fix(app): back off and offer retry when timeline sync is refused

A host can refuse a timeline read for as long as it likes. Codex does
this whenever another writer holds the thread: every catch-up fails with
"already has an active writer" until the other session lets go. The app
retried that on a fixed 1s timer with no ceiling, so an idle window sat
there issuing a request and logging a warning every second, indefinitely.
The refusal is normal lifecycle; the traffic was the bug.

Catch-up and subscription retries now double from 1s to a 30s ceiling,
resetting on success, reconnect, delivery-mode change, and visibility
change so recovery stays immediate once the condition clears.

Backing off alone would leave the user waiting up to 30s with no way in,
so both failure surfaces get a retry they can press. Background retries
stay silent; a retry the user asked for is a fallible user action and
owns its pending state, so "retrying" is a status the sync model
publishes rather than a React boolean.

The blocking "Failed to load agent" screen was the worse half: its load
effect self-gates on the error state, so a refused first resume was
terminal until the user navigated away. Its retry resets the state to
idle, which re-arms the effect.
This commit is contained in:
Mohamed Boudra
2026-08-18 14:36:27 +02:00
parent 05ee7a1d31
commit 0f2e58d4bb
17 changed files with 461 additions and 51 deletions
@@ -79,10 +79,11 @@ class TimelineWorld {
schedule: (task, delayMs) => {
const scheduled = { task, delayMs };
this.scheduled.push(scheduled);
const waiter = this.retryWaiters.shift();
if (waiter && delayMs === 1_000) {
const waiterIndex = this.retryWaiters.findIndex((waiter) => waiter.delayMs === delayMs);
if (waiterIndex >= 0) {
const [waiter] = this.retryWaiters.splice(waiterIndex, 1);
this.scheduled.splice(this.scheduled.indexOf(scheduled), 1);
waiter(task);
waiter.resolve(task);
}
return () => {
const index = this.scheduled.indexOf(scheduled);
@@ -100,7 +101,10 @@ class TimelineWorld {
}> = [];
private readonly errorWaiters: Array<(message: string) => void> = [];
private readonly scheduled: Array<{ task: () => void; delayMs: number }> = [];
private readonly retryWaiters: Array<(retry: () => void) => void> = [];
private readonly retryWaiters: Array<{
delayMs: number;
resolve(retry: () => void): void;
}> = [];
private fetchTimeline(
agentId: string,
@@ -147,10 +151,10 @@ class TimelineWorld {
return new Promise((resolve) => this.errorWaiters.push(resolve));
}
nextRetry(): Promise<() => void> {
const index = this.scheduled.findIndex((entry) => entry.delayMs === 1_000);
nextRetry(delayMs = 1_000): Promise<() => void> {
const index = this.scheduled.findIndex((entry) => entry.delayMs === delayMs);
if (index >= 0) return Promise.resolve(this.scheduled.splice(index, 1)[0].task);
return new Promise((resolve) => this.retryWaiters.push(resolve));
return new Promise((resolve) => this.retryWaiters.push({ delayMs, resolve }));
}
elapse(elapsedMs: number): void {
@@ -411,6 +415,74 @@ test("a failed catch-up reports once and retries through the explicit retry poli
world.expectNoPendingMembership();
});
test("a failed catch-up retries with exponential backoff", async () => {
const world = new TimelineWorld();
world.sync.setConnected(true);
world.sync.replaceVisibleAgentIds("workspace", ["agent-a"]);
const membership = await world.nextMembership();
membership.succeed();
const first = await world.nextFetch("agent-a");
first.fail("timeline unavailable");
const [firstError, retryAfterFirstFailure] = await Promise.all([
world.nextError(),
world.nextRetry(),
]);
expect(firstError).toBe("timeline unavailable");
retryAfterFirstFailure();
const second = await world.nextFetch("agent-a");
second.fail("timeline unavailable");
await world.nextError();
const retryAfterSecondFailure = await world.nextRetry(2_000);
expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("error");
retryAfterSecondFailure();
const third = await world.nextFetch("agent-a");
third.respond({ hasNewer: false });
await vi.waitFor(() => expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("ready"));
});
test("manual retries can immediately re-attempt a failed catch-up", async () => {
const world = new TimelineWorld();
world.sync.setConnected(true);
world.sync.replaceVisibleAgentIds("workspace", ["agent-a"]);
