惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

J
Java Code Geeks
博客园 - 司徒正美
博客园 - 【当耐特】
爱范儿
爱范儿
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
IT之家
IT之家
人人都是产品经理
人人都是产品经理
雷峰网
雷峰网
酷 壳 – CoolShell
酷 壳 – CoolShell
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
大猫的无限游戏
大猫的无限游戏
月光博客
月光博客
宝玉的分享
宝玉的分享
V
V2EX
S
SegmentFault 最新的问题
V
Visual Studio Blog
阮一峰的网络日志
阮一峰的网络日志
Martin Fowler
Martin Fowler
Jina AI
Jina AI
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
博客园_首页
L
LangChain Blog
D
Docker
腾讯CDC

Recent Commits to openclaw:main

test: merge chat side-result checks · openclaw/openclaw@ddd2c2a test: merge cron history checks · openclaw/openclaw@f7eb746 test: merge responsive navigation shell checks · openclaw/openclaw@c2e4b47 docs(changelog): add codex oauth fixes · openclaw/openclaw@628e6cd test: merge navigation routing cases · openclaw/openclaw@5d8cecb Tests: mock channel registry bundled fallback · openclaw/openclaw@2b08233 Secrets: avoid broad web search discovery for single plugin config · openclaw/openclaw@a464f59 test: merge config view browser checks · openclaw/openclaw@20cf511 fix(status): align oauth health with runtime · openclaw/openclaw@eed7116 feat: add macOS screen snapshots for monitor preview (#67954) thanks … · openclaw/openclaw@f377db1 fix: report shared auth scopes in hello-ok (#67810) thanks @BunsDev · openclaw/openclaw@0b6c39b Auto-reply: avoid eager bundled route fallback · openclaw/openclaw@3ea1bf4 Tests: narrow session binding contract setup · openclaw/openclaw@54e4e16 fix(macOS): enable undo/redo in webchat composer text input (#34962) · openclaw/openclaw@00951dc Tests: speed up channel setup promotion · openclaw/openclaw@82b529a Docs: refresh agent instructions · openclaw/openclaw@5775fe2 fix(auth): serialize OAuth refresh across agents to fix #26322 (#67876) · openclaw/openclaw@8e79080 test: allow ollama public surface boundary test · openclaw/openclaw@7d4f1a6 Docs: add test performance guardrails · openclaw/openclaw@89706d3 Tests: restore context-engine usage proof · openclaw/openclaw@e4c4f95 Tests: slim context engine runtime coverage · openclaw/openclaw@74c198f ci: retry failed custom checkouts · openclaw/openclaw@0ee5baf test: trim duplicate provider auth onboarding cases · openclaw/openclaw@1ffc02e matrix: fix sessions_spawn --thread subagent session spawning (#67643) · openclaw/openclaw@1ce2596 test: reduce auth choice fixture churn · openclaw/openclaw@857b9cd test: mock health status config boundaries · openclaw/openclaw@9d5ab4a test: mock onboard config io boundary · openclaw/openclaw@299694d test: mock legacy state plugin boundaries · openclaw/openclaw@2713089 test: mock channel install boundaries · openclaw/openclaw@b945248 test: mock doctor preview channel boundaries · openclaw/openclaw@b1a3ad4
fix: classify provider conversation state errors (#82616)...
dutifulbob · 2026-05-17 · via Recent Commits to openclaw:main

@@ -1,5 +1,4 @@

11

import crypto from "node:crypto";

2-

import fs from "node:fs";

32

import {

43

hasOutboundReplyContent,

54

resolveSendableOutboundReplyParts,

@@ -48,7 +47,6 @@ import { runEmbeddedPiAgent } from "../../agents/pi-embedded.js";

4847

import { buildAgentRuntimeOutcomePlan } from "../../agents/runtime-plan/build.js";

4948

import {

5049

resolveGroupSessionKey,

51-

resolveSessionTranscriptPath,

5250

type SessionEntry,

5351

updateSessionStore,

5452

} from "../../config/sessions.js";

@@ -98,6 +96,10 @@ import {

9896

} from "./agent-runner-utils.js";

9997

import { type BlockReplyPipeline } from "./block-reply-pipeline.js";

10098

import { resolveOriginMessageProvider } from "./origin-routing.js";

99+

import {

100+

classifyProviderRequestError,

101+

PROVIDER_CONVERSATION_STATE_ERROR_USER_MESSAGE,

102+

} from "./provider-request-error-classifier.js";

