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 blackwell moved the nvlink fault family to xid 149 with netir subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events. Validating against the symptom list in this entry is the fastest way to confirm the diagnosis.

Is this a hardware or a software issue?

Hardware 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 hardware 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 NVIDIA B100 and B200 GPUs 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 #blackwell, #nvlink, #xid-149. 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 blackwell is upgraded. Pin compatible versions and rerun the prevention smoke test on every dependency change.

Is there a fingerprint I can monitor?

Yes, Kernel log line of the form: NVRM: Xid (PCI:0018:01:00): 149, NETIR_LINK_DOWN Fatal 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.

Xid 149 NETIR_LINK_DOWN on Blackwell: the NVLink fault that Xid 74 knowledge misses

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 blackwell moved the nvlink fault family to xid 149 with netir subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events 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.

On B100, B200 and GB200 class systems NVIDIA reports NVLink faults as Xid 149 with a NETIR subcode instead of the Xid 74 used on earlier generations. An operator or a runbook that only greps for Xid 74 sees nothing while the link is down and the GPU has already been pulled out of the NVLink domain.

Quick answer

On B100, B200 and GB200 class systems NVIDIA reports NVLink faults as Xid 149 with a NETIR subcode instead of the Xid 74 used on earlier generations.

Symptom: NVRM: Xid (PCI:0018:01:00): 149, NETIR_LINK_EVT Fatal XC0 i0 Link 08 (0x004505c6 0x00000000 0x00000000 0x00000000)
Root cause: Blackwell moved the NVLink fault family to Xid 149 with NETIR subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events. A fatal NETIR_LINK_DOWN takes the GPU out of the NVLink domain and marks it Degraded. The degraded marking is sticky, so the GPU stays out of the domain until it is reset even though nvidia-smi may look normal.
Recommended fix: Capture the complete Xid 149 line. The subcode name (NETIR_LINK_DOWN versus NETIR_LINK_EVT) and the hex payload are what separate a hard link-down from a bit-error-rate event, so a truncated log is not actionable.
How Denpex helps: Denpex matches Xid 149 NETIR_LINK_DOWN on Blackwell: the NVLink fault that Xid 74 knowledge misses 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.

Hardware#blackwell#nvlink#xid-149#netir#gb200#b200

What this failure is

Xid 149 NETIR_LINK_DOWN on Blackwell: the NVLink fault that Xid 74 knowledge misses is a Hardware failure seen during ML training runs. On B100, B200 and GB200 class systems NVIDIA reports NVLink faults as Xid 149 with a NETIR subcode instead of the Xid 74 used on earlier generations. An operator or a runbook that only greps for Xid 74 sees nothing while the link is down and the GPU has already been pulled out of the NVLink domain. Common tags: Blackwell, Nvlink, Xid 149, Netir.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about Xid 149 NETIR_LINK_DOWN on Blackwell: the NVLink fault that Xid 74 knowledge misses. 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:

Zero-Trust Client Redaction Active. No sensitive values detected.

3 free diagnoses/day

Want 30 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)

Blackwell moved the NVLink fault family to Xid 149 with NETIR subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events. A fatal NETIR_LINK_DOWN takes the GPU out of the NVLink domain and marks it Degraded. The degraded marking is sticky, so the GPU stays out of the domain until it is reset even though nvidia-smi may look normal. 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

Collectives fail or hang on a Blackwell node while every Xid 74 check stays silent
nvidia-smi looks healthy after a reboot but the same rank keeps failing collectives
Existing NVLink runbooks written for Hopper do not fire on Blackwell hardware

Common symptoms and what they mean

Symptom	Why it happens
Kernel log line of the form: NVRM: Xid (PCI:0018:01:00): 149, NETIR_LINK_DOWN Fatal	Blackwell moved the NVLink fault family to Xid 149 with NETIR subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events
Xid 154 GPU recovery action changed logged immediately after the 149	A fatal NETIR_LINK_DOWN takes the GPU out of the NVLink domain and marks it Degraded
knvlinkSetDegradedMode_IMPL reporting that the GPU is marked Degraded	The degraded marking is sticky, so the GPU stays out of the domain until it is reset even though nvidia-smi may look normal
NCCL falling back to a slower path or failing ring setup on one rank only	Blackwell moved the NVLink fault family to Xid 149 with NETIR subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events

Which systems are affected

NVIDIA B100 and B200 GPUs
GB200 NVL72 and GB300 racks
Any Blackwell NVLink domain, including DGX Spark GB10

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.

✓grep the kernel log for "Xid" and for "149" separately; a runbook that greps only for 74 will return nothing on Blackwell
✓Look for the Xid 154 recovery-action line and the Degraded marking following the 149
✓Confirm the affected link id repeats across resets before escalating to hardware replacement

Searchable error signature

search key

NVRM: Xid (PCI:0018:01:00): 149, NETIR_LINK_EVT Fatal XC0 i0 Link 08 (0x004505c6 0x00000000 0x00000000 0x00000000)
NVRM: Xid (PCI:0018:01:00): 149, NETIR_LINK_DOWN Fatal
NVRM: Xid (PCI:0018:01:00): 154, GPU recovery action changed
NVRM: knvlinkSetDegradedMode_IMPL: GPU 0000:18:01.0 marked Degraded
nvidia-smi nvlink -s: GPU 0: Link 8: <inactive>

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

Root cause

Blackwell moved the NVLink fault family to Xid 149 with NETIR subcodes; the subcode name distinguishes a hard link-down from bit-error-rate and data-error events
A fatal NETIR_LINK_DOWN takes the GPU out of the NVLink domain and marks it Degraded
The degraded marking is sticky, so the GPU stays out of the domain until it is reset even though nvidia-smi may look normal

The fix and how to prevent it

Evaluate Denpex on your own logs

Request a Scale evaluation code. A verified workplace organization activates 30 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 Hardware errors

NVSwitch SXid errors: what aborts your CUDA jobs and what is only informational

Hardware · critical

GPU Memory Clock Throttle

Hardware · medium

NVLink Error

Hardware · high

GPU Thermal Throttling

Hardware · high