From 5cc9ee25056ef44372a7539806ae46d499ddaa41 Mon Sep 17 00:00:00 2001
From: MOVZX <movzx@yahoo.com>
Date: Thu, 15 Oct 2026 06:37:23 +0000
Subject: [PATCH] chore: Reset vLLM logging interval back to default

Co-Authored-By: Hinata <hinata@idihore.id>
Signed-off-by: MOVZX <movzx@yahoo.com>
---
 .env.example       | 2 --
 README.md          | 1 -
 docker-compose.yml | 1 -
 start.sh           | 2 --
 4 files changed, 6 deletions(-)

diff --git a/.env.example b/.env.example
index 9eae304..13ae041 100644
--- a/.env.example
+++ b/.env.example
@@ -31,8 +31,6 @@ TOPO_TIMEOUT=20
 GUARD=warn
 # 1 adds the two logging-only counters (exact-restart counters /30 s, FP8 KV clips /300 s).
 COUNTERS=0
-# Seconds between engine stats lines (throughput, KV usage, queue depth).
-VLLM_LOG_STATS_INTERVAL=1
 
 # Extracted from orcarouter/Qwen3.8-Flash-Next-Uncensored, hidden 2560
 QWEN38_ABLITERATION=Qwen3.8-Flash-Next-Uncensored
diff --git a/README.md b/README.md
index ae60e95..0f9572b 100644
--- a/README.md
+++ b/README.md
@@ -252,5 +252,4 @@ docker run -d --name vllm --gpus all --stop-timeout 20 --shm-size=16g \
 | `CACHE_HOST_DIR`          | `.vllm-cache` in this directory                       | Where the compile caches live. The host path uses the same folder.               |
 | `CUDA_VISIBLE_DEVICES`    | `0,1,2,3`                                             | The GPU lane, in bus order.                                                      |
 | `PLE_HOME`                | `0`                                                   | The card that hosts the PLE offload. That card gets no CUDA context for it.      |
-| `VLLM_LOG_STATS_INTERVAL` | `1`                                                   | Seconds between engine stats lines (throughput, KV use, queue).                  |
 | `MAX_JOBS`, `ARCH`        | `32`, `8.6`                                           | Compile jobs and GPU architecture, for the build only.                           |
diff --git a/docker-compose.yml b/docker-compose.yml
index b2f0d1e..d2433a6 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -40,7 +40,6 @@ services:
       TOPO_TIMEOUT: ${TOPO_TIMEOUT:-20}
       GUARD: ${GUARD:-warn}
       COUNTERS: ${COUNTERS:-0}
-      VLLM_LOG_STATS_INTERVAL: ${VLLM_LOG_STATS_INTERVAL:-1}
       QWEN38_ABLITERATION: ${QWEN38_ABLITERATION:-Qwen3.8-Flash-Next-Uncensored}
     volumes:
       - "${MODEL_DIR:?Set MODEL_DIR in .env}:${MODEL_MOUNT_PATH:-/models/Qwen3.8-Flash-Next}:ro"
diff --git a/start.sh b/start.sh
index 0cb179b..9fe487c 100755
--- a/start.sh
+++ b/start.sh
@@ -22,7 +22,6 @@
 #   PLE_HOME    GPU that hosts the PLE offload home (0)
 #   GUARD       bounds-guard telemetry off|warn|raise (warn)
 #   COUNTERS=1    logging-only counters (top-k exact restarts /30s, FP8 KV clips /300s)
-#   VLLM_LOG_STATS_INTERVAL  seconds between engine stats lines (unset = vLLM default 1 s)
 #   CUDA_HOME   toolkit for first-serve JIT; default: the venv's CUDA 13.0 wheels
 #   AUTO_TOPO   0 (default, fixed shipped settings) | 1 | strict -> scripts/topo_select.py
 #   AUTO_TOPO_NCCL=default  AUTO_TOPO_BIND=1  LANE_GPUS=auto  LANE_CPUS=...  TOPO_TIMEOUT=20
@@ -204,7 +203,6 @@ export VLLM_HOST_EMBED_TABLE=1
 export VLLM_USE_BREAKABLE_CUDAGRAPH=0
 export VLLM_QSA_KV_SCALES="${SCALES:-$CKPT/qsa_kv_scales_swift.json}"
 # engine stats line (throughput, KV usage, queue). Set empty to keep vLLM's default.
-if [ -n "${VLLM_LOG_STATS_INTERVAL:-}" ]; then export VLLM_LOG_STATS_INTERVAL; fi
 export HF_HUB_OFFLINE=1 HF_DATASETS_OFFLINE=1
 # the kernel toggles (the shipped values). These names are the tree's own: envs.py in the
 # serving tree registers MERLIN_* for the kernel toggles and VLLM_E2_* for the runtime knobs.