[bugfix] deprecate disable log request (#885)

ruizhang0101 · gemini-code-assist[bot] · web-flow · commit 80e2f84ea09b · 2026-03-12T13:19:01.000-07:00
* bugfix: deprecate disable log request

Signed-off-by: Rui Zhang &lt;zrfishnoodles@gmail.com&gt;

* Update helm/README.md

Co-authored-by: gemini-code-assist[bot] &lt;176961590+gemini-code-assist[bot]@users.noreply.github.com&gt;
Signed-off-by: Rui Zhang &lt;51696593+ruizhang0101@users.noreply.github.com&gt;

---------

Signed-off-by: Rui Zhang &lt;zrfishnoodles@gmail.com&gt;
Signed-off-by: Rui Zhang &lt;51696593+ruizhang0101@users.noreply.github.com&gt;
Co-authored-by: gemini-code-assist[bot] &lt;176961590+gemini-code-assist[bot]@users.noreply.github.com&gt;
diff --git a/.github/values-01-2pods-minimal-example.yaml b/.github/values-01-2pods-minimal-example.yaml
@@ -26,7 +26,7 @@ servingEngineSpec:
 
     vllmConfig:
       maxModelLen: 1024
-      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.4"]
+      extraArgs: ["--gpu-memory-utilization", "0.4"]
 
 routerSpec:
   repository: "localhost:5000/git-act-router"
diff --git a/benchmarks/multi-round-qa/model.yaml b/benchmarks/multi-round-qa/model.yaml
@@ -20,7 +20,7 @@ servingEngineSpec:
       enablePrefixCaching: false
       maxModelLen: 4096
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+      extraArgs: ["--gpu-memory-utilization", "0.8"]
 
     lmcacheConfig:
       enabled: true
diff --git a/deployment_on_cloud/gcp/production_stack_specification_basic.yaml b/deployment_on_cloud/gcp/production_stack_specification_basic.yaml
@@ -20,6 +20,6 @@ servingEngineSpec:
         enablePrefixCaching: false
         maxModelLen: 4096
         dtype: "bfloat16"
-        extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+        extraArgs: ["--gpu-memory-utilization", "0.8"]
 
       hf_token: <YOUR HF TOKEN>
diff --git a/docs/source/deployment/crd.rst b/docs/source/deployment/crd.rst
@@ -164,7 +164,6 @@ Sample Configurations
        tensorParallelSize: 1
        gpuMemoryUtilization: "0.8"
        maxLoras: 4
-       extraArgs: ["--disable-log-requests"]
        v1: true
        port: 8000
        # Environment variables
diff --git a/docs/source/deployment/helm.rst b/docs/source/deployment/helm.rst
@@ -73,7 +73,7 @@ Example Snippet
          enablePrefixCaching: false
          maxModelLen: 16384
          dtype: "bfloat16"
-         extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+         extraArgs: ["--gpu-memory-utilization", "0.8"]
 
        hf_token: <YOUR HF TOKEN>
 
@@ -182,7 +182,7 @@ So far, you have configured and deployed a vLLM serving engine with a single GPU
          maxModelLen: 4096
          tensorParallelSize: 2
          dtype: "bfloat16"
-         extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+         extraArgs: ["--gpu-memory-utilization", "0.8"]
        hf_token: <YOUR HF TOKEN>
        shmSize: "20Gi"
 
diff --git a/examples/disaggregated_prefill/start_decode.sh b/examples/disaggregated_prefill/start_decode.sh
@@ -24,7 +24,6 @@ docker run --runtime nvidia --gpus all \
     $IMAGE \
      meta-llama/Llama-3.1-8B-Instruct \
     --port 8200 \
-    --disable-log-requests \
     --enforce-eager \
     --kv-transfer-config \
     '{"kv_connector":"LMCacheConnectorV1","kv_role":"kv_consumer","kv_connector_extra_config": {"discard_partial_chunks": false, "lmcache_rpc_port": "consumer1"}}'
diff --git a/examples/disaggregated_prefill/start_prefill.sh b/examples/disaggregated_prefill/start_prefill.sh
@@ -25,7 +25,6 @@ docker run --runtime nvidia --gpus all \
     $IMAGE \
      meta-llama/Llama-3.1-8B-Instruct \
     --port 8100 \
-    --disable-log-requests \
     --enforce-eager \
     --kv-transfer-config \
     '{"kv_connector":"LMCacheConnectorV1","kv_role":"kv_producer","kv_connector_extra_config": {"discard_partial_chunks": false, "lmcache_rpc_port": "producer1"}}'
diff --git a/helm/README.md b/helm/README.md
@@ -143,7 +143,7 @@ This table documents all available configuration values for the Production Stack
 | `servingEngineSpec.modelSpec[].vllmConfig.gpuMemoryUtilization` | number | `0.9` | The fraction of GPU memory to be used for the model executor (0-1) |
 | `servingEngineSpec.modelSpec[].vllmConfig.runner` | string | `""` | The runner type for the model, can be "auto" or "pooling" |
 | `servingEngineSpec.modelSpec[].vllmConfig.convert` | string | `""` | The conversion type for the model, can be "token_embed", "embed", "token_classify", "classify", or "score" |
-| `servingEngineSpec.modelSpec[].vllmConfig.extraArgs` | list | `["--disable-log-requests"]` | Extra command line arguments to pass to vLLM |
+| `servingEngineSpec.modelSpec[].vllmConfig.extraArgs` | list | `["--trust-remote-code"]` | Extra command line arguments to pass to vLLM |
 
