Dual 3090 AI Workload Causes System Hard Lockup
fgoricha · reddit · 2026-07-18
The author is troubleshooting a dual RTX 3090 rig: both cards run extended AI tasks perfectly fine individually, but under combined load, the entire machine hard-locks after a few hours. Monitors lose signal, the network drops, fans and LEDs keep spinning, and a hard reboot is required.
Known details:
- Motherboard: EVGA X299 FTW K, CPU: i9-7940X, RAM: 128GB, PSU: 1600W.
- Both 3090s are in dedicated x16 PCIe slots with a one-slot gap between them.
- The issue occurs at both 100% and 80% power limits, with temps and power draw appearing normal before the crash.
The author is seeking next diagnostic steps, highlighting broader stability issues with multi-GPU local training/inference.
More from Infra
- Resin lets an agent build standalone LLM inference engines; vLLM and SGLang optimize DeepSeek kernels differently — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- Alchemy adds effect/Workflow support across Durable Objects, Workflows and Lambda Durable Functions — samgoodwin89 · 2026-10-27
- Alchemy's Lambda Web Functions run plain Node web servers on Lambda, no OpenNext needed — samgoodwin89 · 2026-10-27
- Alchemy Cuts Cloudflare Bindings CPU Time by Up to 166× Via Profiling — samgoodwin89 · 2026-10-27
- Foldkit Adds SSR Support With a Few Lines of Config on Cloudflare and AWS — samgoodwin89 · 2026-10-27