Lightning-AI / Lightning-AI/pytorch-lightning

Iterable dataset + DDP + SLURM + MultiGPU : Training stuck - error: The client socket has failed to connect to [ip6-localhost]:24355 (errno: 99 - Cannot assign requested address).

Open
#18,338 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug repro needed ver: 2.0.x
Dominant language
Python
Stars
31.4k
Forks
3.8k
Avg merge
6d 7h
Merged PRs (30d)
6

Description

### Bug description

I am trying to train a network on iterable style dataset. It works when I train on interactive node but when I submit job the training is stuck. I am posting this here after doing a lot of search on internet and on lightning git issues.

I get the error:
`
[W socket.cpp:601] [c10d] The client socket has failed to connect to [ip6-localhost]:24355 (errno: 99 - Cannot assign requested address).`

### What version are you seeing the problem on?

v2.0

### How to reproduce the bug

_No response_

### Error messages and logs

```
# Error messages and logs here please
```

### Environment

Current environment

```
#- Lightning Component (e.g. Trainer, LightningModule, LightningApp, LightningWork, LightningFlow):
#- PyTorch Lightning Version (e.g., 1.5.0):
#- Lightning App Version (e.g., 0.5.2):
#- PyTorch Version (e.g., 2.0):
#- Python version (e.g., 3.9):
#- OS (e.g., Linux):
#- CUDA/cuDNN version:
#- GPU models and configuration:
#- How you installed Lightning(`conda`, `pip`, source):
#- Running environment of LightningApp (e.g. local, cloud):
```

### More info

_No response_

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the v2.0 training setup with an iterable dataset, DDP, MultiGPU, and a submitted SLURM job rather than an interactive node. Inspect the client socket error and compare the two launch environments; done means the submitted training job no longer stalls.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
distributed-systems, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.