fix(failover): classify z.ai network_error stop reason as retryable timeout (#43884)

Merged via squash.

Prepared head SHA: 9660f6cd5bcb8d073fc5575bbba2bf3792b29de3
Co-authored-by: hougangdev <105773686+hougangdev@users.noreply.github.com>
Co-authored-by: altaywtf <9790196+altaywtf@users.noreply.github.com>
Reviewed-by: @altaywtf

CHANGELOG.md

@@ -54,6 +54,7 @@ Docs: https://docs.openclaw.ai
 - Doctor/gateway service audit: earlier groundwork for this fix landed in the superseded #28338 branch. Thanks @realriphub.
 - Gateway/session stores: regenerate the Swift push-test protocol models and align Windows native session-store realpath handling so protocol checks and sync session discovery stop drifting on Windows. (#44266) thanks @jalehman.
 - Context engine/session routing: forward optional `sessionKey` through context-engine lifecycle calls so plugins can see structured routing metadata during bootstrap, assembly, post-turn ingestion, and compaction. (#44157) thanks @jalehman.
+- Agents/failover: classify z.ai `network_error` stop reasons as retryable timeouts so provider connectivity failures trigger fallback instead of surfacing raw unhandled-stop-reason errors. (#43884) Thanks @hougangdev.
 
 ## 2026.3.11
 

src/agents/failover-error.test.ts

@@ -289,6 +289,9 @@ describe("failover-error", () => {
     expect(resolveFailoverReasonFromError({ message: "stop reason: error" })).toBe("timeout");
     expect(resolveFailoverReasonFromError({ message: "reason: abort" })).toBe("timeout");
     expect(resolveFailoverReasonFromError({ message: "reason: error" })).toBe("timeout");
+    expect(
+      resolveFailoverReasonFromError({ message: "Unhandled stop reason: network_error" }),
+    ).toBe("timeout");
   });
 
   it("infers timeout from connection/network error messages", () => {

src/agents/pi-embedded-helpers.isbillingerrormessage.test.ts

@@ -576,6 +576,19 @@ describe("isFailoverErrorMessage", () => {
     }
   });
 
+  it("matches z.ai network_error stop reason as timeout", () => {
+    const samples = [
+      "Unhandled stop reason: network_error",
+      "stop reason: network_error",
+      "reason: network_error",
+    ];
+    for (const sample of samples) {
+      expect(isTimeoutErrorMessage(sample)).toBe(true);
+      expect(classifyFailoverReason(sample)).toBe("timeout");
+      expect(isFailoverErrorMessage(sample)).toBe(true);
+    }
+  });
+
   it("does not classify MALFORMED_FUNCTION_CALL as timeout", () => {
     const sample = "Unhandled stop reason: MALFORMED_FUNCTION_CALL";
     expect(isTimeoutErrorMessage(sample)).toBe(false);

src/agents/pi-embedded-helpers/failover-matches.ts

@@ -47,9 +47,9 @@ const ERROR_PATTERNS = {
     /\benotfound\b/i,
     /\beai_again\b/i,
     /without sending (?:any )?chunks?/i,
-    /\bstop reason:\s*(?:abort|error|malformed_response)\b/i,
-    /\breason:\s*(?:abort|error|malformed_response)\b/i,
-    /\bunhandled stop reason:\s*(?:abort|error|malformed_response)\b/i,
+    /\bstop reason:\s*(?:abort|error|malformed_response|network_error)\b/i,
+    /\breason:\s*(?:abort|error|malformed_response|network_error)\b/i,
+    /\bunhandled stop reason:\s*(?:abort|error|malformed_response|network_error)\b/i,
   ],
   billing: [
     /["']?(?:status|code)["']?\s*[:=]\s*402\b|\bhttp\s*402\b|\berror(?:\s+code)?\s*[:=]?\s*402\b|\b(?:got|returned|received)\s+(?:a\s+)?402\b|^\s*402\s+payment/i,