From 4e1f59010ef62b78a749c5da0921841b73a89add Mon Sep 17 00:00:00 2001
From: Josh Avant <830519+joshavant@users.noreply.github.com>
Date: Thu, 14 May 2026 01:39:46 -0500
Subject: [PATCH] fix(gateway): suppress startup liveness warnings (#81699)

* fix(gateway): suppress startup liveness warnings

* docs(changelog): note diagnostic startup grace fix
---
 CHANGELOG.md                   |  1 +
 src/gateway/server.impl.ts     |  5 +++-
 src/logging/diagnostic.test.ts | 42 ++++++++++++++++++++++++++++++++++
 src/logging/diagnostic.ts      |  8 ++++++-
 4 files changed, 54 insertions(+), 2 deletions(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 25c63bd1073..e1f1a467fe0 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -26,6 +26,7 @@ Docs: https://docs.openclaw.ai
 - Hooks: load workspace-relative legacy hook modules from dot-dot-prefixed directories without treating the filename prefix as parent traversal.
 - Plugins: preserve installed package metadata and persisted registry freshness checks for plugin package paths under dot-dot-prefixed directories.
 - Agents: allow dot-dot-prefixed filenames such as `..note.txt` through sandbox FS bridge, remote sandbox reads, and apply_patch summaries without mistaking the name for parent traversal.
+- Gateway/diagnostics: suppress cold-start liveness warnings during the startup grace window while still sampling liveness metrics. Fixes #79915. (#81699) Thanks @joshavant.
 - CLI/migrate: hide per-item source/plugin hints on non-conflicting Codex skill and plugin selection prompts, keeping the hint text reserved for rows that actually need attention. Thanks @sjf.
 - Codex harness: treat high-confidence app-server OAuth refresh invalidation as a terminal auth-profile failure, stopping repeated raw token-refresh errors without turning entitlement or usage-limit payloads into re-auth prompts.
 - CLI/migrate: humanize Codex conflict-status messaging across the migrate UI so selection prompts and plan/result rows say "Codex skill already installed in workspace" instead of surfacing internal `MIGRATION_REASON_*` codes. Thanks @sjf.
diff --git a/src/gateway/server.impl.ts b/src/gateway/server.impl.ts
index 7997105bb09..5c6ade78775 100644
--- a/src/gateway/server.impl.ts
+++ b/src/gateway/server.impl.ts
@@ -593,7 +593,10 @@ export async function startGatewayServer(
   const diagnosticsEnabled = isDiagnosticsEnabled(cfgAtStart);
   setDiagnosticsEnabledForProcess(diagnosticsEnabled);
   if (diagnosticsEnabled) {
-    startDiagnosticHeartbeat(undefined, { getConfig: getRuntimeConfig });
+    startDiagnosticHeartbeat(undefined, {
+      getConfig: getRuntimeConfig,
+      startupGraceMs: 60_000,
+    });
   }
   setGatewaySigusr1RestartPolicy({ allowExternal: isRestartEnabled(cfgAtStart) });
   let getActiveTaskCount = () => 0;
diff --git a/src/logging/diagnostic.test.ts b/src/logging/diagnostic.test.ts
index 89bea0240e6..8d47c8c0b3b 100644
--- a/src/logging/diagnostic.test.ts
+++ b/src/logging/diagnostic.test.ts
@@ -1055,6 +1055,48 @@ describe("stuck session diagnostics threshold", () => {
     );
   });
 
+  it("suppresses liveness warnings during startupGraceMs while still sampling", () => {
+    const warnSpy = vi.spyOn(diagnosticLogger, "warn").mockImplementation(() => undefined);
+    const events: string[] = [];
+    const sampleLiveness = vi.fn(() => ({
+      reasons: ["event_loop_delay" as const],
+      intervalMs: 30_000,
+      eventLoopDelayP99Ms: 1_500,
+      eventLoopDelayMaxMs: 2_000,
+    }));
+    const unsubscribe = onDiagnosticEvent((event) => events.push(event.type));
+
+    try {
+      startDiagnosticHeartbeat(
+        {
+          diagnostics: {
+            enabled: true,
+          },
+        },
+        {
+          emitMemorySample: createEmitMemorySampleMock(),
+          sampleLiveness,
+          startupGraceMs: 60_000,
+        },
+      );
+
+      logMessageQueued({ sessionId: "s1", sessionKey: "main", source: "test" });
+      vi.advanceTimersByTime(30_000);
+
+      expect(sampleLiveness).toHaveBeenCalledTimes(1);
+      expectNoLoggerMessageContaining(warnSpy, "liveness warning:");
+      expect(events).not.toContain("diagnostic.liveness.warning");
+
+      vi.advanceTimersByTime(30_000);
+
+      expect(sampleLiveness).toHaveBeenCalledTimes(2);
+      expectLoggerMessageContaining(warnSpy, "liveness warning:");
+      expect(events).toContain("diagnostic.liveness.warning");
+    } finally {
+      unsubscribe();
+    }
+  });
+
   it("warns for liveness samples when diagnostic work is open", () => {
     const warnSpy = vi.spyOn(diagnosticLogger, "warn").mockImplementation(() => undefined);
 
diff --git a/src/logging/diagnostic.ts b/src/logging/diagnostic.ts
index 9f29fc7d6a5..58bce7a0517 100644
--- a/src/logging/diagnostic.ts
+++ b/src/logging/diagnostic.ts
@@ -123,6 +123,7 @@ type StartDiagnosticHeartbeatOptions = {
   emitMemorySample?: EmitDiagnosticMemorySample;
   sampleLiveness?: SampleDiagnosticLiveness;
   recoverStuckSession?: RecoverStuckSession;
+  startupGraceMs?: number;
 };
 
 let diagnosticLivenessMonitor: EventLoopDelayMonitor | null = null;
@@ -939,6 +940,8 @@ export function startDiagnosticHeartbeat(
     return;
   }
   startDiagnosticLivenessSampler();
+  const livenessGraceUntil =
+    opts?.startupGraceMs != null && opts.startupGraceMs > 0 ? Date.now() + opts.startupGraceMs : 0;
   heartbeatInterval = setInterval(() => {
     let heartbeatConfig = config;
     if (!heartbeatConfig) {
@@ -953,7 +956,10 @@ export function startDiagnosticHeartbeat(
     const now = Date.now();
     pruneDiagnosticSessionStates(now, true);
     const work = getDiagnosticWorkSnapshot(now);
-    const livenessSample = (opts?.sampleLiveness ?? sampleDiagnosticLiveness)(now, work);
+    const inStartupGrace = livenessGraceUntil > 0 && now < livenessGraceUntil;
+    const rawLivenessSample = (opts?.sampleLiveness ?? sampleDiagnosticLiveness)(now, work);
+    // Keep sampling during grace so event-loop delay baselines reset, but suppress startup-only reports.
+    const livenessSample = inStartupGrace ? null : rawLivenessSample;
     const shouldEmitLivenessEvent =
       livenessSample !== null && shouldEmitDiagnosticLivenessEvent(now);
     const shouldEmitLivenessWarning =