Skip to main content
This is a pre-production deployment of Warehouse. Changes made here affect the production instance of PyPI (
Help us improve Python packaging - Donate today!

A simple, fast, pure-python load balancer

Project Description

PumpkinLB is a fast multi-process TCP load balancer / port forwarder, compatible with: Linux, Cygwin, and Windows environments.


PumpkinLB listens for requests on ports local to the machine on which it is running, and farms them out to any number of workers.

You can use it to very quickly setup a load balancer, e.x. from 1 entry-point to 5 different apache workers on various servers.

Each incoming port is waited-on by a distinct process, and each connection is yet another process, thus it performs very well even under heavy load.

Requests are generally handled round-robin between the various workers. If a request fails on a backend worker, it will be retried on another random worker until it succeeds, and a message will be logged.


Execute by running [cfgFile]

Where [cfgFile] is the path to your config file. There is a sample “example.cfg” included.

Config Sections

The Config file is broken up into sections, definable by [$SectionName], followed by variables in format of key=value.


  • pre_resolve_workers=0/1 - Default 1

    Any workers defined with a hostname will be evaluated at the time the config is read.

    This is preferable as it saves a DNS trip for every request, and should be enabled

    unless your DNS is likely to change and you want the workers to match the change.

  • buffer_size=N - Default 4096

    Default read/write buffer size (in bytes) used on socket operations. 4096 is a good default for most, but you may be able to tune better depending on your application.


  • localaddr:inport=worker1:port,worker2:port…

    Listen on interface defined by “localaddr” on port “inport”. Farm out to worker addresses and ports.


  • inport=worker1:port,worker2:port…

    Listen on all interfaces on port “inport”, and farm out to worker addresses with given ports.

    Ex: 80=,

So an example to listen on port 80 localhost and farm out to 3 apache servers on your local subnet:


Graceful Shutdown

Sending SIGTERM, SIGINT, or pressing control+c will do a graceful shutdown (it will wait for up to 6 seconds to finish any active requests, and then terminate).

Release History

Release History

This version
History Node


History Node


History Node


History Node


History Node


History Node


Download Files

Download Files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

File Name & Checksum SHA256 Checksum Help Version File Type Upload Date
PumpkinLB-2.0.0.tar.gz (22.7 kB) Copy SHA256 Checksum SHA256 Source Jan 10, 2017

Supported By

WebFaction WebFaction Technical Writing Elastic Elastic Search Pingdom Pingdom Monitoring Dyn Dyn DNS Sentry Sentry Error Logging CloudAMQP CloudAMQP RabbitMQ Heroku Heroku PaaS Kabu Creative Kabu Creative UX & Design Fastly Fastly CDN DigiCert DigiCert EV Certificate Rackspace Rackspace Cloud Servers DreamHost DreamHost Log Hosting