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 non-fatal sxids are informational; fabric manager keeps running jobs alive, though a brief performance drop or stall can occur. 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 NVSwitch based systems including DGX and HGX baseboards 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 #nvswitch, #sxid, #fabric-manager. 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 nvswitch 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: nvidia-nvswitch0: SXid (PCI:0000:00:00.0): 20034, Fatal, Link 30 LTSSM Fault Up 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.

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

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 non-fatal sxids are informational; fabric manager keeps running jobs alive, though a brief performance drop or stall can occur 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.

NVSwitch reports its own error family as SXid, not Xid. Whether an SXid aborts running jobs depends on severity and on which port reported it, and that distinction decides whether you reset one GPU or the whole baseboard pair.

Quick answer

NVSwitch reports its own error family as SXid, not Xid.

Symptom: nvidia-nvswitch3: SXid (PCI:0000:05:00.0): 12028, Non-fatal, Link 32 egress non-posted PRIV error (First)
Root cause: Non-fatal SXids are informational; Fabric Manager keeps running jobs alive, though a brief performance drop or stall can occur. A fatal SXid on a GPU-to-NVSwitch port is propagated to that GPU, aborting its CUDA jobs, and commonly surfaces as Xid 74 plus Xid 45. A fatal SXid on a port connecting two GPU baseboards makes Fabric Manager abort every running CUDA job and block new launches, with Xid 45 reported as part of the abort.
Recommended fix: Read the severity field in the SXid line first. Non-fatal entries are informational and do not by themselves justify draining the node.
How Denpex helps: Denpex matches NVSwitch SXid errors: what aborts your CUDA jobs and what is only informational 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#nvswitch#sxid#fabric-manager#nvlink#ltssm

What this failure is

NVSwitch SXid errors: what aborts your CUDA jobs and what is only informational is a Hardware failure seen during ML training runs. NVSwitch reports its own error family as SXid, not Xid. Whether an SXid aborts running jobs depends on severity and on which port reported it, and that distinction decides whether you reset one GPU or the whole baseboard pair. Common tags: Nvswitch, Sxid, Fabric Manager, Nvlink.

Live diagnosis, no signup

Is this what broke your run? Paste your log.

You're reading about NVSwitch SXid errors: what aborts your CUDA jobs and what is only informational. 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)

Non-fatal SXids are informational; Fabric Manager keeps running jobs alive, though a brief performance drop or stall can occur. A fatal SXid on a GPU-to-NVSwitch port is propagated to that GPU, aborting its CUDA jobs, and commonly surfaces as Xid 74 plus Xid 45. A fatal SXid on a port connecting two GPU baseboards makes Fabric Manager abort every running CUDA job and block new launches, with Xid 45 reported as part of the abort. 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

All CUDA jobs on a node abort at once and new launches are refused
Operators grep for Xid and find nothing because the switch reports SXid
Unclear whether an observed SXid requires action or is informational noise

Common symptoms and what they mean

Symptom	Why it happens
Kernel log line: nvidia-nvswitch0: SXid (PCI:0000:00:00.0): 20034, Fatal, Link 30 LTSSM Fault Up	Non-fatal SXids are informational; Fabric Manager keeps running jobs alive, though a brief performance drop or stall can occur
Kernel log line: nvidia-nvswitch3: SXid (PCI:0000:05:00.0): 12028, Non-fatal, Link 32 egress non-posted PRIV error (First)	A fatal SXid on a GPU-to-NVSwitch port is propagated to that GPU, aborting its CUDA jobs, and commonly surfaces as Xid 74 plus Xid 45
Correlated Xid 74 and Xid 45 on GPUs when the fatal SXid is on a GPU-to-NVSwitch port	A fatal SXid on a port connecting two GPU baseboards makes Fabric Manager abort every running CUDA job and block new launches, with Xid 45 reported as part of the abort
Fabric Manager refusing new CUDA job launches	Non-fatal SXids are informational; Fabric Manager keeps running jobs alive, though a brief performance drop or stall can occur

Which systems are affected

NVSwitch based systems including DGX and HGX baseboards
GB200 NVL72 NVLink domains
Any multi-GPU node running the Fabric Manager service

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 dmesg and syslog for "SXid" and read the severity token that follows the code number
✓Check whether the reporting device is a GPU-facing port or a baseboard trunk port
✓Check the Fabric Manager service log for the matching abort record

Searchable error signature

search key

nvidia-nvswitch3: SXid (PCI:0000:05:00.0): 12028, Non-fatal, Link 32 egress non-posted PRIV error (First)
nvidia-nvswitch0: SXid (PCI:0000:00:00.0): 20034, Fatal, Link 30 LTSSM Fault Up
nv-fabricmanager: detected NVSwitch fatal error, aborting all running CUDA jobs
NVRM: Xid (PCI:0000:8a:00): 74, NVLink error detected
NVRM: Xid (PCI:0000:8a:00): 45, Ch 00000010

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

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

Root cause

Non-fatal SXids are informational; Fabric Manager keeps running jobs alive, though a brief performance drop or stall can occur
A fatal SXid on a GPU-to-NVSwitch port is propagated to that GPU, aborting its CUDA jobs, and commonly surfaces as Xid 74 plus Xid 45
A fatal SXid on a port connecting two GPU baseboards makes Fabric Manager abort every running CUDA job and block new launches, with Xid 45 reported as part of the abort

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

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

Hardware · critical

GPU Memory Clock Throttle

Hardware · medium

NVLink Error

Hardware · high

GPU Thermal Throttling

Hardware · high