Winston-rate-limiter NPM

winston-rate-limiter

a simple winston format which can be used to limit the amount of logs getting out from a system. It uses token bucket algorithm internally.

npm install winston-rate-limiter

const { rateLimitingFormat } = require("winston-rate-limiter").utils;

on a basic level this format can be added to a winston logger or to winston.format.combine
this format takes a key called config which has multiple optional keywords with key as log_level and value as a dict with below 3 keys

"DEBUG": {
    "tokensPerSec": 1,
    "startingTokens": 1, # optional, default=0
    "maxTokensBalance": 1 # optional, default=Infinity
}

tokensPerSec: the number of tokens(logs) which will be available for a given log level
startingTokens: the number of tokens already available at the start of the logging initialisation
maxTokensBalance: the maximum number of tokens that can be accumulated if the logs for a particular level are not being requested as much as the tokensPerSec allowed. kind of limited burst balance.
besides providing this config for particular log levels, you can supply a default key with this config.
if a default is supplied and there is no config supplied for a requested log level then this requested log level will be checked against this default config. useful for scenario where you want an overall rate shared across multiple log levels.
besides this an optional resultCallback can be supplied which will be called everytime a log record goes through format function
the resultCallback must accept 2 params result which is a boolean telling whether the log record was filtered or not and record the original log record.

build the docker-image of the package docker-compose build --no-cache
run the image docker-compose up
go to the container docker exec -it winstonformattester sh
to run tests first install jest npm install jest then jest to run the actual tests.

3 years ago

3 years ago