 #### LMCache Configuration
 
diff --git a/helm/values.yaml b/helm/values.yaml
@@ -86,7 +86,7 @@ servingEngineSpec:
   #   - gpuMemoryUtilization: (optional, float) The fraction of GPU memory to be used for the model executor, which can range from 0 to 1. e.g., 0.95
   #   - runner: (optional, string) The runner type for the model, can be "auto" or "pooling". e.g., "pooling"
   #   - convert: (optional, string) The conversion type for the model, can be "token_embed", "embed", "token_classify", "classify", or "score". e.g., "embed"
-  #   - extraArgs: (optional, list) Extra command line arguments to pass to vLLM, e.g., ["--disable-log-requests"]
+  #   - extraArgs: (optional, list) Extra command line arguments to pass to vLLM, e.g., ["--gpu-memory-utilization", "0.4"]
   #
   # - lmcacheConfig: (optional, map) The configuration of the LMCache for KV offloading, supported options are:
   #   - enabled: (optional, bool) Enable LMCache, e.g., true
@@ -204,7 +204,7 @@ servingEngineSpec:
   #     maxNumSeqs: 32
   #     gpuMemoryUtilization: 0.95
   #     maxLoras: 4
-  #     extraArgs: ["--disable-log-requests", "--trust-remote-code"]
+  #     extraArgs: ["--trust-remote-code"]
   #
   #   lmcacheConfig:
   #     enabled: true
diff --git a/operator/config/samples/production-stack_v1alpha1_vllmruntime.yaml b/operator/config/samples/production-stack_v1alpha1_vllmruntime.yaml
@@ -29,7 +29,6 @@ spec:
     tensorParallelSize: 1
     gpuMemoryUtilization: "0.8"
     maxLoras: 4
-    extraArgs: ["--disable-log-requests"]
     v1: true
     port: 8000
     # Environment variables
diff --git a/tutorials/02-basic-vllm-config.md b/tutorials/02-basic-vllm-config.md
@@ -66,7 +66,7 @@ servingEngineSpec:
       enablePrefixCaching: false
       maxModelLen: 16384
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+      extraArgs: ["--gpu-memory-utilization", "0.8"]
 
     hf_token: <YOUR HF TOKEN>
 ```
@@ -172,7 +172,7 @@ servingEngineSpec:
       maxModelLen: 4096
       tensorParallelSize: 2
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+      extraArgs: [ "--gpu-memory-utilization", "0.8"]
     hf_token: <YOUR HF TOKEN>
     shmSize: "20Gi"
 ```
diff --git a/tutorials/07-benchmark-multi-round-qa-single-gpu.md b/tutorials/07-benchmark-multi-round-qa-single-gpu.md
@@ -48,7 +48,7 @@ servingEngineSpec:
       enablePrefixCaching: false
       maxModelLen: 32000
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--swap-space", 0]
+      extraArgs: ["--swap-space", 0]
     lmcacheConfig:
       enabled: true
       cpuOffloadingBufferSize: "60"
@@ -118,7 +118,7 @@ servingEngineSpec:
     vllmConfig:
       enableChunkedPrefill: false
       maxModelLen: 32000
-      extraArgs: ["--disable-log-requests", "--swap-space", 0]
+      extraArgs: [ "--swap-space", 0]
 
