fix(gateway): harden message action channel fallback and startup grace

steipete · steipete · commit 71cd33713727 · 2026-03-03T01:17:27.000Z
Take the safe, tested subset from #32367:\n- per-channel startup connect grace in health monitor\n- tool-context channel-provider fallback for message actions\n\nCo-authored-by: Munem Hashmi <munem.hashmi@gmail.com>
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -32,6 +32,7 @@ Docs: https://docs.openclaw.ai
 
 ### Fixes
 
+- Gateway/message tool reliability: avoid false `Unknown channel` failures when `message.*` actions receive platform-specific channel ids by falling back to `toolContext.currentChannelProvider`, and prevent health-monitor restart thrash for channels that just (re)started by adding a per-channel startup-connect grace window. (from #32367) Thanks @MunemHashmi.
 - WebChat/markdown tables: ensure GitHub-flavored markdown table parsing is explicitly enabled at render time and add horizontal overflow handling for wide tables, with regression coverage for table-only and mixed text+table content. (#32365) Thanks @BlueBirdBack.
 - Feishu/default account resolution: always honor explicit `channels.feishu.defaultAccount` during outbound account selection (including top-level-credential setups where the preferred id is not present in `accounts`), instead of silently falling back to another account id. (#32253) Thanks @bmendonca3.
 - Gemini schema sanitization: coerce malformed JSON Schema `properties` values (`null`, arrays, primitives) to `{}` before provider validation, preventing downstream strict-validator crashes on invalid plugin/tool schemas. (#32332) Thanks @webdevtodayjason.
diff --git a/src/gateway/channel-health-monitor.test.ts b/src/gateway/channel-health-monitor.test.ts
@@ -201,6 +201,7 @@ describe("channel-health-monitor", () => {
   });
 
   it("restarts a stuck channel (running but not connected)", async () => {
+    const now = Date.now();
     const manager = createSnapshotManager({
       whatsapp: {
         default: {
@@ -209,6 +210,7 @@ describe("channel-health-monitor", () => {
           enabled: true,
           configured: true,
           linked: true,
+          lastStartAt: now - 300_000,
         },
       },
     });
@@ -219,6 +221,41 @@ describe("channel-health-monitor", () => {
     monitor.stop();
   });
 
+  it("skips recently-started channels while they are still connecting", async () => {
+    const now = Date.now();
+    const manager = createSnapshotManager({
+      discord: {
+        default: {
+          running: true,
+          connected: false,
+          enabled: true,
+          configured: true,
+          lastStartAt: now - 5_000,
+        },
+      },
+    });
+    await expectNoRestart(manager);
+  });
+
+  it("respects custom per-channel startup grace", async () => {
+    const now = Date.now();
+    const manager = createSnapshotManager({
+      discord: {
+        default: {
+          running: true,
+          connected: false,
+          enabled: true,
+          configured: true,
+          lastStartAt: now - 30_000,
+        },
+      },
+    });
+    const monitor = await startAndRunCheck(manager, { channelStartupGraceMs: 60_000 });
+    expect(manager.stopChannel).not.toHaveBeenCalled();
+    expect(manager.startChannel).not.toHaveBeenCalled();
+    monitor.stop();
+  });
+
   it("restarts a stopped channel that gave up (reconnectAttempts >= 10)", async () => {
     const manager = createSnapshotManager({
       discord: {
diff --git a/src/gateway/channel-health-monitor.ts b/src/gateway/channel-health-monitor.ts
@@ -17,6 +17,7 @@ const ONE_HOUR_MS = 60 * 60_000;
  * alive (health checks pass) but Slack silently stops delivering events.
  */
 const DEFAULT_STALE_EVENT_THRESHOLD_MS = 30 * 60_000;
+const DEFAULT_CHANNEL_STARTUP_GRACE_MS = 120_000;
 
 export type ChannelHealthMonitorDeps = {
   channelManager: ChannelManager;
@@ -25,6 +26,7 @@ export type ChannelHealthMonitorDeps = {
   cooldownCycles?: number;
   maxRestartsPerHour?: number;
   staleEventThresholdMs?: number;
+  channelStartupGraceMs?: number;
   abortSignal?: AbortSignal;
 };
 
