MB2: parameterize vLLM roles (kv_producer + kv_consumer default)

start_vllm_pair.sh
  ROLE_A / ROLE_B env vars (default kv_producer / kv_consumer for strict
  PD-disagg). Override to kv_both for the kv_both control. The role is
  injected into --kv-transfer-config so vLLM imposes the role restriction.

mb2_kv_transfer.py
  --skip-verify flag drops step 3 (the plain completion sanity-check on
  the destination), required when the dst is kv_consumer-only since a
  kv_consumer instance refuses to serve a request without
  do_remote_prefill. The transfer-time itself is still measured from
  step 2 (do_remote_prefill on the consumer).

Also: per-step client-side wall-clock timestamps (t_step1_client_unix,
t_step2_client_unix, t_step2_end_unix) are now captured so the
post-hoc breakdown analyzer can join with the per-instance JSONL logs
on absolute time.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

microbench/fresh_setup/start_vllm_pair.sh

@@ -29,6 +29,8 @@ BP_A=8998
 BP_B=8999
 MASTER_A=29500
 MASTER_B=29501
+ROLE_A="${ROLE_A:-kv_producer}"  # or kv_both
+ROLE_B="${ROLE_B:-kv_consumer}"  # or kv_both
 
 MB2_LOG_ROOT="${FRESH_ROOT}/mb2_transfer_logs"
 SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
@@ -78,8 +80,9 @@ fi
 mkdir -p "${MB2_LOG_ROOT}/A" "${MB2_LOG_ROOT}/B"
 
 launch() {
-    local idx="$1" gpu="$2" port="$3" bp="$4" master="$5"
-    echo "[mb2] launching instance ${idx} on GPU ${gpu}, port ${port}, bp ${bp}"
+    local idx="$1" gpu="$2" port="$3" bp="$4" master="$5" role="$6"
+    local cfg="{\"kv_connector\":\"MooncakeConnector\",\"kv_role\":\"${role}\"}"
+    echo "[mb2] launching ${idx}: gpu=${gpu} port=${port} bp=${bp} role=${role}"
     PYTHONHASHSEED=42 \
     VLLM_MOONCAKE_BOOTSTRAP_PORT="${bp}" \
     CUDA_VISIBLE_DEVICES="${gpu}" \
@@ -91,15 +94,15 @@ launch() {
         --trust-remote-code --enable-prefix-caching \
         --dtype auto --gpu-memory-utilization 0.9 \
         --max-model-len 200000 \
-        --kv-transfer-config '{"kv_connector":"MooncakeConnector","kv_role":"kv_both"}' \
+        --kv-transfer-config "${cfg}" \
         --enable-prompt-tokens-details \
         > "${LOGS_DIR}/vllm_${idx}_gpu${gpu}.log" 2>&1 &
     disown
 }
 
-launch A "${GPU_A}" "${PORT_A}" "${BP_A}" "${MASTER_A}"
+launch A "${GPU_A}" "${PORT_A}" "${BP_A}" "${MASTER_A}" "${ROLE_A}"
 sleep 3
-launch B "${GPU_B}" "${PORT_B}" "${BP_B}" "${MASTER_B}"
+launch B "${GPU_B}" "${PORT_B}" "${BP_B}" "${MASTER_B}" "${ROLE_B}"
 
 echo "[mb2] waiting for both /health endpoints..."
 for port in "${PORT_A}" "${PORT_B}"; do