config.limit.yml Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of rate-limit Show documentation
Show all versions of rate-limit Show documentation
A handler that does rate limit in order to prevent attacks for public facing services.
---
# Rate Limit Handler Configuration
# If this handler is enabled or not. It is disabled by default as this handle might be in
# most http-sidecar, light-proxy and light-router instances. However, it should only be used
# internally to throttle request for a slow backend service or externally for DDoS attacks.
enabled: ${limit.enabled:false}
# Maximum concurrent requests allowed per second on the entire server. This is property is
# here to keep backward compatible. New users should use the rateLimit property for config
# with different keys and different time unit.
concurrentRequest: ${limit.concurrentRequest:2}
# This property is kept to ensure backward compatibility. Please don't use it anymore. All
# requests will return the rate limit headers with error messages after the limit is reached.
queueSize: ${limit.queueSize:-1}
# If the rate limit is exposed to the Internet to prevent DDoS attacks, it will return 503
# error code to trick the DDoS client/tool to stop the attacks as it considers the server
# is down. However, if the rate limit is used internally to throttle the client requests to
# protect a slow backend API, it will return 429 error code to indicate too many requests
# for the client to wait a grace period to resent the request. By default, 429 is returned.
errorCode: ${limit.errorCode:429}
# Default request rate limit 10 requests per second and 10000 quota per day. This is the
# default for the server shared by all the services. If the key is not server, then the
# quota is not applicable.
# 10 requests per second limit and 10000 requests per day quota.
rateLimit: ${limit.rateLimit:10/s 10000/d}
# Key of the rate limit: server, address, client, user
# server: The entire server has one rate limit key, and it means all users share the same.
# address: The IP address is the key and each IP will have its rate limit configuration.
# client: The client id in the JWT token so that we can give rate limit per client.
# user: The user id in the JWT token so that we can set rate limit and quota based on user.
key: ${limit.key:server}
# If server is the key, we can set up different rate limit per request path prefix.
server: ${limit.server:}
# If address is the key, we can set up different rate limit per address and optional per
# path or service for certain addresses. All other un-specified addresses will share the
# limit defined in rateLimit.
address: ${limit.address:}
# If client is the key, we can set up different rate limit per client and optional per
# path or service for certain clients. All other un-specified clients will share the limit
# defined in rateLimit. When client is select, the rate-limit handler must be after the
## JwtVerifierHandler so that the client_id can be retrieved from the auditInfo attachment.
client: ${limit.client:}
# If user is the key, we can set up different rate limit per user and optional per
# path or service for certain users. All other un-specified users will share the limit
# defined in rateLimit. When user is select, the rate-limit handler must be after the
# JwtVerifierHandler so that the user_id can be retrieved from the auditInfo attachment.
user: ${limit.user:}
# Client id Key Resolver.
clientIdKeyResolver: ${limit.clientIdKeyResolver:com.networknt.limit.key.JwtClientIdKeyResolver}
# Ip Address Key Resolver.
addressKeyResolver: ${limit.addressKeyResolver:com.networknt.limit.key.RemoteAddressKeyResolver}
# User Id Key Resolver.
userIdKeyResolver: ${limit.userIdKeyResolver:com.networknt.limit.key.JwtUserIdKeyResolver}