Partial Checkpoint Save
Partial checkpoint saves write incomplete data when training is interrupted, leaving a file that appears valid but is truncated.
Partial checkpoint saves write incomplete data when training is interrupted, leaving a file that appears valid but is truncated.
What this failure is
Partial Checkpoint Save is a Data Integrity failure seen during ML training runs. Partial checkpoint saves write incomplete data when training is interrupted, leaving a file that appears valid but is truncated. Common tags: Checkpoint, Partial, Save, Atomic.
Is this what broke your run? Paste your log.
You're reading about Partial Checkpoint Save. 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.
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.
Why it happens (the mechanism)
Job was preempted or killed during checkpoint write. Disk full during checkpoint write. NFS timeout during checkpoint write. Multiple processes writing to the same checkpoint file. 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
- Checkpoint file exists but is smaller than expected
- Resume from checkpoint crashes with pickle error
- File size differs between saves of the same model
Common symptoms and what they mean
| Symptom | Why it happens |
|---|---|
| Checkpoint file is smaller than expected (e.g., 500MB instead of 2GB) | Job was preempted or killed during checkpoint write |
| UnpicklingError when loading checkpoint | Disk full during checkpoint write |
| torch.load fails halfway through | NFS timeout during checkpoint write |
Which systems are affected
- Training with preemptible instances
- Long-running training with periodic checkpoints
- Network filesystem (NFS) checkpoint targets
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: Checkpoint file is smaller than expected (e.g., 500MB instead of 2GB)
- ✓Verified signal present: UnpicklingError when loading checkpoint
- ✓Verified signal present: torch.load fails halfway through
- ✓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.
Sign up free. Unlock the full analysisNo 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 extensionRelated failures to investigate next
Root cause
- Job was preempted or killed during checkpoint write
- Disk full during checkpoint write
- NFS timeout during checkpoint write
- Multiple processes writing to the same checkpoint file
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.
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.