fix: fail stale codex thread resume
This commit is contained in:
@@ -452,9 +452,7 @@ async function runCodexStdioTurnWithSession(options: CodexStdioTurnOptions, sess
|
||||
emitEvent({ type: "backend_status", payload: { phase: "thread/resume:completed", threadId } });
|
||||
} catch (error) {
|
||||
const failure = normalizeFailure(error);
|
||||
if (!isMissingRolloutThreadResumeFailure(failure)) throw error;
|
||||
emitEvent({ type: "backend_status", payload: threadResumeNonResumablePayload(options.threadId, failure) });
|
||||
threadId = await startThread("thread/replacement-start");
|
||||
throw threadResumeFailure(options.threadId, failure);
|
||||
}
|
||||
} else {
|
||||
threadId = await startThread();
|
||||
@@ -910,12 +908,6 @@ function normalizeFailure(error: unknown): CodexStdioFailure {
|
||||
return new CodexStdioFailure(classifyMessageFailureKind(message, "backend-protocol-error"), message, "codex-stdio");
|
||||
}
|
||||
|
||||
function isMissingRolloutThreadResumeFailure(error: CodexStdioFailure): boolean {
|
||||
if (error.phase !== "response:thread/resume") return false;
|
||||
const text = `${error.message}\n${JSON.stringify(error.details)}`.toLowerCase();
|
||||
return text.includes("no rollout found for thread id");
|
||||
}
|
||||
|
||||
function threadResumeFailure(threadId: string, error: CodexStdioFailure): CodexStdioFailure {
|
||||
return new CodexStdioFailure(
|
||||
"thread-resume-failed",
|
||||
@@ -931,18 +923,6 @@ function threadResumeFailure(threadId: string, error: CodexStdioFailure): CodexS
|
||||
);
|
||||
}
|
||||
|
||||
function threadResumeNonResumablePayload(threadId: string, error: CodexStdioFailure): JsonRecord {
|
||||
return {
|
||||
phase: "thread/resume:non-resumable",
|
||||
requestedThreadId: threadId,
|
||||
originalFailureKind: error.failureKind,
|
||||
originalPhase: error.phase,
|
||||
originalDetails: redactJson(error.details),
|
||||
replacement: "thread/start",
|
||||
valuesPrinted: false,
|
||||
};
|
||||
}
|
||||
|
||||
function classifyCodexErrorRecord(error: JsonRecord, fallback: FailureKind): FailureKind {
|
||||
const parts: string[] = [];
|
||||
if (typeof error.message === "string") parts.push(error.message);
|
||||
|
||||
@@ -105,19 +105,19 @@ const selfTest: SelfTestCase = async (context) => {
|
||||
env: { CODEX_HOME: context.codexHome, AGENTRUN_FAKE_CODEX_MODE: "resume-no-rollout" },
|
||||
oneShot: true,
|
||||
}) as JsonRecord;
|
||||
assert.equal(staleThreadResult.terminalStatus, "completed", "stale thread resume should create a replacement thread and continue the turn");
|
||||
assert.equal(staleThreadResult.failureKind, null, "replacement stale thread turn should not expose thread-resume-failed as terminal failure");
|
||||
assert.equal(staleThreadResult.terminalStatus, "failed", "stale thread resume must fail instead of starting a replacement thread");
|
||||
assert.equal(staleThreadResult.failureKind, "thread-resume-failed", "stale thread resume must expose thread-resume-failed as the terminal failure");
|
||||
const staleEnvelope = await client.get(`/api/v1/runs/${staleThread.runId}/commands/${staleThread.commandId}/result`) as JsonRecord;
|
||||
assert.equal(staleEnvelope.terminalStatus, "completed");
|
||||
assert.equal(staleEnvelope.failureKind, null);
|
||||
assert.equal(staleEnvelope.completed, true);
|
||||
assert.equal((staleEnvelope.sessionRef as JsonRecord).threadId, "thread_selftest_1");
|
||||
assert.equal(staleEnvelope.terminalStatus, "failed");
|
||||
assert.equal(staleEnvelope.failureKind, "thread-resume-failed");
|
||||
assert.equal(staleEnvelope.completed, false);
|
||||
assert.equal((staleEnvelope.sessionRef as JsonRecord).threadId, "thread_missing_rollout");
|
||||
const staleEvents = await client.get(`/api/v1/runs/${staleThread.runId}/events?afterSeq=0&limit=100`) as { items?: Array<{ type: string; payload: unknown }> };
|
||||