const membership = await world.nextMembership();
membership.succeed();
const failed = await world.nextFetch("agent-a");
failed.fail("timeline unavailable");
await world.nextRetry();
expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("error");
world.sync.retryVisibleAgentTimeline("agent-a");
expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("retrying");
const retry = await world.nextFetch("agent-a");
retry.respond({ hasNewer: false });
await vi.waitFor(() => expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("ready"));
});
test("a manual retry that fails returns to the error state", async () => {
const world = new TimelineWorld();
world.sync.setConnected(true);
world.sync.replaceVisibleAgentIds("workspace", ["agent-a"]);
const membership = await world.nextMembership();
membership.succeed();
const failed = await world.nextFetch("agent-a");
failed.fail("timeline unavailable");
await world.nextRetry();
world.sync.retryVisibleAgentTimeline("agent-a");
expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("retrying");
const retry = await world.nextFetch("agent-a");
retry.fail("timeline still unavailable");
await vi.waitFor(() => expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("error"));
});
test("gap recovery supersedes completed catch-up and pages through the current tail", async () => {
const world = new TimelineWorld();
world.sync.setConnected(true);
@@ -486,6 +558,34 @@ test("membership failure autonomously retries without another visibility declara
});
});
test("membership failures retry with exponential backoff", async () => {
const world = new TimelineWorld();
world.sync.setConnected(true);
world.sync.replaceVisibleAgentIds("workspace", ["agent-a"]);
const first = await world.nextMembership();
first.fail("subscription unavailable");
const [firstError, retryAfterFirstFailure] = await Promise.all([
world.nextError(),
world.nextRetry(),
]);
expect(firstError).toBe("subscription unavailable");
retryAfterFirstFailure();
const second = await world.nextMembership();
second.fail("subscription unavailable again");
const retryAfterSecondFailure = await world.nextRetry(2_000);
expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("error");
retryAfterSecondFailure();
const retry = await world.nextMembership();
retry.succeed();
const catchUp = await world.nextFetch("agent-a");
catchUp.respond({ hasNewer: false });
await vi.waitFor(() => expect(world.sync.getAgentTimelineStatus("agent-a")).toBe("ready"));
});
test("backgrounding preserves the hot membership without catch-up on return", async () => {
const world = new TimelineWorld();
world.sync.setConnected(true);
@@ -23,12 +23,13 @@ interface ViewedTimelineSyncPorts {
}
export type TimelineDeliveryMode = "legacy" | "selective";
export type ViewedTimelineStatus = "ready" | "pending" | "error";
export type ViewedTimelineStatus = "ready" | "pending" | "error" | "retrying";
export interface ViewedTimelineUiBridge {
replaceVisibleAgentIds(sourceId: string, agentIds: string[]): void;
subscribe(listener: () => void): () => void;
getAgentTimelineStatus(agentId: string): ViewedTimelineStatus;
retryVisibleAgentTimeline(agentId: string): void;
}
export interface ViewedTimelineSync extends ViewedTimelineUiBridge {
@@ -40,6 +41,7 @@ export interface ViewedTimelineSync extends ViewedTimelineUiBridge {
}
const RETRY_DELAY_MS = 1_000;
const MAX_RETRY_DELAY_MS = 30_000;
const VIEWED_TIMELINE_HOT_AGENT_LIMIT = 5;
type CatchUpStatus = "running" | "complete" | "error";
@@ -49,8 +51,16 @@ interface CatchUpState {
status: CatchUpStatus;
request?: ProjectedTimelineForwardFetchPlan;
cancelRetry?: () => void;
retryDelayMs?: number;
}
const getNextRetryDelayMs = (previousDelayMs: number | undefined): number => {
if (previousDelayMs == null) {
return RETRY_DELAY_MS;
}
return Math.min(previousDelayMs * 2, MAX_RETRY_DELAY_MS);
};
function isSameCatchUpRequest(
left: ProjectedTimelineForwardFetchPlan | undefined,
right: ProjectedTimelineForwardFetchPlan | undefined,
@@ -102,6 +112,9 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
const pendingCatchUps = new Map<string, ProjectedTimelineForwardFetchPlan>();
const visibilityCatchUpPending = new Set<string>();
const visibilityCatchUpErrors = new Set<string>();
// User-initiated retries only. Background retries stay silent; a retry the user asked for
// owes them a pending state until it settles.
