pjandro/OmniRoute
1
0
Fork 0
mirror of https://github.com/diegosouzapw/OmniRoute.git synced 2026-08-26 09:02:11 +03:00
Code Issues Packages Projects Releases Wiki Activity

feat(resilience): add hierarchical concurrency admission

Browse Source
This commit is contained in:
Ravi Tharuma
2026-08-24 05:05:34 +02:00
committed by Markus Hartung
parent 943b9aaa84
commit 2b9f25accb
15 changed files with 485 additions and 248 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
src/app/(dashboard)/dashboard/settings/components/ResilienceTab.tsx
View File

@@ -13,6 +13,7 @@ type RequestQueueSettings = {
requestsPerMinute: number;
minTimeBetweenRequestsMs: number;
concurrentRequests: number;
globalConcurrentRequests: number;
maxWaitMs: number;
};
@@ -240,13 +241,21 @@ function RequestQueueCard({
}
/>
<NumberField
label={t("resilienceConcurrentRequests")}
label={t("resilienceConnectionScopeConcurrentRequests")}
value={draft.concurrentRequests}
min={1}
onChange={(concurrentRequests) =>
setDraft((prev) => ({ ...prev, concurrentRequests }))
}
/>
<NumberField
label={t("resilienceGlobalConcurrentRequests")}
value={draft.globalConcurrentRequests}
min={0}
onChange={(globalConcurrentRequests) =>
setDraft((prev) => ({ ...prev, globalConcurrentRequests }))
}
/>
<NumberField
label={t("resilienceMaxQueueWait")}
value={draft.maxWaitMs}
@@ -278,11 +287,21 @@ function RequestQueueCard({
</div>
</div>
<div className="rounded-xl border border-border bg-bg-subtle p-4">
<div className="text-xs text-text-muted">{t("resilienceConcurrentRequests")}</div>
<div className="text-xs text-text-muted">
{t("resilienceConnectionScopeConcurrentRequests")}
</div>
<div className="mt-1 text-sm font-semibold text-text-main">
{value.concurrentRequests}
</div>
</div>
<div className="rounded-xl border border-border bg-bg-subtle p-4">
<div className="text-xs text-text-muted">
{t("resilienceGlobalConcurrentRequests")}
</div>
<div className="mt-1 text-sm font-semibold text-text-main">
{value.globalConcurrentRequests || t("statusDisabled")}
</div>
</div>
<div className="rounded-xl border border-border bg-bg-subtle p-4">
<div className="text-xs text-text-muted">{t("resilienceMaxQueueWait")}</div>
<div className="mt-1 text-sm font-semibold text-text-main">

2
src/i18n/messages/de.json
View File

@@ -7538,6 +7538,8 @@
"resilienceRequestsPerMinute": "Requests per minute",
"resilienceMinTimeBetweenRequests": "Minimum time between requests",
"resilienceConcurrentRequests": "Concurrent requests",
"resilienceConnectionScopeConcurrentRequests": "Gleichzeitige Anfragen pro Verbindung/Quota-Bereich",
"resilienceGlobalConcurrentRequests": "Globale gleichzeitige Upstream-Anfragen (0 = deaktiviert)",
"resilienceMaxQueueWaitTime": "Maximale Wartezeit in der Warteschlange",
"resilienceBaseCooldown": "Base cooldown",
"resilienceUseUpstreamRetryHints": "Use upstream retry hints",

2
src/i18n/messages/en.json
View File

@@ -7550,6 +7550,8 @@
"resilienceRequestsPerMinute": "Requests per minute",
"resilienceMinTimeBetweenRequests": "Minimum time between requests",
"resilienceConcurrentRequests": "Concurrent requests",
"resilienceConnectionScopeConcurrentRequests": "Concurrent requests per connection/quota scope",
"resilienceGlobalConcurrentRequests": "Global concurrent upstream requests (0 = disabled)",
"resilienceMaxQueueWaitTime": "Maximum queue wait time",
"resilienceBaseCooldown": "Base cooldown",
"resilienceUseUpstreamRetryHints": "Use upstream retry hints",

