For ml researchers and fine-tuning developers

Isolate motherboard PCIe P2P restrictions and /dev/shm limits on Vast.ai hosts.

Verify IOMMU peer-to-peer memory access, Docker shared memory allocation, and spot instance preemption markers before blaming model code on heterogeneous Vast.ai rigs.

Replay one resolved incident Try 30 days free

Free 30-day Team evaluation. No card or automatic subscription. Work email verification required.

Live diagnosis, no signup

Replay a resolved Vast.ai multi-GPU crash incident

Upload your log file. Review the selected evidence and masking, then run diagnosis. Keep the original log for any follow-up questions.

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

Try Denpex for 30 days. FREE.

No credit card. No automatic subscription. See whether Denpex helps with an incident you already understand.

Start your free 30-day Team evaluation

Up to 50 diagnoses a day for 30 days. Verify your work email to activate an eligible company evaluation, then replay a resolved incident and compare the evidence and next steps with your team's confirmed cause.

What a useful evaluation should establish

Establish whether the failure stems from host motherboard topology limitations or a container configuration gap that can be resolved with environment flags.

Use Denpex alongside your existing telemetry tools to investigate submitted incident evidence. With monitoring configured on a supported plan, detected failures can trigger alerts and an automatic diagnosis from the captured evidence.

Inside Denpex

See the evidence, not just the claim

Follow the causal chain, inspect the quoted evidence and prepare the next handoff. These actual product captures use founder-account examples and the VS Code extension, with the source identified under each image. Open a capture to inspect it at full size.

Denpex founder-account example: a causal graph connects an Xid 48 ECC event to a rank exit and downstream collective failures, with quoted evidence for each stage. — Live founder-account capture from the bundled NCCL/ECC example. Follow the hardware event, rank exit and downstream symptoms with a supporting log quote at each stage.Inspect original image

Denpex hardware handoff summary asks the operator to capture diagnostic evidence and verify RMA eligibility with the vendor. — Actual hardware handoff summary from the same founder-account example. Capture diagnostic evidence before an approved reset, then confirm RMA eligibility with the vendor.Inspect original image

Denpex historical example says a collective timeout does not establish the initiating rank, cause or failure domain and requests distinguishing evidence before a remedy. — Founder-account example restored from history. The original timeout remains unattributed, and the UI keeps the missing-evidence assessment visible.Inspect original image

Denpex VS Code extension displaying a bundled OOM-before-NCCL example, matched log evidence, verification guidance and an investigation handoff. — Actual VS Code extension capture of the bundled OOM-before-NCCL example. Review matched log evidence, verification guidance and the investigation handoff without leaving the editor.Inspect original image

Failure Signature Triage

Vast.ai - Observed Runtime Evidence

Read Vast.ai documentation

Incident pattern summary

A 4x RTX 4090 instance on Vast.ai crashes immediately on torch.distributed.init_process_group with ncclUnhandledSystemError. The log trace points to failed P2P CUDA IPC mapping across consumer motherboard PCIe switches, resolved by testing NCCL_P2P_DISABLE=1 or migrating hosts.

Terminal traceback excerpt (Vast.ai)

vast-instance-8912:88:104 [1] NCCL WARN Cuda failure 'peer access is not supported between these two devices'
vast-instance-8912:88:104 [1] init.cc:902 NCCL WARN initTransportsRank: P2P initialization failed
torch.distributed.DistBackendError: NCCL error in: ../torch/csrc/distributed/c10d/ProcessGroupNCCL.cpp:1275, unhandled system error, NCCL version 2.19.3

Recommended verification flags

Set these environment flags in your training container or launch script to capture high-resolution collective logs:

NCCL_P2P_DISABLE=1NCCL_SHM_DISABLE=0NCCL_DEBUG=INFOCUDA_LAUNCH_BLOCKING=1

Diagnostic Scope Boundary

Denpex cannot modify host BIOS IOMMU settings or prevent marketplace host reclamation on Vast.ai. The diagnosis clarifies whether a crash requires environment workarounds or immediate instance replacement.