feat: Add request rate limiter based on IP address (#8174)

This commit is contained in:
Daniel
2023-01-06 23:39:02 +11:00
committed by GitHub
parent 0eac5dc6d4
commit 6c79f6a69e
13 changed files with 713 additions and 50 deletions

View File

@@ -292,6 +292,29 @@ export interface ParseServerOptions {
/* An array of keys and values that are prohibited in database read and write requests to prevent potential security vulnerabilities. It is possible to specify only a key (`{"key":"..."}`), only a value (`{"value":"..."}`) or a key-value pair (`{"key":"...","value":"..."}`). The specification can use the following types: `boolean`, `numeric` or `string`, where `string` will be interpreted as a regex notation. Request data is deep-scanned for matching definitions to detect also any nested occurrences. Defaults are patterns that are likely to be used in malicious requests. Setting this option will override the default patterns.
:DEFAULT: [{"key":"_bsontype","value":"Code"},{"key":"constructor"},{"key":"__proto__"}] */
requestKeywordDenylist: ?(RequestKeywordDenylist[]);
/* Options to limit repeated requests to Parse Server APIs. This can be used to protect sensitive endpoints such as `/requestPasswordReset` from brute-force attacks or Parse Server as a whole from denial-of-service (DoS) attacks.<br><br> Mind the following limitations:<br>- rate limits applied per IP address; this limits protection against distributed denial-of-service (DDoS) attacks where many requests are coming from various IP addresses<br>- if multiple Parse Server instances are behind a load balancer or ran in a cluster, each instance will calculate it's own request rates, independent from other instances; this limits the applicability of this feature when using a load balancer and another rate limiting solution that takes requests across all instances into account may be more suitable<br>- this feature provides basic protection against denial-of-service attacks, but a more sophisticated solution works earlier in the request flow and prevents a malicious requests to even reach a server instance; it's therefore recommended to implement a solution according to architecture and user case.
:DEFAULT: [] */
rateLimit: ?(RateLimitOptions[]);
}
export interface RateLimitOptions {
/* The path of the API route to be rate limited. Route paths, in combination with a request method, define the endpoints at which requests can be made. Route paths can be strings, string patterns, or regular expression. See: https://expressjs.com/en/guide/routing.html */
requestPath: string;
/* The window of time in milliseconds within which the number of requests set in `requestCount` can be made before the rate limit is applied. */
requestTimeWindow: ?number;
/* The number of requests that can be made per IP address within the time window set in `requestTimeWindow` before the rate limit is applied. */
requestCount: ?number;
/* The error message that should be returned in the body of the HTTP 429 response when the rate limit is hit. Default is `Too many requests.`.
:DEFAULT: Too many requests. */
errorResponseMessage: ?string;
/* Optional, the HTTP request methods to which the rate limit should be applied, default is all methods. */
requestMethods: ?(string[]);
/* Optional, if `true` the rate limit will also apply to requests using the `masterKey`, default is `false`. Note that a public Cloud Code function that triggers internal requests using the `masterKey` may circumvent rate limiting and be vulnerable to attacks.
:DEFAULT: false */
includeMasterKey: ?boolean;
/* Optional, if `true` the rate limit will also apply to requests that are made in by Cloud Code, default is `false`. Note that a public Cloud Code function that triggers internal requests may circumvent rate limiting and be vulnerable to attacks.
:DEFAULT: false */
includeInternalRequests: ?boolean;
}
export interface SecurityOptions {