Kova OpenClaw Runtime Findings

Run: kova-2026-05-03T130714Z
Target: local-build:/home/runner/_work/openclaw/openclaw
Mode: execution
Platform: linux 6.5.13 (x64)

Failure Brief

Decision: FAIL
Primary blocker: agent-cold-warm-message/mock-openai-provider
Why: ocm @'kova-agent-cold-warm-message-mock-openai-provider-r1-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 80080ms, over threshold 45000ms
Evidence:
- peakRssMb: 3591
- cpuPercentMax: 301.3
- coldAgentTurnMs: 80080
- warmAgentTurnMs: 50517
- agentColdWarmDeltaMs: 29563
- cold provider was fast (2ms), but OpenClaw spent 79777ms before provider work.
- agent-cli: 3591MB RSS, 301.3% CPU
- agent-process: 3591MB RSS, 301.3% CPU
Likely owner: OpenClaw
Paste to fixer:
Investigate OpenClaw release gate failure agent-cold-warm-message/mock-openai-provider. Kova decision was FAIL on linux-x64. Primary evidence: ocm @'kova-agent-cold-warm-message-mock-openai-provider-r1-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 80080ms, over threshold 45000ms. Measurements: peakRssMb: 3591; cpuPercentMax: 301.3; coldAgentTurnMs: 80080; warmAgentTurnMs: 50517; agentColdWarmDeltaMs: 29563; cold provider was fast (2ms), but OpenClaw spent 79777ms before provider work.; agent-cli: 3591MB RSS, 301.3% CPU; agent-process: 3591MB RSS, 301.3% CPU. Dependency/plugin load errors were zero, so focus on startup, memory, CPU, gateway readiness, runtime deps staging, provider/model load, and UI asset initialization. Likely owner area: OpenClaw.

Recommended next scenario

Reason: ocm @'kova-agent-cold-warm-message-mock-openai-provider-r1-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 80080ms, over threshold 45000ms
Command: node bin/kova.mjs run --target local-build:/home/runner/_work/openclaw/openclaw --scenario agent-cold-warm-message --state mock-openai-provider --execute --profile-on-failure --retain-on-failure --json

Scenario: agent-cold-warm-message
Result: FAIL
Cleanup: destroyed
Measurements:
- startup: listening unknown; health unknown; readiness unknown; gateway disabled; restarts 0
- resources: peak RSS 3591 MB; max CPU 301.3%; samples 144; roles agent-cli 3591MB/301.3%, agent-process 3591MB/301.3%, command-tree 3591MB/301.3%, package-manager 2809MB/260%
- agent: turn 80080ms; cold/warm 80080ms/50517ms; cold-warm delta 29563ms; pre-provider 79777ms; provider 2ms; cleanup unknown; diagnosis pre-provider-stall; leaks 0
- plugins/runtime: missing deps 0; plugin failures 0; runtime deps 6030.67ms (acpx); warm restages unknown; warm reuse unknown
- diagnostics: timeline available; slowest span runtimeDeps.stage 6030.67ms; embedded traces 0; liveness warnings 0; open spans 0 (0 required); node CPU/heap/trace 0/0/0
Violations:
- ocm @'kova-agent-cold-warm-message-mock-openai-provider-r1-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 80080ms, over threshold 45000ms
- ocm @'kova-agent-cold-warm-message-mock-openai-provider-r1-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 50517ms, over threshold 45000ms
- peak RSS 3591 MB exceeded threshold 900 MB
- command-tree peak RSS 3591 MB exceeded threshold 1400 MB
- package-manager peak RSS 2809 MB exceeded threshold 900 MB
- agent-cli peak RSS 3591 MB exceeded threshold 900 MB
- agent-cli max CPU 301.3% exceeded threshold 300%
- agent-process peak RSS 3591 MB exceeded threshold 900 MB
- agent-process max CPU 301.3% exceeded threshold 300%
- cold agent spent 79777ms before provider work, over threshold 10000ms
- warm agent spent 50348ms before provider work, over threshold 10000ms
- cold agent turn took 80080ms, over threshold 45000ms
- warm agent turn took 50517ms, over threshold 15000ms
- cold pre-provider latency was 79777ms, over threshold 10000ms
- warm pre-provider latency was 50348ms, over threshold 10000ms
- cold provider was fast (2ms), but OpenClaw spent 79777ms before provider work.