101103

import type { FollowupRun } from "./queue.js";

102104

import { createBlockReplyDeliveryHandler } from "./reply-delivery.js";

103105

import type { ReplyMediaContext } from "./reply-media-paths.js";

@@ -456,16 +458,6 @@ function isPureBillingSummary(err: unknown): boolean {

456458

);

457459

}

458460459-

function isToolResultTurnMismatchError(message: string): boolean {

460-

const lower = normalizeLowercaseStringOrEmpty(message);

461-

return (

462-

lower.includes("toolresult") &&

463-

lower.includes("tooluse") &&

464-

lower.includes("exceeds the number") &&

465-

lower.includes("previous turn")

466-

);

467-

}

468-469461

function collapseRepeatedFailureDetail(message: string): string {

470462

const parts = message

471463

.split(/\s+\|\s+/u)

@@ -582,6 +574,13 @@ function buildExternalRunFailureReply(

582574

options?: { includeDetails?: boolean; isHeartbeat?: boolean },

583575

): ExternalRunFailureReply {

584576

const normalizedMessage = collapseRepeatedFailureDetail(message);

577+

const providerRequestError = classifyProviderRequestError(normalizedMessage);

578+

if (providerRequestError) {

579+

return {

580+

text: providerRequestError.userMessage,

581+

isGenericRunnerFailure: false,

582+

};

583+

}

585584

const missingApiKeyFailure = buildMissingApiKeyFailureText(normalizedMessage);

586585

if (missingApiKeyFailure) {

587586

return { text: missingApiKeyFailure, isGenericRunnerFailure: false };

@@ -603,12 +602,6 @@ function buildExternalRunFailureReply(

603602

if (options?.isHeartbeat) {

604603

return { text: HEARTBEAT_EXTERNAL_RUN_FAILURE_TEXT, isGenericRunnerFailure: false };

605604

}

606-

if (isToolResultTurnMismatchError(normalizedMessage)) {

607-

return {

608-

text: "⚠️ Session history got out of sync. Please try again, or use /new to start a fresh session.",

609-

isGenericRunnerFailure: false,

610-

};

611-

}

612605

const cliBackendTimeoutFailure = buildCliBackendTimeoutFailureText(normalizedMessage);

613606

if (cliBackendTimeoutFailure) {

614607

return { text: cliBackendTimeoutFailure, isGenericRunnerFailure: false };

@@ -2254,16 +2247,18 @@ export async function runAgentTurnWithFallback(params: {

22542247

};

22552248

}

22562249

if (embeddedError?.kind === "role_ordering") {

2257-

const didReset = await params.resetSessionAfterRoleOrderingConflict(embeddedError.message);

2258-

if (didReset) {

2259-

params.replyOperation?.fail("run_failed", embeddedError);

2260-

return {

2261-

kind: "final",

2262-

payload: markAgentRunFailureReplyPayload({

2263-

text: "⚠️ Message ordering conflict. I've reset the conversation - please try again.",

2264-

}),

2265-

};

2266-

}

2250+

const providerRequestError = classifyProviderRequestError(embeddedError);

2251+

params.replyOperation?.fail("run_failed", embeddedError);

2252+

const embeddedErrorText = formatErrorMessage(embeddedError).replace(/\.\s*$/, "");

2253+

return {

2254+

kind: "final",

2255+

payload: markAgentRunFailureReplyPayload({

2256+

text: shouldSurfaceToControlUi

2257+

? `⚠️ Agent failed before reply: ${embeddedErrorText}.\nLogs: openclaw logs --follow`

2258+

: (providerRequestError?.userMessage ??

2259+

PROVIDER_CONVERSATION_STATE_ERROR_USER_MESSAGE),

2260+

}),

2261+

};

22672262

}

2268226322692264

break;

@@ -2318,8 +2313,8 @@ export async function runAgentTurnWithFallback(params: {

23182313

: isBillingErrorMessage(message);

23192314

const isContextOverflow = !isBilling && isLikelyContextOverflowError(message);

23202315

const isCompactionFailure = !isBilling && isCompactionFailureError(message);

2321-

const isSessionCorruption = /function call turn comes immediately after/i.test(message);

2322-

const isRoleOrderingError = /incorrect role information|roles must alternate/i.test(message);

2316+

const providerRequestError =

2317+

!isBilling && !shouldSurfaceToControlUi ? classifyProviderRequestError(err) : undefined;

