Skip to content

fix(client-runtime): tolerate transient foreground probe timeouts #5198

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Zeus-Deus wants to merge 1 commit into
base: main
Choose a base branch
from
+115 −7
Open

fix(client-runtime): tolerate transient foreground probe timeouts #5198

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
83 changes: 83 additions & 0 deletions packages/client-runtime/src/connection/supervisor.test.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -949,6 +949,89 @@ describe("EnvironmentSupervisor", () => {
}).pipe(Effect.provide(TestClock.layer())),
);

it.effect("reconnects only after consecutive foreground liveness probe timeouts", () =>
Effect.gen(function* () {
const firstProbeStarted = yield* Deferred.make<void>();
const secondProbeStarted = yield* Deferred.make<void>();
const probeCount = yield* Ref.make(0);
const harness = yield* makeHarness({
probe: () =>
Ref.updateAndGet(probeCount, (count) => count + 1).pipe(
Effect.tap((count) =>
Deferred.succeed(count === 1 ? firstProbeStarted : secondProbeStarted, undefined),
),
Effect.andThen(Effect.never),
),
});
const supervisor = yield* EnvironmentSupervisor.make(TARGET_ENTRY, {
initiallyDesired: true,
}).pipe(Effect.provide(harness.dependencies));

yield* awaitState(supervisor.state, (state) => state.phase === "connected");
yield* harness.wake("application-active");
yield* Deferred.await(firstProbeStarted);
yield* TestClock.adjust("15 seconds");
yield* Effect.yieldNow;

expect((yield* SubscriptionRef.get(supervisor.state)).phase).toBe("connected");
expect(yield* Ref.get(harness.sessionCount)).toBe(1);
expect(yield* Ref.get(harness.releaseCount)).toBe(0);

yield* harness.wake("application-active");
yield* Deferred.await(secondProbeStarted);
yield* TestClock.adjust("15 seconds");
yield* awaitState(supervisor.state, (state) => state.phase === "backoff");

expect(yield* Ref.get(harness.sessionCount)).toBe(1);
expect(yield* Ref.get(harness.releaseCount)).toBe(1);
}).pipe(Effect.provide(TestClock.layer())),
);

it.effect("resets the foreground probe timeout count after a successful probe", () =>
Effect.gen(function* () {
const firstProbeStarted = yield* Deferred.make<void>();
const secondProbeStarted = yield* Deferred.make<void>();
const thirdProbeStarted = yield* Deferred.make<void>();
const probeCount = yield* Ref.make(0);
const harness = yield* makeHarness({
probe: () =>
Ref.updateAndGet(probeCount, (count) => count + 1).pipe(
Effect.tap((count) =>
count === 1
? Deferred.succeed(firstProbeStarted, undefined)
: count === 2
? Deferred.succeed(secondProbeStarted, undefined)
: count === 3
? Deferred.succeed(thirdProbeStarted, undefined)
: Effect.void,
),
Effect.flatMap((count) => (count === 2 ? Effect.void : Effect.never)),
),
});
const supervisor = yield* EnvironmentSupervisor.make(TARGET_ENTRY, {
initiallyDesired: true,
}).pipe(Effect.provide(harness.dependencies));

yield* awaitState(supervisor.state, (state) => state.phase === "connected");
yield* harness.wake("application-active");
yield* Deferred.await(firstProbeStarted);
yield* TestClock.adjust("15 seconds");

yield* harness.wake("application-active");
yield* Deferred.await(secondProbeStarted);
yield* Effect.yieldNow;

yield* harness.wake("application-active");
yield* Deferred.await(thirdProbeStarted);
yield* TestClock.adjust("15 seconds");
yield* Effect.yieldNow;

expect((yield* SubscriptionRef.get(supervisor.state)).phase).toBe("connected");
expect(yield* Ref.get(harness.sessionCount)).toBe(1);
expect(yield* Ref.get(harness.releaseCount)).toBe(0);
}).pipe(Effect.provide(TestClock.layer())),
);

it.effect("quickly times out a stalled mobile foreground liveness probe", () =>
Effect.gen(function* () {
const harness = yield* makeHarness({
Expand Down
39 changes: 32 additions & 7 deletions packages/client-runtime/src/connection/supervisor.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,7 @@ const RETRY_DELAYS_MS = [1_000, 2_000, 4_000, 8_000, 16_000] as const;
const CONNECTION_ESTABLISHMENT_TIMEOUT = "15 seconds";
const CONNECTION_PROBE_TIMEOUT = "15 seconds";
const MOBILE_CONNECTION_PROBE_TIMEOUT = "3 seconds";
const FOREGROUND_PROBE_TIMEOUTS_BEFORE_RECONNECT = 2;
const BACKOFF_RESET_AFTER_MS = 30_000;

interface SupervisorIntent {
Expand Down Expand Up @@ -392,6 +393,7 @@ export const make = Effect.fn("EnvironmentSupervisor.make")(function* (
const monitorConnectedLease = Effect.fnUntraced(function* (
lease: ConnectionDriver.EnvironmentConnectionLease,
) {
let foregroundProbeTimeoutCount = 0;
for (;;) {
const next = yield* Queue.take(signals);
switch (next._tag) {
Expand All @@ -415,19 +417,23 @@ export const make = Effect.fn("EnvironmentSupervisor.make")(function* (
return true;
}
if (next.reason === "application-active" || next.reason === "application-active-probe") {
const tolerateProbeTimeout = next.reason === "application-active";
const probe = yield* lease.session.probe.pipe(
Effect.as(true),
Effect.timeoutOrElse({
duration:
next.reason === "application-active-probe"
? MOBILE_CONNECTION_PROBE_TIMEOUT
: CONNECTION_PROBE_TIMEOUT,
orElse: () =>
Effect.fail(
new ConnectionTransientError({
reason: "timeout",
detail: `${target.label} did not respond to a connection health check.`,
}),
),
tolerateProbeTimeout
? Effect.succeed(false)
: Effect.fail(
new ConnectionTransientError({
reason: "timeout",
detail: `${target.label} did not respond to a connection health check.`,
}),
),
}),
Effect.forkChild,
);
Expand All @@ -441,7 +447,26 @@ export const make = Effect.fn("EnvironmentSupervisor.make")(function* (
),
);
if (probeEvent._tag === "ProbeCompleted") {
yield* probeEvent.exit;
const responded = yield* probeEvent.exit;
if (responded) {
foregroundProbeTimeoutCount = 0;
break;
}
foregroundProbeTimeoutCount += 1;
if (foregroundProbeTimeoutCount >= FOREGROUND_PROBE_TIMEOUTS_BEFORE_RECONNECT) {
return yield* new ConnectionTransientError({
reason: "timeout",
detail: `${target.label} did not respond to a connection health check.`,
});
}
yield* Effect.logWarning(
"Foreground connection health check timed out; keeping the existing session.",
).pipe(
Effect.annotateLogs({
"environment.id": target.environmentId,
"environment.label": target.label,
}),
);
break;
}
switch (probeEvent.signal._tag) {
Expand Down
Loading