Bug type
Regression (worked before, now fails)
Beta release blocker
No
Summary
After upgrading to version 5.12, I sent a simple "Hello, are you there?" message through Feishu. The reply started to appear but then timed out after one minute. Almost everything timed out—the model, the memory system, etc. I waited about 5 minutes and finally got a normal reply. A simple question took five minutes because of timeouts. This needs to be fixed.
Steps to reproduce
My computer is a Legion R9000P, Node.js is sufficient. I only have proactive and improving active skills, no plugins, and I'm using the default memory feature. Feishu was just upgraded to Lark 5.13. I don't know how else to describe it because I've been using 4.23 all along—it's the most stable, with no errors and no reply timeouts. Long tasks are just a bit slower, that's all.
That's all for the configuration. All I did was send a "Hello" message in Feishu Lark.
Expected behavior
The new version added many features and fixed a lot of bugs. Based on my errors below, I'm sure others have the same timeout issue. I hope these timeout problems can be completely fixed because wasting too much time on timeouts is very frustrating
It's much more powerful now, but stability—like the 4.23 version—is still what many people want most.
Actual behavior
Everything that could time out did time out. It eventually gives an answer, but a simple "Hello" takes five minutes.
OpenClaw version
5.12
Operating system
windows11
Install method
npm
Model
minimax
Provider / routing chain
openclaw -> minimax-portal/MiniMax-M2.7
Additional provider/model setup details
Primary: openclaw -> minimax-portal/MiniMax-M2.7
Fallback: openclaw -> custom-api-deepseek-com/deepseek-chat
I'm not sure about the intermediate gateway. I only configured MiniMax and DeepSeek.
Logs, screenshots, and evidence
12:12:24 [feishu] feishu[default]: message om_x100b6f5d9ca7c0b0c3788aadf88f372 in chat oc_22a98a24f1cb233ee06bdf5da052ddb9 — immediate
12:12:24 plugin "openclaw-lark" runtime config.loadConfig() is deprecated (runtime-config-load-write); use config.current(). Source: C:\Users\Lenovo\.openclaw\extensions\openclaw-lark\index.js
12:12:25 [feishu] feishu[default]: received message from ou_46455b0cca3c46506b766aeef317a259 in oc_22a98a24f1cb233ee06bdf5da052ddb9 (p2p)
12:12:25 [feishu] feishu[default]: dispatching to agent (session=agent:main:feishu:direct:ou_46455b0cca3c46506b766aeef317a259)
12:13:06 [diagnostic] liveness warning: reasons=event_loop_delay,event_loop_utilization,cpu interval=41s eventLoopDelayP99Ms=40567.3 eventLoopDelayMaxMs=40567.3 eventLoopUtilization=1 cpuCoreRatio=0.999 active=1 waiting=0 queued=1 phase=channels.feishu.start-account recentPhases=sidecars.main-session-recovery:4ms,sidecars.restart-sentinel:21ms,post-attach.update-sentinel:14ms,sidecars.model-prewarm:5250ms,sidecars.session-locks:3128ms,post-ready.maintenance:1096ms work=[active=agent:main:feishu:direct:ou_46455b0cca3c46506b766aeef317a259(processing,q=1,age=41s) queued=agent:main:feishu:direct:ou_46455b0cca3c46506b766aeef317a259(processing,q=1,age=41s)]
12:14:02 [feishu] feishu[lobster-bot]: reaction event on message om_x100b6f5d9ca7c0b0c3788aadf88f372
12:14:02 [feishu] feishu[lobster-bot]: reaction event on message om_x100b6f5d9ca7c0b0c3788aadf88f372
12:14:02 [feishu] feishu[lobster-bot]: duplicate reaction om_x100b6f5d9ca7c0b0c3788aadf88f372:reaction:Typing:, skipping
12:14:02 [info]: [ '[ws]', 'reconnect' ]
12:14:02 [agent/embedded] [trace:embedded-run] startup stages: runId=7c2d7359-290e-49c9-b88e-1df024d56a3a sessionId=554a4b5f-7a1f-483b-95b0-6c5d8b90e9e4 phase=attempt-dispatch totalMs=51381 stages=workspace:1ms@1ms,runtime-plugins:70ms@71ms,hooks:0ms@71ms,model-resolution:283ms@354ms,auth:48727ms@49081ms,context-engine:2ms@49083ms,attempt-dispatch:2298ms@51381ms
12:14:09 [tools] tools.profile (full) allowlist contains unknown entries (canvas). These entries are shipped core tools but unavailable in the current runtime/provider/model/config.