@@ -50,14 +52,20 @@ function isChannelHealthy(
     lastEventAt?: number | null;
     lastStartAt?: number | null;
   },
-  opts: { now: number; staleEventThresholdMs: number },
+  opts: { now: number; staleEventThresholdMs: number; channelStartupGraceMs: number },
 ): boolean {
   if (!isManagedAccount(snapshot)) {
     return true;
   }
   if (!snapshot.running) {
     return false;
   }
+  if (snapshot.lastStartAt != null) {
+    const upDuration = opts.now - snapshot.lastStartAt;
+    if (upDuration < opts.channelStartupGraceMs) {
+      return true;
+    }
+  }
   if (snapshot.connected === false) {
     return false;
   }
@@ -88,6 +96,7 @@ export function startChannelHealthMonitor(deps: ChannelHealthMonitorDeps): Chann
     cooldownCycles = DEFAULT_COOLDOWN_CYCLES,
     maxRestartsPerHour = DEFAULT_MAX_RESTARTS_PER_HOUR,
     staleEventThresholdMs = DEFAULT_STALE_EVENT_THRESHOLD_MS,
+    channelStartupGraceMs = DEFAULT_CHANNEL_STARTUP_GRACE_MS,
     abortSignal,
   } = deps;
 
@@ -132,7 +141,7 @@ export function startChannelHealthMonitor(deps: ChannelHealthMonitorDeps): Chann
           if (channelManager.isManuallyStopped(channelId as ChannelId, accountId)) {
             continue;
           }
-          if (isChannelHealthy(status, { now, staleEventThresholdMs })) {
+          if (isChannelHealthy(status, { now, staleEventThresholdMs, channelStartupGraceMs })) {
             continue;
           }
 
diff --git a/src/infra/outbound/message-action-runner.test.ts b/src/infra/outbound/message-action-runner.test.ts
@@ -349,6 +349,37 @@ describe("runMessageAction context isolation", () => {
     expect(result.channel).toBe("slack");
   });
 
+  it("falls back to tool-context provider when channel param is an id", async () => {
+    const result = await runDrySend({
+      cfg: slackConfig,
+      actionParams: {
+        channel: "C12345678",
+        target: "#C12345678",
+        message: "hi",
+      },
+      toolContext: { currentChannelId: "C12345678", currentChannelProvider: "slack" },
+    });
+
+    expect(result.kind).toBe("send");
+    expect(result.channel).toBe("slack");
+  });
+
+  it("falls back to tool-context provider for broadcast channel ids", async () => {
+    const result = await runDryAction({
+      cfg: slackConfig,
+      action: "broadcast",
+      actionParams: {
+        targets: ["channel:C12345678"],
+        channel: "C12345678",
+        message: "hi",
+      },
+      toolContext: { currentChannelProvider: "slack" },
+    });
+
+    expect(result.kind).toBe("broadcast");
+    expect(result.channel).toBe("slack");
+  });
+
   it("blocks cross-provider sends by default", async () => {
     await expect(
       runDrySend({
diff --git a/src/infra/outbound/message-action-runner.ts b/src/infra/outbound/message-action-runner.ts
@@ -217,13 +217,28 @@ async function maybeApplyCrossContextMarker(params: {
   });
 }
 
-async function resolveChannel(cfg: OpenClawConfig, params: Record<string, unknown>) {
+async function resolveChannel(
+  cfg: OpenClawConfig,
+  params: Record<string, unknown>,
+  toolContext?: { currentChannelProvider?: string },
+) {
   const channelHint = readStringParam(params, "channel");
-  const selection = await resolveMessageChannelSelection({
-    cfg,
-    channel: channelHint,
-  });
-  return selection.channel;
+  try {
+    const selection = await resolveMessageChannelSelection({
+      cfg,
+      channel: channelHint,
+    });
+    return selection.channel;
+  } catch (error) {
+    if (channelHint && toolContext?.currentChannelProvider) {
+      const fallback = normalizeMessageChannel(toolContext.currentChannelProvider);
+      if (fallback && isDeliverableMessageChannel(fallback)) {
+        params.channel = fallback;
+        return fallback;
+      }
+    }
+    throw error;
+  }
 }
 
 async function resolveActionTarget(params: {
@@ -317,7 +332,7 @@ async function handleBroadcastAction(
   }
   const targetChannels =
     channelHint && channelHint.trim().toLowerCase() !== "all"
-      ? [await resolveChannel(input.cfg, { channel: channelHint })]
+      ? [await resolveChannel(input.cfg, { channel: channelHint }, input.toolContext)]
       : configured;
   const results: Array<{
     channel: ChannelId;
@@ -754,7 +769,7 @@ export async function runMessageAction(
     }
   }
 
-  const channel = await resolveChannel(cfg, params);
+  const channel = await resolveChannel(cfg, params, input.toolContext);
   let accountId = readStringParam(params, "accountId") ?? input.defaultAccountId;
   if (!accountId && resolvedAgentId) {
     const byAgent = buildChannelAccountBindings(cfg).get(channel);
