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 previous training process not fully killed. Validating against the symptom list in this entry is the fastest way to confirm the diagnosis.

Is this a hardware or a software issue?

Reliability 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 reliability 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 Training with elastic restart 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 #restart, #stuck, #cleanup. 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 restart is upgraded. Pin compatible versions and rerun the prevention smoke test on every dependency change.

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 previous training process not fully killed 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.

Is there a fingerprint I can monitor?

Yes. New training process hangs at init 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.

Training Restart Stuck

Training restarts after failure can get stuck if cleanup didn't complete properly, blocking new training jobs.

Quick answer

Training restarts after failure can get stuck if cleanup didn't complete properly, blocking new training jobs.

Root cause: Previous training process not fully killed. CUDA context not released. Distributed training group not destroyed.
Recommended fix: Kill all related processes: pkill -9 -f python
How Denpex helps: Denpex investigates Training Restart Stuck using the evidence you provide or your connected workload collects. Earlier rank, host or application evidence is needed to distinguish an initiating failure from a downstream report.

Reliability#restart#stuck#cleanup#reliability#zombie#cuda-context

What this failure is

Training Restart Stuck is a Reliability failure seen during ML training runs. Training restarts after failure can get stuck if cleanup didn't complete properly, blocking new training jobs. Common tags: Restart, Stuck, Cleanup, Reliability.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about Training Restart Stuck. Paste your own traceback and relevant evidence for an investigation of your workload, with a next action or a specific missing fact. A reference entry does not establish your cause. No account or card for the free diagnosis. Review data handling before submitting sensitive logs.

Before uploading, review cloud data handling and local options.

training_logs.txt

No log to hand? Try one:

Upload or drop a plain-text log

Upload your log file. Large files are processed in your browser. Only selected evidence is submitted when you run diagnosis.

3 free diagnoses/day

Why it happens (the mechanism)

Previous training process not fully killed. CUDA context not released. Distributed training group not destroyed. File handles not closed properly. 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

Restarted training hangs at startup
Previous training's resources still held
GPU memory not released from previous run

Common symptoms and what they mean

Symptom	Why it happens
New training process hangs at init	Previous training process not fully killed
GPU shows memory used by previous process	CUDA context not released
SLURM/K8s shows old process still running	Distributed training group not destroyed

Which systems are affected

Training with elastic restart
Training after manual restart
Training after node failure

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: New training process hangs at init
✓Verified signal present: GPU shows memory used by previous process
✓Verified signal present: SLURM/K8s shows old process still running
✓A targeted fix from the "How to fix it" section eliminates or substantially reduces the symptom within one validation pass.

Root cause

Previous training process not fully killed
CUDA context not released
Distributed training group not destroyed
File handles not closed properly

The fix and how to prevent it

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

CUDA errors in context

CUDA reports errors asynchronously, so the traceback usually points at whatever line synchronised next rather than the one at fault. The hub covers every common CUDA error and how to make it report honestly.

Compare every cuda 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 →

Nfs Mount 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. nfs mount failure →

Checkpoint Partial Save 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. checkpoint partial save →

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 Compare a solved incident Browse all failures

Related Reliability errors

GB200 NVL72 driver hang with Xid 150 and 154 after long uptime is a firmware bug, not a dead GPU

Reliability · high

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