mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-09 16:02:36 +01:00
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - The restart chaos harness and the runner-e2e recover-controller cells stop the control plane with `replaceController` (`tests/runner-e2e/controller-outage.ts`), hold it down, then start it again > - `parseControllerOutage` refuses holds longer than 300000 ms > - Restart chaos needs longer holds. One example is proving that a Daytona sandbox survives an outage longer than its old 5-minute auto-stop but inside the 7-minute runner reconnect grace (RC-54) > - With the cap, such a fault errors before the controller stops, so the scenario cannot run > - This pull request raises the test-harness cap to 600000 ms > - The benefit is that chaos runs can now inject 5–10 minute outages; production code and qualification contracts are unchanged ## Linked Issues or Issue Description **Bug description** A restart chaos replay with a 360000 ms K4 hold failed with "Controller outage requires SIGTERM/SIGKILL and 0..300000 integer milliseconds". The controller was never stopped. **Expected behavior** The test harness accepts outages of up to 10 minutes. The recover-controller receipt check keeps its own readiness limit (`readyAfterMs < 420000`). Refs #19544 ## What Changed - `tests/runner-e2e/controller-outage.ts`: adds `MAX_CONTROLLER_OUTAGE_MS = 600_000` and uses it in `parseControllerOutage` and its error message. `validControllerOutageReceipt` is unchanged and still requires readiness inside the 7-minute grace. - `tests/runner-e2e/controller-outage.test.ts`: the rejection cases now use 600001. A new test accepts 360000 (SIGKILL) and 600000 (SIGTERM). The existing five-minute contract test is unchanged. ## Verification - `vitest run tests/runner-e2e/controller-outage.test.ts tests/runner-e2e/server-restart-request.test.ts`: 52/52 pass. - `server-restart-request.test.ts` now uses 600001 as its out-of-range `controllerOutage`. The separate restart-request `controllerDowntimeMs` limit (five minutes) is unchanged. ## Risks - This change is test-only. No product or server code changes. - A suite that sets `PAPERCLIP_RUNNER_E2E_CONTROLLER_OUTAGE_MS` above 300000 now runs instead of being refused. The recover-controller receipt check still fails any receipt whose readiness is outside the grace. > For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and discuss it in `#dev` before opening the PR. Feature PRs that overlap with planned core work may need to be redirected — check the roadmap first. See `CONTRIBUTING.md`. ## Model Used - Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, in Claude Code with tool use (shell, file edits, GitHub CLI). Default context window, extended thinking. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] All Paperclip CI gates are green - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
116 lines
7.2 KiB
TypeScript
116 lines
7.2 KiB
TypeScript
import { mkdtemp, rm, writeFile } from "node:fs/promises";
|
|
import { tmpdir } from "node:os";
|
|
import { join } from "node:path";
|
|
import { afterEach, beforeEach, expect, it, vi } from "vitest";
|
|
import { runnerMatrix } from "./catalog.js";
|
|
import { controllerOutageForExecution } from "./controller-outage.js";
|
|
import { readRestartRequest, kimiCandidateAdmissionForExecution } from "./server-restart-request.js";
|
|
|
|
let root: string;
|
|
let requestPath: string;
|
|
const profilesModule = "../../packages/paperclip-runner/src/drivers/acpx/qualified-profiles.js";
|
|
beforeEach(async () => {
|
|
root = await mkdtemp(join(tmpdir(), "paperclip-supervisor-request-"));
|
|
requestPath = join(root, "restart.json");
|
|
});
|
|
afterEach(async () => {
|
|
vi.doUnmock(profilesModule);
|
|
vi.resetModules();
|
|
await rm(root, { recursive: true, force: true });
|
|
});
|
|
|
|
it.each(["rolling-web-deployment", "rolling-web-deployment-warm"])("accepts the assembled %s restart request", async suite => {
|
|
const execution = runnerMatrix.find(row => row.suite.id === suite && row.profile.id === "runner-codex")!;
|
|
const recoveryBinding = { companyId: "11111111-1111-4111-8111-111111111111",
|
|
issueId: "22222222-2222-4222-8222-222222222222", runId: "33333333-3333-4333-8333-333333333333" };
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "rolling-request", rollingDeployment: true,
|
|
discardControllerCache: true, controllerOutage: controllerOutageForExecution(execution, {}), recoveryBinding }));
|
|
const request = await readRestartRequest(requestPath, [execution.id]);
|
|
expect(request).toMatchObject({ rollingDeployment: true, discardControllerCache: true, recoveryBinding });
|
|
expect(request).not.toHaveProperty("controllerOutage");
|
|
});
|
|
|
|
const cells = [
|
|
{ suite: "everyday-workflows", provider: "codex" },
|
|
{ suite: "everyday-workflows", provider: "acpx" },
|
|
{ suite: "daytona-restart-continuity", provider: "codex" },
|
|
].map(({ suite, provider }) => {
|
|
const execution = runnerMatrix.find(row => row.suite.id === suite && row.task.id === "recover-controller"
|
|
&& row.environment.id === "daytona" && row.profile.generation === "native" && row.profile.provider === provider);
|
|
if (!execution) throw new Error(`Missing real ${suite}/${provider} continuity cell`);
