NCCL RAS Query Crashes Job (Memory Corruption, 2.27.3)
Querying the NCCL RAS subsystem during training on NCCL 2.27.3 returns corrupted communicator data and can crash the job via TCPStore connection failures.
Querying the NCCL RAS subsystem during training on NCCL 2.
What this failure is
NCCL RAS Query Crashes Job (Memory Corruption, 2.27.3) is a Communication failure seen during ML training runs. Querying the NCCL RAS subsystem during training on NCCL 2.27.3 returns corrupted communicator data and can crash the job via TCPStore connection failures. Common tags: Nccl, Ras, Memory Corruption, Tcpstore.
Is this what broke your run? Paste your log.
You're reading about NCCL RAS Query Crashes Job (Memory Corruption, 2.27.3). 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)
Memory corruption within the NCCL RAS subsystem on 2.27.3. Corrupted state propagates to TCPStore, dropping connections. Triggered by the RAS status query path itself. 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
- Polling NCCL RAS status crashes a running job
- RAS reports impossible rank numbers / missing processes
- High reproduction probability at large scale
Common symptoms and what they mean
| Symptom | Why it happens |
|---|---|
| recvValueWithTimeout failed on SocketImpl ... Connection was likely closed (TCPStore) | Memory corruption within the NCCL RAS subsystem on 2.27.3 |
| RAS VERBOSE STATUS returns corrupted communicator data | Corrupted state propagates to TCPStore, dropping connections |
| Job terminates shortly after a RAS query | Triggered by the RAS status query path itself |
Which systems are affected
- NCCL 2.27.3 (RAS subsystem)
- Large jobs (100+ GPUs) where RAS is polled for health
- PyTorch 2.8.0 + NCCL 2.27.3
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: recvValueWithTimeout failed on SocketImpl ... Connection was likely closed (TCPStore)
- ✓Verified signal present: RAS VERBOSE STATUS returns corrupted communicator data
- ✓Verified signal present: Job terminates shortly after a RAS query
- ✓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 extensionNCCL 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 sideRelated failures to investigate next
Root cause
- Memory corruption within the NCCL RAS subsystem on 2.27.3
- Corrupted state propagates to TCPStore, dropping connections
- Triggered by the RAS status query path itself
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.