fix: 收敛 stale thread 和 tool-call 错误归因
This commit is contained in:
@@ -105,15 +105,18 @@ 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 start a replacement thread and complete the turn");
|
||||
assert.equal(staleThreadResult.terminalStatus, "failed", "stale thread resume must fail instead of creating a replacement thread");
|
||||
assert.equal(staleThreadResult.failureKind, "thread-resume-failed", "stale thread resume failure kind");
|
||||
const staleEnvelope = await client.get(`/api/v1/runs/${staleThread.runId}/commands/${staleThread.commandId}/result`) as JsonRecord;
|
||||
assert.equal(staleEnvelope.terminalStatus, "completed");
|
||||
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 }> };
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "backend_status" && eventPayload(event).phase === "thread/resume:stale-thread-fallback"), true, "stale resume fallback must be visible");
|
||||
assert.equal(staleEvents.items?.some((event) => event.type === "backend_status" && eventPayload(event).phase === "thread/start:after-stale-resume:completed"), true, "stale resume must start a replacement thread");
|
||||
const stalePhases = (staleEvents.items ?? []).filter((event) => event.type === "backend_status").map((event) => String(eventPayload(event).phase ?? ""));
|
||||
assert.equal(stalePhases.some((phase) => phase.startsWith("thread/start")), 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"), true, "stale resume failure must be visible as an error event");
|
||||
assertNoSecretLeak({ staleThreadResult, staleEnvelope, staleEvents });
|
||||
|
||||
const live = await createRunWithCommand(client, context, "hello live events", "selftest-live-tool-events", 15_000);
|
||||
@@ -173,6 +176,7 @@ const selfTest: SelfTestCase = async (context) => {
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "missing-turn-result", expectedStatus: "failed", expectedFailureKind: "backend-response-invalid" });
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-401-rpc-error", expectedStatus: "failed", expectedFailureKind: "provider-auth-failed" });
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-429-terminal", expectedStatus: "failed", expectedFailureKind: "provider-rate-limited" });
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-invalid-tool-call", expectedStatus: "failed", expectedFailureKind: "provider-invalid-tool-call" });
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-503-rpc-error", expectedStatus: "failed", expectedFailureKind: "provider-unavailable" });
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-503-terminal", expectedStatus: "failed", expectedFailureKind: "provider-unavailable" });
|
||||
await runFailureCase({ client, managerUrl: server.baseUrl, context, mode: "provider-503-retry-event", expectedStatus: "failed", expectedFailureKind: "provider-unavailable", expectRetryError: true });
|
||||
@@ -181,7 +185,7 @@ 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-fallback", "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-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-fails", "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()));
|
||||
}
|
||||
@@ -264,7 +268,7 @@ async function createStaleThreadRun(client: ManagerClient, context: SelfTestCont
|
||||
},
|
||||
traceSink: null,
|
||||
}) as { id: string };
|
||||
const command = await client.post(`/api/v1/runs/${run.id}/commands`, { type: "turn", payload: { prompt: "hello stale thread" }, idempotencyKey: "selftest-stale-thread-fallback" }) as { id: string };
|
||||
const command = await client.post(`/api/v1/runs/${run.id}/commands`, { type: "turn", payload: { prompt: "hello stale thread" }, idempotencyKey: "selftest-stale-thread-resume-fails" }) as { id: string };
|
||||
return { runId: run.id, commandId: command.id };
|
||||
}
|
||||
|
||||
|
||||
@@ -100,6 +100,21 @@ for await (const line of rl) {
|
||||
respond(message.id, { turn });
|
||||
continue;
|
||||
}
|
||||
if (mode === "provider-invalid-tool-call") {
|
||||
turnCounter += 1;
|
||||
const turn = {
|
||||
id: `turn_selftest_${turnCounter}`,
|
||||
status: "failed",
|
||||
error: {
|
||||
message: "invalid params, invalid function arguments json string, tool_call_id: call_function_selftest_2 (2013)",
|
||||
code: "invalid_prompt",
|
||||
},
|
||||
};
|
||||
notify("turn/started", { turn: { id: turn.id, status: "running" } });
|
||||
notify("turn/completed", { turn });
|
||||
respond(message.id, { turn });
|
||||
continue;
|
||||
}
|
||||
if (mode === "provider-503-retry-event") {
|
||||
turnCounter += 1;
|
||||
const turn = {
|
||||
|
||||
Reference in New Issue
Block a user