const manualRetries = new Set<string>();
const listeners = new Set<() => void>();
let active = true;
let connected = false;
@@ -113,6 +126,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
let reconciling = false;
let reconcileRequested = false;
let membershipNeedsRetry = false;
let membershipRetryDelayMs: number | undefined;
let cancelMembershipRetry: (() => void) | null = null;
let recentlyViewedAgentIds: string[] = [];
@@ -152,12 +166,14 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
const setVisibilityCatchUpReady = (agentId: string) => {
const wasPending = visibilityCatchUpPending.delete(agentId);
const hadError = visibilityCatchUpErrors.delete(agentId);
if (wasPending || hadError) notifyListeners();
const wasRetrying = manualRetries.delete(agentId);
if (wasPending || hadError || wasRetrying) notifyListeners();
};
const setVisibilityCatchUpError = (agentIds: string[]) => {
let changed = false;
for (const agentId of agentIds) {
if (manualRetries.delete(agentId)) changed = true;
if (!visibilityCatchUpPending.delete(agentId)) continue;
visibilityCatchUpErrors.add(agentId);
changed = true;
@@ -210,12 +226,19 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
setVisibilityCatchUpReady(agentId);
} catch (error) {
if (catchUps.get(agentId)?.generation === generation) {
const nextRetryDelayMs = getNextRetryDelayMs(catchUps.get(agentId)?.retryDelayMs);
const cancelRetry = ports.schedule(() => {
const current = catchUps.get(agentId);
if (current?.generation !== generation || current.status !== "error") return;
startCatchUp(agentId);
}, RETRY_DELAY_MS);
catchUps.set(agentId, { generation, status: "error", cancelRetry });
}, nextRetryDelayMs);
catchUps.set(agentId, {
generation,
status: "error",
request,
cancelRetry,
retryDelayMs: nextRetryDelayMs,
});
setVisibilityCatchUpError([agentId]);
ports.reportError(error);
}
@@ -247,7 +270,14 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
current?.cancelRetry?.();
const generation = (catchUpGenerations.get(agentId) ?? 0) + 1;
catchUpGenerations.set(agentId, generation);
catchUps.set(agentId, { generation, status: "running", request: nextRequest });
const retryDelayMs =
supersede || current?.status !== "error" ? undefined : current.retryDelayMs;
catchUps.set(agentId, {
generation,
status: "running",
request: nextRequest,
retryDelayMs,
});
pendingCatchUps.delete(agentId);
void fetchUntilCurrent(
agentId,
@@ -279,6 +309,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
membershipNeedsRetry = true;
setVisibilityCatchUpError(requested);
cancelMembershipRetry?.();
const nextRetryDelayMs = getNextRetryDelayMs(membershipRetryDelayMs);
cancelMembershipRetry = ports.schedule(() => {
cancelMembershipRetry = null;
if (
@@ -290,12 +321,14 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
return;
}
void reconcileMembership();
}, RETRY_DELAY_MS);
}, nextRetryDelayMs);
membershipRetryDelayMs = nextRetryDelayMs;
ports.reportError(error);
return;
}
cancelMembershipRetry?.();
cancelMembershipRetry = null;
membershipRetryDelayMs = undefined;
if (disposed || !connected || deliveryMode !== "selective") return;
acknowledged = requested;
if (generation !== membershipGeneration) {
@@ -338,6 +371,25 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
}
};
const retryVisibleAgentTimeline = (agentId: string) => {
