homelab-frontend/tasks/4.1-gpu-slot-semaphore.md

4.1 — GPU slot semaphore for reasoning (GREEN)

Phase: 4 — Limits and budgets Stage: GREEN Depends on: 4.3

The reasoning upstream is 2 replicas x --max-num-seqs=4 = 8 concurrent sequence slots cluster-wide. That 8 is a vLLM concurrency setting, not a GPU count — worker-1 has 4 physical GPUs and the two numbers are unrelated. Exceeding 8 does not fail fast; it silently queues inside vLLM where the gateway has no visibility and no ability to cancel.

The gateway serves /v1/* (OpenAI-compatible) over the same predictors. The Anthropic /llm/* dialect was dropped (see tasks/INDEX.md Phase 2) — this controller only ever needs to know about /v1/*.

• Concurrent in-flight requests to reasoning are capped at a configured limit strictly below 8, leaving operator headroom
• The cap, the queue depth, and the queue wait timeout are all explicit in configuration with no silent defaults
• Requests over the cap wait in a bounded queue rather than being rejected immediately
• Once the queue is full, further requests are rejected with a retryable status and a Retry-After, as an application/problem+json document
• A slot is released when the response completes, when the upstream errors, and when the client disconnects mid-stream — no path leaks a slot
• A client that disconnects while still queued never reaches the upstream and never consumes a slot
• The cap applies only to reasoning; other upstreams are unaffected and are not blocked behind its queue
• The controller is keyed by UPSTREAM, never by route or path prefix — reasoning-predictor.llm-serving:80 has exactly one cap and one queue
• Slot occupancy and queue depth are reported per upstream

Verify

# Fire 40 concurrent chat requests at a stub reasoning upstream that holds each for 2s,
# with cap=6 and queue=8 configured.
seq 40 | xargs -P40 -I{} curl -s -o /dev/null -w '%{http_code}\n' \
  -X POST localhost:8080/v1/chat/completions \
  -H 'content-type: application/json' -d '{"model":"reasoning","messages":[]}' | sort | uniq -c
# expected: a mix of 200 and 503; zero 500s

# expected: the stub logged at most 6 simultaneous in-flight requests, never 7
grep -c 'max_concurrent=7' /tmp/stub-reasoning.log
# expected: 0

curl -s -D - -o /dev/null -X POST localhost:8080/v1/chat/completions \
  -H 'content-type: application/json' -d '{"model":"reasoning","messages":[]}' | grep -i retry-after
# expected: Retry-After present on the rejected request