Scenario: agent-cold-warm-message
Result: FAIL
Cleanup: destroyed
Measurements:
- startup: listening unknown; health unknown; readiness unknown; gateway disabled; restarts 0
- resources: peak RSS 3317.6 MB; max CPU 319.7%; samples 132; roles agent-cli 3317.6MB/319.7%, agent-process 3317.6MB/319.7%, command-tree 3317.6MB/319.7%, package-manager 2497.1MB/273%
- agent: turn 71311ms; cold/warm 71311ms/46853ms; cold-warm delta 24458ms; pre-provider 71095ms; provider 2ms; cleanup unknown; diagnosis pre-provider-stall; leaks 0
- plugins/runtime: missing deps 0; plugin failures 0; runtime deps 4370.28ms (acpx); warm restages unknown; warm reuse unknown
- diagnostics: timeline available; slowest span runtimeDeps.stage 4370.28ms; embedded traces 0; liveness warnings 0; open spans 0 (0 required); node CPU/heap/trace 0/0/0
Violations:
- ocm @'kova-agent-cold-warm-message-mock-openai-provider-r2-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 71311ms, over threshold 45000ms
- ocm @'kova-agent-cold-warm-message-mock-openai-provider-r2-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 46853ms, over threshold 45000ms
- peak RSS 3317.6 MB exceeded threshold 900 MB
- command-tree peak RSS 3317.6 MB exceeded threshold 1400 MB
- package-manager peak RSS 2497.1 MB exceeded threshold 900 MB
- agent-cli peak RSS 3317.6 MB exceeded threshold 900 MB
- agent-cli max CPU 319.7% exceeded threshold 300%
- agent-process peak RSS 3317.6 MB exceeded threshold 900 MB
- agent-process max CPU 319.7% exceeded threshold 300%
- cold agent spent 71095ms before provider work, over threshold 10000ms
- warm agent spent 46641ms before provider work, over threshold 10000ms
- cold agent turn took 71311ms, over threshold 45000ms
- warm agent turn took 46853ms, over threshold 15000ms
- cold pre-provider latency was 71095ms, over threshold 10000ms
- warm pre-provider latency was 46641ms, over threshold 10000ms
- cold provider was fast (2ms), but OpenClaw spent 71095ms before provider work.

Scenario: agent-cold-warm-message
Result: FAIL
Cleanup: destroyed
Measurements:
- startup: listening unknown; health unknown; readiness unknown; gateway disabled; restarts 0
- resources: peak RSS 3529 MB; max CPU 304.2%; samples 130; roles agent-cli 3529MB/304.2%, agent-process 3529MB/304.2%, command-tree 3529MB/304.2%, package-manager 2761.7MB/258%
- agent: turn 70908ms; cold/warm 70908ms/45861ms; cold-warm delta 25047ms; pre-provider 70676ms; provider 2ms; cleanup unknown; diagnosis pre-provider-stall; leaks 0
- plugins/runtime: missing deps 0; plugin failures 0; runtime deps 4631.53ms (acpx); warm restages unknown; warm reuse unknown
- diagnostics: timeline available; slowest span runtimeDeps.stage 4631.53ms; embedded traces 0; liveness warnings 0; open spans 0 (0 required); node CPU/heap/trace 0/0/0
Violations:
- ocm @'kova-agent-cold-warm-message-mock-openai-provider-r3-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 70908ms, over threshold 45000ms
- ocm @'kova-agent-cold-warm-message-mock-openai-provider-r3-kova-2026-05-03t130714z' -- agent --local --agent main --session-id kova-agent-cold-warm --message 'Reply with exact ASCII text KOVA_AGENT_OK only.' --thinking off --timeout 120 --json took 45861ms, over threshold 45000ms
- peak RSS 3529 MB exceeded threshold 900 MB
- command-tree peak RSS 3529 MB exceeded threshold 1400 MB
- package-manager peak RSS 2761.7 MB exceeded threshold 900 MB
- agent-cli peak RSS 3529 MB exceeded threshold 900 MB
- agent-cli max CPU 304.2% exceeded threshold 300%
- agent-process peak RSS 3529 MB exceeded threshold 900 MB
- agent-process max CPU 304.2% exceeded threshold 300%
- cold agent spent 70676ms before provider work, over threshold 10000ms
- warm agent spent 45713ms before provider work, over threshold 10000ms
- cold agent turn took 70908ms, over threshold 45000ms
- warm agent turn took 45861ms, over threshold 15000ms
- cold pre-provider latency was 70676ms, over threshold 10000ms
- warm pre-provider latency was 45713ms, over threshold 10000ms
- cold provider was fast (2ms), but OpenClaw spent 70676ms before provider work.
