For distributed training and cluster infrastructure engineers
Distinguish an InfiniBand mlx5 fabric flap from an application rank stall on Lambda.
Correlate Mellanox RDMA transport retry counters, NVSwitch health telemetry, and rank-level Python tracebacks to isolate which 8x H100 node triggered the collective abort.
Free 30-day Team evaluation. No card or automatic subscription. Work email verification required.
Replay a resolved Lambda 8x H100 cluster 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.
Upload your log file. Large files are processed in your browser. Only selected evidence is submitted when you run diagnosis.
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
Verify whether the triage report isolates the specific degraded HCA port and node instead of falsely flagging healthy victim ranks waiting at the collective barrier.
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.
Follow the failure through the cascade

Prepare the hardware escalation

Ask for the evidence that is missing

Keep the investigation in the editor

Lambda - Observed Runtime Evidence
Incident pattern summary
A 4-node 8x H100 training job on Lambda aborts mid-epoch with Watchdog collective timeout across 32 ranks. Inspecting raw logs reveals Rank 19 logged an IB Verbs transport retry counter exceeded error 45 seconds before the PyTorch watchdog fired.
lambda-node-03:4120:4298 [3] misc/ibvwrap.cc:284 NCCL WARN Call to ibv_poll_cq returned error 12 (Transport retry counter exceeded)
lambda-node-03:4120:4298 [3] NCCL INFO comm 0x7f8b94001090 rank 19 nranks 32 cudaDev 3 nvmlDev 3 - Abort COMPLETE
[E ProcessGroupNCCL.cpp:563] [Rank 19] Some NCCL operations have failed or timed out.Recommended verification flags
Set these environment flags in your training container or launch script to capture high-resolution collective logs:
NCCL_DEBUG=INFONCCL_DEBUG_SUBSYS=INIT,NETNCCL_IB_TIMEOUT=22NCCL_IB_RETRY_CNT=7Diagnostic Scope Boundary
Denpex cannot query physical Quantum-2 switch fabric counters or reseat optical transceivers on Lambda infrastructure. The evaluation isolates the suspect node and HCA port for faster Lambda support escalation.
Other GPU Cloud Failure Workflows
Compare evidence patterns across multi-node fabrics, Kubernetes pods, and bare-metal environments.