|
|
return execution;
|
|
});
|
|
|
|
it.each(cells)("accepts the real supervisor request file for $id", async execution => {
|
|
const outage = controllerOutageForExecution(execution, {})!;
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "actual-restart-request",
|
|
discardControllerCache: true, controllerOutage: outage }));
|
|
await expect(readRestartRequest(requestPath, [execution.id])).resolves.toMatchObject({
|
|
requestId: "actual-restart-request", discardControllerCache: true, controllerOutage: outage,
|
|
});
|
|
});
|
|
|
|
it("rejects cache replacement for unselected, local and legacy cells", async () => {
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "unsupported-restart", discardControllerCache: true }));
|
|
const local = runnerMatrix.find(row => row.environment.id === "local")!;
|
|
const legacy = runnerMatrix.find(row => row.profile.generation === "legacy")!;
|
|
expect(local).toBeDefined(); expect(legacy).toBeDefined();
|
|
for (const ids of [[], ["not-a-cell"], [local.id], [legacy.id]]) {
|
|
await expect(readRestartRequest(requestPath, ids)).rejects.toThrow(/cache replacement requires/i);
|
|
}
|
|
});
|
|
|
|
it.each([
|
|
{ controllerOutage: { signal: "SIGKILL", holdMs: 125_000 }, discardControllerCache: false },
|
|
{ controllerOutage: { signal: "SIGKILL", holdMs: 600_001 }, discardControllerCache: true },
|
|
{ controllerOutage: { signal: "SIGKILL", holdMs: 125_000 }, discardControllerCache: true, preserveRunner: {} },
|
|
{ controllerOutage: { signal: "SIGKILL", holdMs: 125_000 }, discardControllerCache: true, crashAfterResult: {} },
|
|
{ recoveryBinding: {} },
|
|
{ controllerDowntimeMs: -1 },
|
|
])("keeps invalid and conflicting fault boundaries rejected: %j", async fault => {
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "conflicting-restart", ...fault }));
|
|
await expect(readRestartRequest(requestPath, [cells[0].id])).rejects.toThrow();
|
|
});
|
|
|
|
it("ignores absent and incomplete atomic request files", async () => {
|
|
await expect(readRestartRequest(requestPath, [cells[0].id])).resolves.toBeNull();
|
|
await writeFile(requestPath, "{");
|
|
await expect(readRestartRequest(requestPath, [cells[0].id])).resolves.toBeNull();
|
|
});
|
|
|
|
it.each(["local", "daytona"])("uses ordinary released Kimi restart without candidate authority on %s", async environment => {
|
|
const execution = runnerMatrix.find(row => row.suite.id === "kimi-qualification" && row.environment.id === environment)!;
|
|
const admission = kimiCandidateAdmissionForExecution(execution, "11111111-1111-4111-8111-111111111111");
|
|
expect(admission).toBeUndefined();
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "kimi-released", kimiCandidateAdmission: admission }));
|
|
const request = await readRestartRequest(requestPath, [execution.id]);
|
|
expect(request).toMatchObject({ requestId: "kimi-released" });
|
|
expect(request).not.toHaveProperty("kimiCandidateAdmission");
|
|
});
|
|
|
|
it.each(["local", "daytona"])("stages only exact pending Kimi company/model/profile/target via owned restart on %s", async environment => {
|
|
const actual = await vi.importActual<typeof import("../../packages/paperclip-runner/src/drivers/acpx/qualified-profiles.js")>(profilesModule);
|
|
vi.doMock(profilesModule, () => ({ ...actual, QUALIFIED_ACPX_PROFILES: { ...actual.QUALIFIED_ACPX_PROFILES,
|
|
kimi: { ...actual.QUALIFIED_ACPX_PROFILES.kimi, qualificationStatus: "pending" } } }));
|
|
vi.resetModules();
|
|
const { readRestartRequest, kimiCandidateAdmissionForExecution } = await import("./server-restart-request.js");
|
|
const execution = runnerMatrix.find(row => row.suite.id === "kimi-qualification" && row.environment.id === environment)!;
|
|
const admission = kimiCandidateAdmissionForExecution(execution, "11111111-1111-4111-8111-111111111111")!;
|
|
expect(admission).toBeDefined();
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "kimi-startup", kimiCandidateAdmission: admission }));
|
|
await expect(readRestartRequest(requestPath, [execution.id])).resolves.toMatchObject({ requestId: "kimi-startup", kimiCandidateAdmission: admission });
|
|
expect(admission).not.toHaveProperty("runId");
|
|
for (const change of [{ companyId: "foreign" }, { agent: "antigravity" }, { model: "different" },
|
|
{ profileDigest: `sha256:${"0".repeat(64)}` }, { nativeClosureDigest: `sha256:${"0".repeat(64)}` },
|
|
{ expiresAt: 0 }, { runId: "injected-run" }, { architecture: "arm64", platform: "linux" }]) {
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "wrong-kimi", kimiCandidateAdmission: { ...admission, ...change } }));
|
|
await expect(readRestartRequest(requestPath, [execution.id])).rejects.toThrow();
|
|
}
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "unselected-kimi", kimiCandidateAdmission: admission }));
|
|
for (const ids of [[], [cells[0].id]]) await expect(readRestartRequest(requestPath, ids)).rejects.toThrow();
|
|
for (const change of [{ controllerDowntimeMs: 1 }, { preserveRunner: {} }, { discardControllerCache: true }]) {
|
|
await writeFile(requestPath, JSON.stringify({ requestId: "mixed-kimi", kimiCandidateAdmission: admission, ...change }));
|
|
await expect(readRestartRequest(requestPath, [execution.id])).rejects.toThrow();
|
|
}
|
|
});
|