openclaw 异常之 increase agents.defaults.timeoutSeconds in your config.

目录异常回复openclaw日志ollama日志解决方法异常回复openclaw 调用本地 ollama 模型进行 cpu 推理时遇到如下回复Request timed out before a response was generated. Please try again, or increase agents.defaults.timeoutSeconds in your config.openclaw日志查看 openclaw 日志显示如下[agent]Profile ollama:default timed out. Trying next account...ollama日志查看 ollama 日志显示如下time2026-04-13T13:08:24.57908:00 levelINFO sourceserver.go:259 msgenabling flash attentiontime2026-04-13T13:08:24.58008:00 levelINFO sourceserver.go:444 msgstarting runner cmd/usr/bin/ollama runner --ollama-engine --model /root/.ollama/models/blobs/sha256-afb707b6b8fac6e475acc42bc8380fc0b8d2e0e4190be5a969fbf62fcc897db5 --port 45997time2026-04-13T13:08:24.58008:00 levelINFO sourcesched.go:484 msgsystem memory total62.7 GiB free56.4 GiB free_swap31.4 GiBtime2026-04-13T13:08:24.58008:00 levelINFO sourceserver.go:771 msgloading model model layers25 requested-1time2026-04-13T13:08:24.58708:00 levelINFO sourcerunner.go:1417 msgstarting ollama enginetime2026-04-13T13:08:24.58708:00 levelINFO sourcerunner.go:1452 msgServer listening on 127.0.0.1:45997time2026-04-13T13:08:24.59108:00 levelINFO sourcerunner.go:1290 msgload request{Operation:fit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:8 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}time2026-04-13T13:08:24.63108:00 levelINFO sourceggml.go:136 msg architectureqwen35 file_typeQ8_0 name description num_tensors536 num_key_values52time2026-04-13T13:08:24.63508:00 levelINFO sourceggml.go:104 msgsystem CPU.0.SSE31 CPU.0.SSSE31 CPU.0.LLAMAFILE1 compilercgo(gcc)time2026-04-13T13:08:26.01108:00 levelINFO sourcerunner.go:1290 msgload request{Operation:alloc LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:8 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}time2026-04-13T13:08:28.05608:00 levelINFO sourcerunner.go:1290 msgload request{Operation:commit LoraPath:[] Parallel:1 BatchSize:512 FlashAttention:Enabled KvSize:262144 KvCacheType: NumThreads:8 GPULayers:[] MultiUserCache:false ProjectorPath: MainGPU:0 UseMmap:false}time2026-04-13T13:08:28.05608:00 levelINFO sourceggml.go:482 msgoffloading 0 repeating layers to GPUtime2026-04-13T13:08:28.05608:00 levelINFO sourceggml.go:486 msgoffloading output layer to CPUtime2026-04-13T13:08:28.05608:00 levelINFO sourceggml.go:494 msgoffloaded 0/25 layers to GPUtime2026-04-13T13:08:28.05608:00 levelINFO sourcedevice.go:245 msgmodel weights deviceCPU size1.2 GiBtime2026-04-13T13:08:28.05608:00 levelINFO sourcedevice.go:256 msgkv cache deviceCPU size3.5 GiBtime2026-04-13T13:08:28.05608:00 levelINFO sourcedevice.go:267 msgcompute graph deviceCPU size1.2 GiBtime2026-04-13T13:08:28.05608:00 levelINFO sourcedevice.go:272 msgtotal memory size5.9 GiBtime2026-04-13T13:08:28.05608:00 levelINFO sourcesched.go:561 msgloaded runners count1time2026-04-13T13:08:28.05608:00 levelINFO sourceserver.go:1364 msgwaiting for llama runner to start respondingtime2026-04-13T13:08:28.05608:00 levelINFO sourceserver.go:1398 msgwaiting for server to become available statusllm server loading modeltime2026-04-13T13:08:28.30708:00 levelINFO sourceserver.go:1402 msgllama runner started in 3.73 seconds[GIN] 2026/04/13 - 13:09:24 |500| 59.908078928s | 172.16.200.251 |POST/api/chat解决方法原因openclaw 默认超时时长为 60 秒cpu 推理较慢所以需延长超时时间。vi /root/.openclaw/openclaw.json增加llm: {idleTimeoutSeconds: 1800}agents: { defaults: { workspace: /root/.openclaw/workspace, model: { primary: xxxxxx }, models: { xxxxxx: {} }, compaction: { mode: safeguard }, llm: {idleTimeoutSeconds: 1800} } },