DeepSpeed ZeRO-3 NVMe Offload Checkpoint Race (FileExistsError)

All ranks race to write into the same consolidation directory during ZeRO-3 NVMe-offload checkpoint save, so only rank 0 succeeds and the rest hit FileExistsError.

Quick answer

All ranks race to write into the same consolidation directory during ZeRO-3 NVMe-offload checkpoint save, so only rank 0 succeeds and the rest hit FileExistsError.

Reliability#deepspeed#zero-3#nvme-offload#checkpoint#race-condition

What this failure is

DeepSpeed ZeRO-3 NVMe Offload Checkpoint Race (FileExistsError) is a Reliability failure seen during ML training runs. All ranks race to write into the same consolidation directory during ZeRO-3 NVMe-offload checkpoint save, so only rank 0 succeeds and the rest hit FileExistsError. Common tags: Deepspeed, Zero 3, Nvme Offload, Checkpoint.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about DeepSpeed ZeRO-3 NVMe Offload Checkpoint Race (FileExistsError). Paste your own crash log or traceback below and get the real root cause for YOUR run, not this generic entry. No account, no card. Logs are masked at ingress and never saved to account history.

training_logs.txt

No log to hand? Try one:

3 free diagnoses/day

Want 14 days on the Scale plan?

Request an evaluation code. A verified workplace organization activates up to 50 diagnoses a day, alerts, history, and follow-up questions. No credit card or automatic subscription.

Evaluate one incident

Why it happens (the mechanism)

All ranks create/copy into the same offloaded_tensors/ directory. Copytree on an existing dir raises FileExistsError. No per-rank isolation of the consolidation path. Taken together, these mechanisms explain why the failure is reproducible, why it tends to surface on specific workloads or scales, and why generic mitigation attempts often fall short without addressing the underlying cause.

What you'll observe

save_checkpoint crashes with ZeRO-3 + NVMe offload
Only happens multi-GPU
Requires gather-16bit-weights on save

Common symptoms and what they mean

Symptom	Why it happens
FileExistsError: [Errno 17] File exists	All ranks create/copy into the same offloaded_tensors/ directory
Error at shutil.copytree() in the consolidation step	copytree on an existing dir raises FileExistsError
ZeRO-3 + NVMe offload only	No per-rank isolation of the consolidation path

Which systems are affected

DeepSpeed ZeRO-3 with NVMe offload
stage3_gather_16bit_weights_on_model_save: true
Multi-GPU saves

How to confirm this is the problem

Use this checklist to test the hypothesis against a small reproduction. No single line proves the root cause, so preserve the preceding events and compare one variable at a time.

✓Reproduce the failure from a clean checkpoint/seed: the symptom must appear without warm-up state from a previous run.
✓Verified signal present: FileExistsError: [Errno 17] File exists
✓Verified signal present: Error at shutil.copytree() in the consolidation step
✓Verified signal present: ZeRO-3 + NVMe offload only
✓A targeted fix from the "How to fix it" section eliminates or substantially reduces the symptom within one validation pass.

The fix and the prevention pattern

The root cause is on this page and stays free. A free account adds the exact remediation steps, saved history, and the fix on every entry in the encyclopedia.

No credit card. Daily allowance follows verified trust tier. Instant access.

Diagnose this failure in VS Code

Select the traceback or open the failed terminal, then run Denpex locally to see the initiating rank, collateral failures, exact fix, and verification command without uploading the log.

Install the free VS Code extension

DeepSpeed errors in context

DeepSpeed changes when parameters, gradients and optimizer state are created, partitioned, gathered and offloaded. The hub separates ZeRO, memory, checkpoint and pipeline failures by lifecycle phase.

Compare every deepspeed error side by side

Related failures to investigate next

Bytedance Sequence Length Straggler often co-occurs with this failure because both share underlying assumptions about reliability; if you fix this entry but still see drift, validate the related primary cause next. bytedance sequence length straggler →

Training Checkpoint Corruption On Write often co-occurs with this failure because both share underlying assumptions about reliability; if you fix this entry but still see drift, validate the related neighboring cause next. training checkpoint corruption on write →

Wandb Tensorboard Failure often co-occurs with this failure because both share underlying assumptions about reliability; if you fix this entry but still see drift, validate the related neighboring cause next. wandb tensorboard failure →

Root cause

All ranks create/copy into the same offloaded_tensors/ directory
copytree on an existing dir raises FileExistsError
No per-rank isolation of the consolidation path

The fix and how to prevent it

Evaluate Denpex on your own logs

Request a Scale evaluation code. A verified workplace organization activates 14 days with up to 50 diagnoses a day. Every account keeps its current diagnosis allowance and gets a verification path. No card or automatic subscription.

References

Don't just read the fix, diagnose your run

The encyclopedia tells you what went wrong. Denpex tells you what went wrong in YOUR training run. With your logs, your config, and your stack.

Diagnose your logs free Browse all failures

Related Reliability errors

Sequence Length Imbalance Causing Distributed Training Stragglers

Reliability · high

Silent Data Corruption from GPU Hardware Faults Causing Loss Spikes and Model Divergence

Reliability · critical

NIXL Firmware Page Registration Fan-Out Triggers Host OOM Kills on HGX H200 and B200

Reliability · critical

MTTF Scaling Inversely with GPU Count in Large ML Research Clusters

Reliability · high