3
src/lib/resilience/settings.ts
View File

@@ -58,6 +58,7 @@ export const DEFAULT_RESILIENCE_SETTINGS: ResilienceSettings = {
requestsPerMinute: DEFAULT_API_LIMITS.requestsPerMinute,
minTimeBetweenRequestsMs: DEFAULT_API_LIMITS.minTimeBetweenRequests,
concurrentRequests: DEFAULT_API_LIMITS.concurrentRequests,
globalConcurrentRequests: 0,
maxWaitMs: DEFAULT_REQUEST_QUEUE_MAX_WAIT_MS,
maxQueueDepth: DEFAULT_REQUEST_QUEUE_MAX_DEPTH,
},
@@ -208,6 +209,8 @@ function buildLegacyFallback(settings: JsonRecord): ResilienceSettings {
DEFAULT_RESILIENCE_SETTINGS.requestQueue.concurrentRequests,
{ min: 1, max: 10_000 }
),
globalConcurrentRequests:
DEFAULT_RESILIENCE_SETTINGS.requestQueue.globalConcurrentRequests,
maxWaitMs: DEFAULT_RESILIENCE_SETTINGS.requestQueue.maxWaitMs,
maxQueueDepth: DEFAULT_RESILIENCE_SETTINGS.requestQueue.maxQueueDepth,
},

16
src/lib/resilience/settings/normalize.ts
View File

@@ -124,6 +124,11 @@ export function normalizeRequestQueueSettings(
min: 1,
max: 10_000,
});
const globalConcurrentRequests = toInteger(
record.globalConcurrentRequests,
fallback.globalConcurrentRequests,
{ min: 0, max: 100_000 }
);
const maxWaitMs = toInteger(record.maxWaitMs, fallback.maxWaitMs, {
min: 1,
max: 24 * 60 * 60 * 1000,
@@ -141,6 +146,7 @@ export function normalizeRequestQueueSettings(
requestsPerMinute,
minTimeBetweenRequestsMs,
concurrentRequests,
globalConcurrentRequests,
maxWaitMs,
maxQueueDepth,
};
@@ -431,8 +437,16 @@ function normalizeProviderQuotaOverrideEntry(raw: unknown): ProviderQuotaOverrid
const out: ProviderQuotaOverrideSettings = {};
const rpm = typeof record.rpm === "number" ? record.rpm : Number(record.rpm);
if (Number.isFinite(rpm) && rpm > 0) out.rpm = Math.trunc(rpm);
const concurrency = typeof record.concurrency === "number" ? record.concurrency : Number(record.concurrency);
const concurrency =
typeof record.concurrency === "number" ? record.concurrency : Number(record.concurrency);
if (Number.isFinite(concurrency) && concurrency > 0) out.concurrency = Math.trunc(concurrency);
const providerConcurrency =
typeof record.providerConcurrency === "number"
? record.providerConcurrency
: Number(record.providerConcurrency);
if (Number.isFinite(providerConcurrency) && providerConcurrency >= 0) {
out.providerConcurrency = Math.trunc(providerConcurrency);
}
return Object.keys(out).length > 0 ? out : null;
}

4
src/lib/resilience/settings/types.ts
View File

@@ -16,6 +16,8 @@ export interface RequestQueueSettings {
requestsPerMinute: number;
minTimeBetweenRequestsMs: number;
concurrentRequests: number;
/** Whole-process upstream concurrency cap. Zero disables the global gate. */
globalConcurrentRequests: number;
/**
* Legacy persisted key used as Bottleneck's post-dispatch execution
* expiration. It does not bound time spent in Bottleneck's QUEUED state.
@@ -169,6 +171,8 @@ export interface ProviderQuotaOverrideSettings {
rpm?: number;
/** Overrides the static per-connection concurrency cap. */
concurrency?: number;
/** Shared concurrency cap across every connection for this provider. */
providerConcurrency?: number;
}
export interface StreamRecoverySettings {

2
src/shared/validation/schemas/settings.ts
View File

@@ -32,6 +32,7 @@ export const legacyResilienceDefaultsSchema = z
requestsPerMinute: z.number().int().min(1).optional(),
minTimeBetweenRequests: z.number().int().min(0).optional(),
concurrentRequests: z.number().int().min(1).optional(),
globalConcurrentRequests: z.number().int().min(0).max(100_000).optional(),
})
.strict();
@@ -170,6 +171,7 @@ export const updateResilienceSchema = z
.object({
rpm: z.number().int().min(1).optional(),
concurrency: z.number().int().min(1).optional(),
providerConcurrency: z.number().int().min(0).max(100_000).optional(),
})
.strict()
)