.67.1 │ │ transformers │ 4.57.1 │ │ trl │ 0.21.0 │ │ typer │ 0.20.0 │ │ vllm │ <not installed> │ │ wandb │ 0.21.4 │ └──────────────────┴─────────────────┘ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Environment variables: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓ ┃ VARIABLE ┃ VALUE ┃ ┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩ │ ACCELERATE_DYNAMO_BACKEND │ <not set> │ │ ACCELERATE_DYNAMO_MODE │ <not set> │ │ ACCELERATE_DYNAMO_USE_DYNAMIC │ <not set> │ │ ACCELERATE_DYNAMO_USE_FULLGRAPH │ <not set> │ │ ACCELERATE_USE_FSDP │ <not set> │ │ CUDA_VISIBLE_DEVICES │ 0,1,2,3,4,5,6,7 │ │ LOCAL_RANK │ <not set> │ │ LOCAL_WORLD_SIZE │ <not set> │ │ OUMI_EXTRA_DEPS_FILE │ <not set> │ │ OUMI_FORCE_EDITABLE_INSTALL │ <not set> │ │ OUMI_SLURM_CONNECTIONS │ <not set> │ │ OUMI_USE_SPOT_VM │ <not set> │ │ RANK │ <not set> │ │ WORLD_SIZE │ <not set> │ └─────────────────────────────────┴─────────────────┘ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ PyTorch information: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ ┌────────────────┬───────────────────────┐ │ CUDA available │ True │ │ CUDA version │ 12.8 │ │ cuDNN version │ 91.0.2 │ │ Number of GPUs │ 8 │ │ GPU type │ NVIDIA H100 80GB HBM3 │ │ GPU memory │ 79.2GB │ └────────────────┴───────────────────────┘ ```