     lmcacheConfig:
       enabled: false
diff --git a/tutorials/08-benchmark-multi-round-qa-multi-gpu.md b/tutorials/08-benchmark-multi-round-qa-multi-gpu.md
@@ -52,7 +52,7 @@ servingEngineSpec:
       enablePrefixCaching: false
       maxModelLen: 32000
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--swap-space", 0]
+      extraArgs: ["--swap-space", 0]
     lmcacheConfig:
       enabled: true
       cpuOffloadingBufferSize: "120"
@@ -124,7 +124,7 @@ servingEngineSpec:
       enableChunkedPrefill: false
       enablePrefixCaching: true
       maxModelLen: 32000
-      extraArgs: ["--disable-log-requests", "--swap-space", 0]
+      extraArgs: ["--swap-space", 0]
 
     lmcacheConfig:
       enabled: false
diff --git a/tutorials/09-lora-enabled-installation.md b/tutorials/09-lora-enabled-installation.md
@@ -330,7 +330,7 @@ servingEngineSpec:
         maxModelLen: 4096
         dtype: "bfloat16"
         v1: 1
-        extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+        extraArgs: ["--gpu-memory-utilization", "0.8"]
 
       # Mount Hugging Face credentials and configure LoRA settings
       env:
@@ -453,7 +453,7 @@ servingEngineSpec:
         maxModelLen: 4096
         dtype: "bfloat16"
         v1: 1
-        extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+        extraArgs: ["--gpu-memory-utilization", "0.8"]
 
       # Mount Hugging Face credentials and configure LoRA settings
       env:
@@ -543,7 +543,7 @@ servingEngineSpec:
         maxModelLen: 4096
         dtype: "bfloat16"
         v1: 1
-        extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+        extraArgs: ["--gpu-memory-utilization", "0.8"]
 
       # Mount Hugging Face credentials and configure LoRA settings
       env:
diff --git a/tutorials/11-secure-vllm-serve.md b/tutorials/11-secure-vllm-serve.md
@@ -84,7 +84,7 @@ servingEngineSpec:
         enablePrefixCaching: false
         maxModelLen: 16384
         dtype: "bfloat16"
-        extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+        extraArgs: ["--gpu-memory-utilization", "0.8"]
 