12:14:09 [tools] agents.main.tools.allow allowlist contains unknown entries (canvas). These entries are shipped core tools but unavailable in the current runtime/provider/model/config.
12:14:09 [agent/embedded] [trace:embedded-run] core-plugin-tool stages: runId=7c2d7359-290e-49c9-b88e-1df024d56a3a sessionId=554a4b5f-7a1f-483b-95b0-6c5d8b90e9e4 phase=core-plugin-tools totalMs=7294 stages=tool-policy:4004ms@4004ms,workspace-policy:0ms@4004ms,base-coding-tools:1ms@4005ms,shell-tools:0ms@4005ms,openclaw-tools:session-workspace:2ms@4007ms,openclaw-tools:image-tool:139ms@4146ms,openclaw-tools:image-generate-tool:47ms@4193ms,openclaw-tools:video-generate-tool:42ms@4235ms,openclaw-tools:music-generate-tool:39ms@4274ms,openclaw-tools:pdf-tool:0ms@4274ms,openclaw-tools:web-search-tool:0ms@4274ms,openclaw-tools:web-fetch-tool:0ms@4274ms,openclaw-tools:message-tool:3ms@4277ms,openclaw-tools:nodes-tool:0ms@4277ms,openclaw-tools:core-tool-list:3ms@4280ms,openclaw-tools:plugin-tools:2979ms@7259ms,openclaw-tools:0ms@7259ms,message-provider-policy:0ms@7259ms,model-provider-policy:0ms@7259ms,authorization-policy:27ms@7286ms,schema-normalization:7ms@7293ms,tool-hooks:0ms@7293ms,abort-wrappers:0ms@7293ms,deferred-followup-descriptions:1ms@7294ms,attempt:create-openclaw-coding-tools:0ms@7294ms,attempt:tools-allow:0ms@7294ms
12:14:46 [agent/embedded] [trace:embedded-run] prep stages: runId=7c2d7359-290e-49c9-b88e-1df024d56a3a sessionId=554a4b5f-7a1f-483b-95b0-6c5d8b90e9e4 phase=stream-ready totalMs=44369 stages=workspace-sandbox:217ms@217ms,skills:1ms@218ms,core-plugin-tools:7295ms@7513ms,bootstrap-context:212ms@7725ms,bundle-tools:5691ms@13416ms,system-prompt:2668ms@16084ms,session-resource-loader:28247ms@44331ms,agent-session:4ms@44335ms,stream-setup:34ms@44369ms
12:14:49 [diagnostic] stuck session: sessionId=unknown sessionKey=agent:main:feishu:direct:ou_46455b0cca3c46506b766aeef317a259 state=processing age=143s queueDepth=1 reason=queued_work_without_active_run classification=stale_session_state recovery=checking
12:14:49 [diagnostic] stuck session recovery outcome: status=skipped action=keep_lane sessionId=554a4b5f-7a1f-483b-95b0-6c5d8b90e9e4 sessionKey=agent:main:feishu:direct:ou_46455b0cca3c46506b766aeef317a259 activeSessionId=554a4b5f-7a1f-483b-95b0-6c5d8b90e9e4 activeWorkKind=embedded_run reason=active_reply_work
12:14:49 [error]: [ '[ws]', 'timeout of 15000ms exceeded' ]
12:14:49 [info]: [ 'ws', 'unable to connect to the server after trying 1 times")' ]
12:15:42 [memory] embeddings rate limited; retrying in 522ms
12:15:42 [memory] embeddings rate limited; retrying in 562ms
12:15:42 [memory] embeddings rate limited; retrying in 582ms
12:15:42 [memory] embeddings rate limited; retrying in 537ms
12:16:04 [memory] embeddings rate limited; retrying in 1161ms
12:16:04 [memory] embeddings rate limited; retrying in 1102ms
12:16:04 [memory] embeddings rate limited; retrying in 1063ms
12:16:04 [memory] embeddings rate limited; retrying in 1157ms
12:16:26 [memory] embeddings rate limited; retrying in 2342ms
12:16:26 [memory] embeddings rate limited; retrying in 2099ms
12:16:26 [memory] embeddings rate limited; retrying in 2348ms
12:16:26 [memory] embeddings rate limited; retrying in 2269ms
Impact and severity