if (!isDesired(agentId) || manualRetries.has(agentId)) return;
const catchUp = catchUps.get(agentId);
const membershipRetryable = deliveryMode === "selective" && membershipNeedsRetry && connected;
if (catchUp?.status !== "error" && !membershipRetryable) return;
manualRetries.add(agentId);
notifyListeners();
if (catchUp?.status === "error") {
catchUp.cancelRetry?.();
startCatchUp(agentId, { request: catchUp.request, supersede: true });
return;
}
cancelMembershipRetry?.();
cancelMembershipRetry = null;
membershipRetryDelayMs = undefined;
membershipNeedsRetry = false;
void reconcileMembership();
};
const commitDesiredMembership = (
nextDesired: string[],
options: { resetCatchUpStatus?: boolean } = {},
@@ -350,6 +402,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
statusChanged = true;
}
if (visibilityCatchUpErrors.delete(agentId)) statusChanged = true;
if (manualRetries.delete(agentId)) statusChanged = true;
}
}
if (sameAgentIds(nextDesired, desired)) {
@@ -364,12 +417,14 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
cancelCatchUp(agentId);
visibilityCatchUpPending.delete(agentId);
visibilityCatchUpErrors.delete(agentId);
manualRetries.delete(agentId);
}
}
for (const agentId of nextDesired) {
if (!desired.includes(agentId)) {
visibilityCatchUpPending.add(agentId);
visibilityCatchUpErrors.delete(agentId);
manualRetries.delete(agentId);
}
}
cancelMembershipRetry?.();
@@ -403,6 +458,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
return () => listeners.delete(listener);
},
getAgentTimelineStatus(agentId) {
if (manualRetries.has(agentId)) return "retrying";
if (visibilityCatchUpErrors.has(agentId)) return "error";
if (!isDesired(agentId) || visibilityCatchUpPending.has(agentId)) return "pending";
return "ready";
@@ -427,6 +483,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
commitDesiredMembership(visible, { resetCatchUpStatus: true });
cancelMembershipRetry?.();
cancelMembershipRetry = null;
membershipRetryDelayMs = undefined;
acknowledged = [];
membershipGeneration += 1;
for (const agentId of desired) cancelCatchUp(agentId);
@@ -445,6 +502,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
deliveryMode = nextMode;
cancelMembershipRetry?.();
cancelMembershipRetry = null;
membershipRetryDelayMs = undefined;
membershipNeedsRetry = false;
membershipGeneration += 1;
for (const agentId of desired) cancelCatchUp(agentId);
@@ -453,6 +511,7 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
desired = visible;
visibilityCatchUpPending.clear();
visibilityCatchUpErrors.clear();
manualRetries.clear();
for (const agentId of desired) visibilityCatchUpPending.add(agentId);
acknowledged = deliveryMode === "legacy" && connected ? desired : [];
notifyListeners();
@@ -470,6 +529,8 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
disposed = true;
cancelMembershipRetry?.();
cancelMembershipRetry = null;
membershipNeedsRetry = false;
membershipRetryDelayMs = undefined;
sources.clear();
membershipGeneration += 1;
for (const agentId of desired) cancelCatchUp(agentId);
@@ -478,8 +539,10 @@ export function createViewedTimelineSync(ports: ViewedTimelineSyncPorts): Viewed
recentlyViewedAgentIds = [];
visibilityCatchUpPending.clear();
visibilityCatchUpErrors.clear();
manualRetries.clear();
notifyListeners();
listeners.clear();
},
retryVisibleAgentTimeline,
};
}