Is this a hardware or a software issue?

Data failures can look like either layer depending on the symptom. The discriminator is the resolution of the symptom to the recommended fix: if applying the steps under "How to fix it" eliminates the symptom, this is a software/configuration class issue regardless of the hardware it surfaced on. Repeat the fix on a fresh node to confirm.

How long does the fix typically take?

Most resolutions from this entry take 10 to 30 minutes of engineer time once the diagnosis is confirmed. Long fixes come from skipping the diagnosis step and chasing apparent symptoms (validation drift, memory growth, slow throughput) instead of the actual root cause listed above. Apply the fix in a small experiment first, then roll it forward.

Does this affect inference or only training?

Primarily this is a data failure observed during training, but the same misconfiguration can leak into downstream evaluation and serving. Treat the prevention checklist as a guardrail: any pipeline that touches a trained checkpoint should re-validate the failure mode before promoting the model.

What's the cheapest way to confirm this is the problem?

Cheapest signal: reproduce the symptom on a 1-GPU toy run with the same configuration. If the toy run exhibits the same behavior, the cause is the configuration rather than the cluster. If only the multi-GPU run is affected, the failure is a distributed interaction, so PyTorch is the most likely layer to inspect.

Can the Denpex platform detect this automatically?

Yes. Denpex correlates the symptoms in this entry against your training logs in real time and surfaces the failure taxonomy, root cause, and exact remediation command on the console. Three full diagnoses are included free during signup, and the encyclopedia entry you are reading is generated from the same signal library that powers the platform.

What related failures should I check next?

Closely related failures cluster around #Rank Desync / Straggler. After applying this fix, validate against the related entries below to confirm no overlapping cause is masking as this one. Engineers routinely discover that the dominant failure mode is actually a second, sibling failure compounded on top.

Will this recur after the fix?

Applied once with the prevention checklist, this failure does not recur under normal operation. The recurring variant most engineers encounter is the re-emergence of the same symptom on a different model or stack after Rank Desync / Straggler is upgraded. Pin compatible versions and rerun the prevention smoke test on every dependency change.

Is there a fingerprint I can monitor?

Yes. Training hangs indefinitely after a specific number of epochs or steps. is the highest-signal fingerprint. Wire a metric or alert on that signal (or its absence after the fix) so you can detect the recurrence automatically. The Denpex platform encodes this exact fingerprint and surfaces it directly in the dashboard.

What is the official documentation that covers this?

Reference documentation is listed in the References section at the bottom of this entry. The most authoritative URLs are the upstream framework docs and the issue trackers cited; the Denpex entry synthesizes both into the executive, engineer, and on-call perspectives so each audience can act without reading every linked source.

Does this entry include working code?

Yes. The "Code Examples" section contains copy-paste-ready snippets that target the most common stacks (PyTorch, Hugging Face Trainer, DeepSpeed, Megatron-LM, vLLM/Transformers). Each snippet is annotated with why it works so you can adapt the pattern to your own framework without losing fidelity.

NCCL Watchdog Timeout due to Uneven Dataset Sharding

Q: What does this failure look like in production?

In production this failure typically appears during the first tens-to-hundreds of training steps: loss curves become non-monotonic, validation metrics plateau above the baseline, and downstream checkpoints either diverge or fail to improve. Engineers frequently misread it as data drift when the underlying mechanism is when sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=false, some ranks will have fewer batches. these ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out.. Validating against the symptom list in this entry is the fastest way to confirm the diagnosis.

Q: How does this scale from 1 GPU to 1000 GPUs?

At single-GPU scale the symptom usually looks like slow convergence or a single anomalous loss value. At 64+ GPU scale, distributed interactions amplify when sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=false, some ranks will have fewer batches. these ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out. through collectives, and the failure can manifest as a full job abort. Always reproduce the fix at the smallest cluster configuration that still exhibits the symptom before promoting to production scale.

When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches. These ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out.

Quick answer

When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches.

Symptom: NCCL WARN local rank * failed to connect to ring
Root cause: When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches. These ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g.
Recommended fix: Ensure all ranks process the exact same number of batches. Use drop_last=True in DistributedSampler or PyTorch's Join context manager. This guarantees all ranks participate in all expected collective operations.
How Denpex helps: Denpex matches NCCL Watchdog Timeout due to Uneven Dataset Sharding across every rank in a distributed run and reports which rank failed first, so you act on the initiating node instead of the loudest one.

Data#Rank Desync / Straggler

What this failure is

NCCL Watchdog Timeout due to Uneven Dataset Sharding is a Data failure seen during ML training runs. When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches. These ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out. Common tags: Rank Desync / Straggler.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about NCCL Watchdog Timeout due to Uneven Dataset Sharding. 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)

The error message 'Watchdog caught collective operation timeout' suggests a network or NCCL failure, making engineers investigate networking hardware or drivers, when the actual cause is a simple data loading desync.

What you'll observe

RuntimeError: Watchdog caught collective operation timeout
NCCL WARN local rank * failed to connect to ring

Common symptoms and what they mean

Symptom	Why it happens
Training hangs indefinitely after a specific number of epochs or steps.	When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches. These ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out.
One rank finishes early and exits, while others are stuck at an all_reduce barrier.	When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches. These ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out.

Which systems are affected

PyTorch
NCCL
DistributedDataParallel

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.

✓Check the number of batches yielded by the dataloader on each rank by logging len(dataloader).
✓Examine if one rank logs a 'finished training' message before the crash.

Searchable error signature

search key

NCCL WARN local rank * failed to connect to ring
RuntimeError: Watchdog caught collective operation timeout

Use this text as a lookup key in logs and upstream issue trackers. It is not presented as a captured customer log. Confirm the cause from your own preceding events, versions, configuration and the cited references.

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

NCCL errors in context

NCCL is where a distributed job reports failure, which is not the same as where it failed. The hub lists every common NCCL error next to what it actually indicates, and the environment variables that tell them apart.

Compare every nccl error side by side

Root cause

When sharding datasets across multiple workers, if the total number of samples is not perfectly divisible by the world size and drop_last=False, some ranks will have fewer batches. These ranks will finish their loop early, while the remaining ranks enter a collective operation (e.g., gradient all_reduce) expecting participation from all ranks, leading to a hang until the watchdog times out.

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

↗Uneven dataset leads to Watchdog timeout

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 Data errors

HDF5 Concurrent Read Deadlock with Multiprocessing

Data · high

WebDataset/TFRecord Decoding Crash Mid-Epoch

Data · medium