Is there anyone else with this problem? Some people have been experiencing this issue consistently in the 5.x versions. I'm sure those with the same symptoms as me also want a stable and secure version. Otherwise, over time, it will gradually kill their enthusiasm. I'm raising this issue to help make OpenClaw better.
Additional information
I think 4.23 is the most stable. Although its features aren't the strongest, the response speed, task execution, and conversations are all fine—no major issues. Starting from 4.27, I began experiencing timeouts, though the speed was still acceptable. With 4.29, it became unusable. The 5.x versions are completely full of timeouts and can't even reply to messages. In 5.12, everything times out but eventually responds after a long delay, wasting a lot of time.
Bug type
Regression (worked before, now fails)
Beta release blocker
No
Summary
After upgrading to version 5.12, I sent a simple "Hello, are you there?" message through Feishu. The reply started to appear but then timed out after one minute. Almost everything timed out—the model, the memory system, etc. I waited about 5 minutes and finally got a normal reply. A simple question took five minutes because of timeouts. This needs to be fixed.
Steps to reproduce
My computer is a Legion R9000P, Node.js is sufficient. I only have proactive and improving active skills, no plugins, and I'm using the default memory feature. Feishu was just upgraded to Lark 5.13. I don't know how else to describe it because I've been using 4.23 all along—it's the most stable, with no errors and no reply timeouts. Long tasks are just a bit slower, that's all.
That's all for the configuration. All I did was send a "Hello" message in Feishu Lark.
Expected behavior
The new version added many features and fixed a lot of bugs. Based on my errors below, I'm sure others have the same timeout issue. I hope these timeout problems can be completely fixed because wasting too much time on timeouts is very frustrating
It's much more powerful now, but stability—like the 4.23 version—is still what many people want most.
Actual behavior
Everything that could time out did time out. It eventually gives an answer, but a simple "Hello" takes five minutes.
OpenClaw version
5.12
Operating system
windows11
Install method
npm
Model
minimax
Provider / routing chain
openclaw -> minimax-portal/MiniMax-M2.7
Additional provider/model setup details
Primary: openclaw -> minimax-portal/MiniMax-M2.7
Fallback: openclaw -> custom-api-deepseek-com/deepseek-chat
I'm not sure about the intermediate gateway. I only configured MiniMax and DeepSeek.
Logs, screenshots, and evidence
Impact and severity
Is there anyone else with this problem? Some people have been experiencing this issue consistently in the 5.x versions. I'm sure those with the same symptoms as me also want a stable and secure version. Otherwise, over time, it will gradually kill their enthusiasm. I'm raising this issue to help make OpenClaw better.
Additional information
I think 4.23 is the most stable. Although its features aren't the strongest, the response speed, task execution, and conversations are all fine—no major issues. Starting from 4.27, I began experiencing timeouts, though the speed was still acceptable. With 4.29, it became unusable. The 5.x versions are completely full of timeouts and can't even reply to messages. In 5.12, everything times out but eventually responds after a long delay, wasting a lot of time.