23232318

const isTransientHttp = isTransientHttpError(message);

2324231923252320

if (isReplyOperationRestartAbort(params.replyOperation)) {

@@ -2382,61 +2377,12 @@ export async function runAgentTurnWithFallback(params: {

23822377

}),

23832378

};

23842379

}

2385-

if (isRoleOrderingError) {

2386-

const didReset = await params.resetSessionAfterRoleOrderingConflict(message);

2387-

if (didReset) {

2388-

params.replyOperation?.fail("run_failed", err);

2389-

return {

2390-

kind: "final",

2391-

payload: markAgentRunFailureReplyPayload({

2392-

text: "⚠️ Message ordering conflict. I've reset the conversation - please try again.",

2393-

}),

2394-

};

2395-

}

2396-

}

2397-2398-

// Auto-recover from Gemini session corruption by resetting the session

2399-

if (

2400-

isSessionCorruption &&

2401-

params.sessionKey &&

2402-

params.activeSessionStore &&

2403-

params.storePath

2404-

) {

2405-

const sessionKey = params.sessionKey;

2406-

const corruptedSessionId = params.getActiveSessionEntry()?.sessionId;

2407-

defaultRuntime.error(

2408-

`Session history corrupted (Gemini function call ordering). Resetting session: ${params.sessionKey}`,

2409-

);

2410-2411-

try {

2412-

// Delete transcript file if it exists

2413-

if (corruptedSessionId) {

2414-

const transcriptPath = resolveSessionTranscriptPath(corruptedSessionId);

2415-

try {

2416-

fs.unlinkSync(transcriptPath);

2417-

} catch {

2418-

// Ignore if file doesn't exist

2419-

}

2420-

}

2421-2422-

// Keep the in-memory snapshot consistent with the on-disk store reset.

2423-

delete params.activeSessionStore[sessionKey];

2424-2425-

// Remove session entry from store using a fresh, locked snapshot.

2426-

await updateSessionStore(params.storePath, (store) => {

2427-

delete store[sessionKey];

2428-

});

2429-

} catch (cleanupErr) {

2430-

defaultRuntime.error(

2431-

`Failed to reset corrupted session ${params.sessionKey}: ${String(cleanupErr)}`,

2432-

);

2433-

}

2434-2435-

params.replyOperation?.fail("session_corruption_reset", err);

2380+

if (providerRequestError) {

2381+

params.replyOperation?.fail("run_failed", err);

24362382

return {

24372383

kind: "final",

24382384

payload: markAgentRunFailureReplyPayload({

2439-

text: "⚠️ Session history was corrupted. I've reset the conversation - please try again!",

2385+

text: providerRequestError.userMessage,

24402386

}),

24412387

};

24422388

}

@@ -2481,7 +2427,6 @@ export async function runAgentTurnWithFallback(params: {

24812427

!(isRateLimit && !isOverloadedErrorMessage(message)) &&

24822428

!rateLimitOrOverloadedCopy &&

24832429

!isContextOverflow &&

2484-

!isRoleOrderingError &&

24852430

!shouldSurfaceToControlUi

24862431

? buildExternalRunFailureReply(message, {

24872432

includeDetails: isVerboseFailureDetailEnabled(params.resolvedVerboseLevel),

@@ -2499,11 +2444,9 @@ export async function runAgentTurnWithFallback(params: {

24992444

? rateLimitOrOverloadedCopy

25002445

: isContextOverflow

25012446

? "⚠️ Context overflow — prompt too large for this model. Try a shorter message or a larger-context model."

2502-

: isRoleOrderingError

2503-

? "⚠️ Message ordering conflict - please try again. If this persists, use /new to start a fresh session."

2504-

: shouldSurfaceToControlUi

2505-

? `⚠️ Agent failed before reply: ${trimmedMessage}.\nLogs: openclaw logs --follow`

2506-

: (externalRunFailureReply?.text ?? genericFallbackText);

2447+

: shouldSurfaceToControlUi

2448+

? `⚠️ Agent failed before reply: ${trimmedMessage}.\nLogs: openclaw logs --follow`

2449+

: (externalRunFailureReply?.text ?? genericFallbackText);

25072450

const userVisibleFallbackText = resolveExternalRunFailureTextForConversation({

25082451

text: fallbackText,

25092452

sessionCtx: params.sessionCtx,