const stalePhases = (staleEvents.items ?? []).filter((event) => event.type === "backend_status").map((event) => String(eventPayload(event).phase ?? ""));
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "backend_status" && eventPayload(event).phase === "thread/resume:non-resumable"), true, "stale resume must be visible as non-resumable before replacement");
|
||||
assert.equal(stalePhases.some((phase) => phase === "thread/replacement-start:completed"), true, "stale resume must start a replacement thread");
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "backend_status" && eventPayload(event).phase === "thread/resume:non-resumable"), false, "stale resume must not be converted into a replacement path");
|
||||
assert.equal(stalePhases.some((phase) => phase === "thread/replacement-start:completed"), false, "stale resume must not start a replacement thread");
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "backend_status" && eventPayload(event).phase === "thread/resume:completed"), false, "stale resume must not be reported as a successful resume");
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "error" && eventPayload(event).failureKind === "thread-resume-failed"), false, "replacement stale resume must not surface terminal thread-resume-failed error");
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "error" && eventPayload(event).failureKind === "thread-resume-failed"), true, "stale resume must surface terminal thread-resume-failed error");
|
||||
assertNoSecretLeak({ staleThreadResult, staleEnvelope, staleEvents });
|
||||
|
||||
const live = await createRunWithCommand(client, context, "hello live events", "selftest-live-tool-events", 15_000);
|
||||
@@ -186,7 +186,7 @@ const selfTest: SelfTestCase = async (context) => {
|
||||
await runSecretFailureCase({ client, managerUrl: server.baseUrl, context });
|
||||
await runSpawnFailureCase({ client, managerUrl: server.baseUrl, context });
|
||||
|
||||
return { name: "codex-stdio", tests: ["runner-lease-heartbeat", "codex-stdio-fake-turn", "codex-stdio-projected-writable-home", "codex-stdio-deepseek-profile-fake-turn", "codex-stdio-minimax-m3-profile-fake-turn", "codex-stdio-deepseek-missing-secret-no-fallback", "codex-stdio-minimax-m3-missing-secret-no-fallback", "codex-stdio-config-model-authoritative", "codex-stdio-explicit-model-forwarded", "codex-stdio-final-agent-message-only", "codex-stdio-stale-thread-replacement", "codex-stdio-live-tool-events", "codex-stdio-noisy-reasoning-suppression", "codex-stdio-missing-turn-result", "codex-stdio-provider-auth-failed", "codex-stdio-provider-rate-limited", "codex-stdio-provider-invalid-tool-call", "codex-stdio-provider-503-rpc-error", "codex-stdio-provider-503-terminal", "codex-stdio-provider-503-retry-event", "codex-stdio-invalid-json", "codex-stdio-timeout", "codex-stdio-secret-unavailable", "codex-stdio-spawn-failure"] };
|
||||
return { name: "codex-stdio", tests: ["runner-lease-heartbeat", "codex-stdio-fake-turn", "codex-stdio-projected-writable-home", "codex-stdio-deepseek-profile-fake-turn", "codex-stdio-minimax-m3-profile-fake-turn", "codex-stdio-deepseek-missing-secret-no-fallback", "codex-stdio-minimax-m3-missing-secret-no-fallback", "codex-stdio-config-model-authoritative", "codex-stdio-explicit-model-forwarded", "codex-stdio-final-agent-message-only", "codex-stdio-stale-thread-resume-failed", "codex-stdio-live-tool-events", "codex-stdio-noisy-reasoning-suppression", "codex-stdio-missing-turn-result", "codex-stdio-provider-auth-failed", "codex-stdio-provider-rate-limited", "codex-stdio-provider-invalid-tool-call", "codex-stdio-provider-503-rpc-error", "codex-stdio-provider-503-terminal", "codex-stdio-provider-503-retry-event", "codex-stdio-invalid-json", "codex-stdio-timeout", "codex-stdio-secret-unavailable", "codex-stdio-spawn-failure"] };
|
||||
} finally {
|
||||
await new Promise<void>((resolve) => server.server.close(() => resolve()));
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user