Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank

A LoRA adapter was requested whose rank is larger than the engine reserved at startup. vLLM preallocates LoRA slots sized by max_lora_rank, so a higher-rank adapter cannot be loaded into the running engine. This entry explains how to confirm the cause, apply the fix, and separate it from adjacent fp8-serving failures.

The literal signature is "Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank". It is a infrastructure failure associated with FP8 serving with SGLang, FlashInfer, and vLLM. The line identifies the failing operation or subsystem, while the surrounding evidence decides whether it is the initiating fault or a downstream symptom.

A LoRA adapter was requested whose rank is larger than the engine reserved at startup. vLLM preallocates LoRA slots sized by max_lora_rank, so a higher-rank adapter cannot be loaded into the running engine. 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 "Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank".
• 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.

• FP8 serving with SGLang, FlashInfer, and vLLM
• 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 "Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank" and preserve at least 100 lines before it.
• ✓Identify which rank, node, device, or process emitted the first related warning.
• ✓inspect the adapters you intend to serve, the rank is the r field in adapter_config.json. Set max_lora_rank to the maximum across all of them, not to the first one that failed.
• ✓Repeat the same input after the targeted change and require the signature to disappear.
• ✓Resume from engine restart with the corrected max_lora_rank only after the control passes.

Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank

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.

restart the engine with --max-lora-rank set to at least the highest rank you serve. It cannot be changed on a live engine. This changes the condition in the causal diagnosis instead of hiding the outer exception. The repeated control proves ownership before recovery from engine restart with the corrected max_lora_rank.

# Preserve evidence before restarting
rg -n -i 'error|exception|timeout|failed' <log-file>
nvidia-smi
python -m torch.utils.collect_env

# Find the exact signature in the complete log
rg -n -F -- "Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank" <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.

• A LoRA adapter was requested whose rank is larger than the engine reserved at startup. vLLM preallocates LoRA slots sized by max_lora_rank, so a higher-rank adapter cannot be loaded into the running engine.
• The decisive evidence is the first log line that precedes "Active LoRA adapter rank exceeds engine pre-allocated max_lora_rank" and differs from a healthy run.
• A nearby failure remains a competing hypothesis until a control separates configuration, capacity, transport, and hardware causes.