Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off

Liquid cooling flow dropped below the safe threshold. This is a facilities fault with a very short fuse, direct-liquid-cooled GPUs have little thermal mass, so an emergency power-off follows quickly and every job on the affected loop dies at once. This entry explains how to confirm the cause, apply the fix, and separate it from adjacent blackwell failures.

The literal signature is "Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off". It is a hardware failure associated with Blackwell, Grace Hopper, NVLink fabrics, and liquid cooling. The line identifies the failing operation or subsystem, while the surrounding evidence decides whether it is the initiating fault or a downstream symptom.

Liquid cooling flow dropped below the safe threshold. This is a facilities fault with a very short fuse, direct-liquid-cooled GPUs have little thermal mass, so an emergency power-off follows quickly and every job on the affected loop dies at once. The failure becomes visible at this call site because the operation first requires the missing resource, valid state, healthy peer, or correct result. Earlier log lines and a known-good control carry more causal value than the final wrapper exception.

• The workload stops or loses forward progress after emitting "Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off".
• A retry on the same configuration reproduces the failure because the causal state has not changed.
• The outer framework exception can hide the rank, node, allocation, or dependency that failed first.
• Increasing timeouts or reducing workload size can suppress the symptom without correcting the cause.

• Blackwell, Grace Hopper, NVLink fabrics, and liquid cooling
• production-shaped multi-accelerator workloads
• containerized and bare-metal deployments of the same stack

Apply the following checklist to a small reproduction: each box below is a positive signal that you are looking at this exact failure rather than a sibling in the same taxonomy.

• ✓Find the first occurrence of "Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off" and preserve at least 100 lines before it.
• ✓Identify which rank, node, device, or process emitted the first related warning.
• ✓check pump status, filter differential pressure, and for trapped air in the secondary loop. A gradual flow decline usually means a clogging filter; a sudden one means a pump or valve.
• ✓Repeat the same input after the targeted change and require the signature to disappear.
• ✓Resume from latest checkpoint, on a different cooling loop only after the control passes.

Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off

Timestamps and exact values vary across runs, but the pattern. An info-level start, an early WARN, an ERROR carrying the symptom. Is the actual fingerprint you should alert on. The Denpex platform flags this combination automatically.

treat as a facility incident, not a GPU one. Confirm whether the power-off already fired, and stop scheduling onto the affected loop. This changes the condition in the causal diagnosis instead of hiding the outer exception. The repeated control proves ownership before recovery from latest checkpoint, on a different cooling loop.

# Preserve evidence before restarting
dmesg -T | grep -iE 'NVRM|Xid|AER|NVLink|ECC'
nvidia-smi -q
dcgmi diag -r 3

# Find the exact signature in the complete log
rg -n -F -- "Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off" <log-file>

Adapt the snippet to your framework. The same pattern holds for PyTorch Lightning, Hugging Face Trainer, DeepSpeed, Megatron-LM, and vLLM training wrappers. Where the wrapper exposes a config flag (for examplelr_scheduler_type in Trainer), prefer the flag over the imperative API to keep the schedule declarative and reproducible.

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.

• Liquid cooling flow dropped below the safe threshold. This is a facilities fault with a very short fuse, direct-liquid-cooled GPUs have little thermal mass, so an emergency power-off follows quickly and every job on the affected loop dies at once.
• The decisive evidence is the first log line that precedes "Coolant Distribution Unit CDU secondary loop flow rate low immediate system power-off" and differs from a healthy run.
• A nearby failure remains a competing hypothesis until a control separates configuration, capacity, transport, and hardware causes.