       hf_token: <YOUR HF TOKEN>
 ```
diff --git a/tutorials/assets/gke-example/values-01-offload-kv-cache-local-disk.yaml b/tutorials/assets/gke-example/values-01-offload-kv-cache-local-disk.yaml
@@ -28,7 +28,6 @@ servingEngineSpec:
       tensorParallelSize: 8
       dtype: "bfloat16"
       extraArgs:
-        - "--disable-log-requests"
         - "--max_num_batched_tokens"
         - "2048"
 
diff --git a/tutorials/assets/values-01-2pods-minimal-example.yaml b/tutorials/assets/values-01-2pods-minimal-example.yaml
@@ -15,4 +15,3 @@ servingEngineSpec:
     vllmConfig:
       maxModelLen: 1024
       gpuMemoryUtilization: 0.4
-      extraArgs: ["--disable-log-requests"]
diff --git a/tutorials/assets/values-02-basic-config.yaml b/tutorials/assets/values-02-basic-config.yaml
@@ -20,6 +20,6 @@ servingEngineSpec:
       enablePrefixCaching: false
       maxModelLen: 4096
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+      extraArgs: ["--gpu-memory-utilization", "0.8"]
 
     hf_token: <YOUR HF TOKEN>
diff --git a/tutorials/assets/values-09-lora-helm.yaml b/tutorials/assets/values-09-lora-helm.yaml
@@ -27,7 +27,7 @@ servingEngineSpec:
         maxModelLen: 4096
         dtype: "bfloat16"
         v1: 1
-        extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+        extraArgs: ["--gpu-memory-utilization", "0.8"]
 
       # Mount Hugging Face credentials and configure LoRA settings
       env:
diff --git a/tutorials/assets/values-14-vllm-v1.yaml b/tutorials/assets/values-14-vllm-v1.yaml
@@ -19,7 +19,7 @@ servingEngineSpec:
       enablePrefixCaching: true
       maxModelLen: 4096
       dtype: "bfloat16"
-      extraArgs: ["--disable-log-requests", "--gpu-memory-utilization", "0.8"]
+      extraArgs: ["--gpu-memory-utilization", "0.8"]
 
     lmcacheConfig:
       enabled: true
diff --git a/tutorials/terraform/coreweave/config/llm-stack/helm/gpu/gpu-gpt-oss-20-cw.tpl b/tutorials/terraform/coreweave/config/llm-stack/helm/gpu/gpu-gpt-oss-20-cw.tpl
@@ -41,7 +41,6 @@ servingEngineSpec:
       enablePrefixCaching: true
       dtype:  "bfloat16"  # Changed from "bfloat16" not supported by Tesla T4 GPU (compute capability 7.5)
       extraArgs:
-        - "--disable-log-requests"
         - "--gpu-memory-utilization=0.8"  # GPU-specific optimization
     env: []        # NEW: CPU env vars removed
     initContainer:
@@ -86,7 +85,6 @@ servingEngineSpec:
       enablePrefixCaching: true
       dtype: "bfloat16"
       extraArgs:
-        - "--disable-log-requests"
         - "--gpu-memory-utilization=0.9"
     initContainer:
       name: downloader
diff --git a/tutorials/terraform/coreweave/config/llm-stack/helm/gpu/gpu-gpt-qwn-gem-glm-cw.tpl b/tutorials/terraform/coreweave/config/llm-stack/helm/gpu/gpu-gpt-qwn-gem-glm-cw.tpl
@@ -44,7 +44,6 @@ servingEngineSpec:
       enablePrefixCaching: true
       dtype: "auto"
       extraArgs:
-        - "--disable-log-requests"      # "default quant_method": "mxfp4" is auto-detected by vLLM
         - "--gpu-memory-utilization=0.95"
         - "--tensor-parallel-size=2"   # Mandatory for 120B to run smoothly
     env:
@@ -100,7 +99,6 @@ servingEngineSpec:
       dtype: "bfloat16"
       maxModelLen: 8192
       extraArgs:
-        - "--disable-log-requests"
         - "--gpu-memory-utilization=0.80"
         - "--enforce-eager"
     lmcacheConfig:
@@ -243,7 +241,6 @@ servingEngineSpec:
 #       maxModelLen: 8192
 #       dtype: "bfloat16"
 #       extraArgs:
-#         - "--disable-log-requests"
 #         - "--gpu-memory-utilization=0.60"
 #         - "--trust-remote-code"
 #         - "--enable-auto-tool-choice"
diff --git a/tutorials/terraform/coreweave/config/llm-stack/helm/gpu/gpu-llama-light-ingress-cw.tpl b/tutorials/terraform/coreweave/config/llm-stack/helm/gpu/gpu-llama-light-ingress-cw.tpl
@@ -39,7 +39,6 @@ servingEngineSpec:
     vllmConfig:
       dtype:  "float16"  # Changed from "bfloat16" not supported by Tesla T4 GPU (compute capability 7.5)
       extraArgs:
-        - "--disable-log-requests"
         - "--gpu-memory-utilization=0.8"  # GPU-specific optimization
     env: []        # NEW: CPU env vars removed
     #   - name: VLLM_CPU_KVCACHE_SPACE
diff --git a/tutorials/terraform/eks/modules/llm-stack/helm/gpu/gpu-tinyllama-light-ingress.tpl b/tutorials/terraform/eks/modules/llm-stack/helm/gpu/gpu-tinyllama-light-ingress.tpl
@@ -34,7 +34,6 @@ servingEngineSpec:
     vllmConfig:
       dtype:  "float16"  # Changed from "bfloat16" not supported by Tesla T4 GPU (compute capability 7.5)
       extraArgs:
-        - "--disable-log-requests"
         - "--gpu-memory-utilization=0.8"  # GPU-specific optimization
     env: []        # NEW: CPU env vars removed
     #   - name: VLLM_CPU_KVCACHE_SPACE
@@ -102,4 +101,4 @@ routerSpec:
   # tls:
   #   - secretName: vllm-tls-secret
   #     hosts:
-  #       - vllm-api.yourdomain.com
+  #       - vllm-api.yourdomain.com
diff --git a/tutorials/terraform/nebius/config/llm-stack/helm/gpu/gpu-tinyllama-light-ingress-nebius.tpl b/tutorials/terraform/nebius/config/llm-stack/helm/gpu/gpu-tinyllama-light-ingress-nebius.tpl
@@ -34,7 +34,6 @@ servingEngineSpec:
     vllmConfig:
       dtype:  "float16"   # Using float16 for broader compatibility, though L40S supports bfloat16.
       extraArgs:
-        - "--disable-log-requests"
         - "--gpu-memory-utilization=0.8"  # GPU-specific optimization
     env: []        # NEW: CPU env vars removed
     #   - name: VLLM_CPU_KVCACHE_SPACE
