ZacharyZhang-NY
927bdc91f2
The July 9 run died at 07:22 when the GPT account hit its plan-side usage limit: every later request failed with "The usage limit has been reached", each failure counted against the 3-attempt retry budget, and 31 tasks were abandoned for good while the worker kept hammering 12 doomed requests a minute until the queue drained. Root fix, three parts sharing one classifier: - isTransientProviderFailure() in provider-runtime-domain recognises provider-side verdicts (usage limit, rate limit, 429, overloaded) plus the internal monthly-quota message; matching is against recorded task_runs.error_message rows, so historical damage heals retroactively. - enqueueHeartbeatRuns computes the retry budget with that predicate instead of an exact match on the internal quota string alone. - A transient failure parks the provider for a 10-minute cooldown: the heartbeat stops enqueuing its tasks and the worker leaves its queued runs unclaimed, then a single probe run re-arms or clears the park. Transitions are logged once each way. Tests: predicate unit tests, budget-exemption regression on the raw provider message, cooldown park/resume with an injected clock, claim exclusion.
22 lines
1.3 KiB
TypeScript
22 lines
1.3 KiB
TypeScript
import { describe, expect, test } from "vitest";
|
|
import { isTransientProviderFailure, PROVIDER_MONTHLY_LIMIT_MESSAGE } from "../../src/shared/provider-runtime-domain";
|
|
|
|
describe("isTransientProviderFailure", () => {
|
|
test("classifies provider-side limit and overload verdicts as transient", () => {
|
|
expect(isTransientProviderFailure(PROVIDER_MONTHLY_LIMIT_MESSAGE)).toBe(true);
|
|
expect(isTransientProviderFailure("The usage limit has been reached")).toBe(true);
|
|
expect(isTransientProviderFailure("Our servers are currently overloaded. Please try again later.")).toBe(true);
|
|
expect(isTransientProviderFailure("Rate limit exceeded, retry after 20s")).toBe(true);
|
|
expect(isTransientProviderFailure("429 Too Many Requests")).toBe(true);
|
|
});
|
|
|
|
test("keeps genuine task failures counting toward the retry budget", () => {
|
|
expect(isTransientProviderFailure(null)).toBe(false);
|
|
expect(isTransientProviderFailure(undefined)).toBe(false);
|
|
expect(isTransientProviderFailure("")).toBe(false);
|
|
expect(isTransientProviderFailure("provider rejected task")).toBe(false);
|
|
expect(isTransientProviderFailure("Agent run did not converge within 12 iterations.")).toBe(false);
|
|
expect(isTransientProviderFailure("Provider returned an empty turn: no text and no tool calls.")).toBe(false);
|
|
});
|
|
});
|