Eric Allam
d343727021
(`OBJECT_STORE_BASE_URL`) and a named protocol provider
(`OBJECT_STORE_DEFAULT_PROTOCOL=s3`), chat.agent session snapshot writes
landed in the named provider but reads fell through to the default — so
the recovery boot couldn't find the snapshot it had just written.
After a mid-stream cancel, the missing snapshot triggered a fallback
replay path that dropped the user's follow-up message, leaving the chat
stuck in `submitted` indefinitely.
Fix:
- New `/api/v1/sessions/:id/snapshot-url` route handles PUT + GET
symmetrically — both prefix unprefixed keys with
`OBJECT_STORE_DEFAULT_PROTOCOL` so they always round-trip through the
same store.
- `Session.chatSnapshotStoragePath` persists the resolved URI on first
write so future protocol changes don't strand existing snapshots.
Reads prefer the stored URI and fall back to the computed default for
pre-column sessions.
- SDK calls `createChatSnapshotUploadUrl` / `getChatSnapshotUrl`; the
generic v1/v2 packets endpoints are unchanged.
## Test plan
- [x] Configure local with two providers (R2 default + MinIO `s3` named)
and `OBJECT_STORE_DEFAULT_PROTOCOL=s3`.
- [x] Reproduce hang: send a message, cancel mid-stream, send another —
without the fix it hangs in `submitted`; with the fix it streams.
- [x] Snapshot lands in the `s3`-protocol bucket and
`Session.chatSnapshotStoragePath` is set after first write.
- [x] SDK unit tests pass; webapp typecheck passes.
83 lines
3.4 KiB
TypeScript
83 lines
3.4 KiB
TypeScript
import { env } from "~/env.server";
|
|
import { authenticateAuthorizationHeader } from "./apiAuth.server";
|
|
import { authorizationRateLimitMiddleware } from "./authorizationRateLimitMiddleware.server";
|
|
import { Duration } from "./rateLimiter.server";
|
|
|
|
export const apiRateLimiter = authorizationRateLimitMiddleware({
|
|
redis: {
|
|
port: env.RATE_LIMIT_REDIS_PORT,
|
|
host: env.RATE_LIMIT_REDIS_HOST,
|
|
username: env.RATE_LIMIT_REDIS_USERNAME,
|
|
password: env.RATE_LIMIT_REDIS_PASSWORD,
|
|
tlsDisabled: env.RATE_LIMIT_REDIS_TLS_DISABLED === "true",
|
|
clusterMode: env.RATE_LIMIT_REDIS_CLUSTER_MODE_ENABLED === "1",
|
|
},
|
|
keyPrefix: "api",
|
|
defaultLimiter: {
|
|
type: "tokenBucket",
|
|
refillRate: env.API_RATE_LIMIT_REFILL_RATE,
|
|
interval: env.API_RATE_LIMIT_REFILL_INTERVAL as Duration,
|
|
maxTokens: env.API_RATE_LIMIT_MAX,
|
|
},
|
|
limiterCache: {
|
|
fresh: 60_000 * 10, // Data is fresh for 10 minutes
|
|
stale: 60_000 * 20, // Date is stale after 20 minutes
|
|
maxItems: 1000,
|
|
},
|
|
limiterConfigOverride: async (authorizationValue) => {
|
|
const authenticatedEnv = await authenticateAuthorizationHeader(authorizationValue, {
|
|
allowPublicKey: true,
|
|
allowJWT: true,
|
|
});
|
|
|
|
if (!authenticatedEnv || !authenticatedEnv.ok) {
|
|
return;
|
|
}
|
|
|
|
if (authenticatedEnv.type === "PUBLIC_JWT") {
|
|
return {
|
|
type: "fixedWindow",
|
|
window: env.API_RATE_LIMIT_JWT_WINDOW,
|
|
tokens: env.API_RATE_LIMIT_JWT_TOKENS,
|
|
};
|
|
} else {
|
|
return authenticatedEnv.environment.organization.apiRateLimiterConfig;
|
|
}
|
|
},
|
|
pathMatchers: [/^\/api/],
|
|
// Allow /api/v1/tasks/:id/callback/:secret
|
|
pathWhiteList: [
|
|
"/api/internal/stripe_webhooks",
|
|
"/api/v1/authorization-code",
|
|
"/api/v1/token",
|
|
"/api/v1/usage/ingest",
|
|
"/api/v1/plain/customer-cards",
|
|
/^\/api\/v1\/tasks\/[^\/]+\/callback\/[^\/]+$/, // /api/v1/tasks/$id/callback/$secret
|
|
/^\/api\/v1\/runs\/[^\/]+\/tasks\/[^\/]+\/callback\/[^\/]+$/, // /api/v1/runs/$runId/tasks/$id/callback/$secret
|
|
/^\/api\/v1\/http-endpoints\/[^\/]+\/env\/[^\/]+\/[^\/]+$/, // /api/v1/http-endpoints/$httpEndpointId/env/$envType/$shortcode
|
|
/^\/api\/v1\/sources\/http\/[^\/]+$/, // /api/v1/sources/http/$id
|
|
/^\/api\/v1\/endpoints\/[^\/]+\/[^\/]+\/index\/[^\/]+$/, // /api/v1/endpoints/$environmentId/$endpointSlug/index/$indexHookIdentifier
|
|
"/api/v1/timezones",
|
|
"/api/v1/usage/ingest",
|
|
"/api/v1/auth/jwt/claims",
|
|
/^\/api\/v1\/runs\/[^\/]+\/attempts$/, // /api/v1/runs/$runFriendlyId/attempts
|
|
/^\/api\/v1\/waitpoints\/tokens\/[^\/]+\/callback\/[^\/]+$/, // /api/v1/waitpoints/tokens/$waitpointFriendlyId/callback/$hash
|
|
/^\/api\/v\d+\/deployments/, // /api/v{1,2,3,n}/deployments/*
|
|
// Internal SDK plumbing — packets are presigned-URL handshakes for
|
|
// payload uploads (v2 PUT) and downloads (v1 GET), authenticated via
|
|
// run-scoped JWT, called once per task/turn boundary by the runtime.
|
|
// Same shape as `/api/v1/runs/$runFriendlyId/attempts` above; not a
|
|
// customer-facing surface so customer rate limits shouldn't apply.
|
|
/^\/api\/v1\/packets\//,
|
|
/^\/api\/v2\/packets\//,
|
|
/^\/api\/v1\/sessions\/[^\/]+\/snapshot-url$/,
|
|
],
|
|
log: {
|
|
rejections: env.API_RATE_LIMIT_REJECTION_LOGS_ENABLED === "1",
|
|
requests: env.API_RATE_LIMIT_REQUEST_LOGS_ENABLED === "1",
|
|
limiter: env.API_RATE_LIMIT_LIMITER_LOGS_ENABLED === "1",
|
|
},
|
|
});
|
|
|
|
export type RateLimitMiddleware = ReturnType<typeof authorizationRateLimitMiddleware>;
|