https://github.com/frmdstryr/c-webserver

A very simple webserver that uses a thread pool to asynchronously handle requests.
https://github.com/frmdstryr/c-webserver

Last synced: about 1 month ago
JSON representation

A very simple webserver that uses a thread pool to asynchronously handle requests.

• Host: GitHub
• URL: https://github.com/frmdstryr/c-webserver
• Owner: frmdstryr
• Created: 2014-04-09T20:37:35.000Z (almost 11 years ago)
• Default Branch: master
• Last Pushed: 2014-04-09T20:42:17.000Z (almost 11 years ago)
• Last Synced: 2024-10-15T18:26:44.576Z (3 months ago)
• Size: 2.09 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readme.md

Awesome Lists containing this project

README

        
>

> readme.md

>

>  Created on: April 9, 2014

>      Author: jairus martin

>

## Description of program (what it does) ##

- A simple webserver using threads to handle requests.

## How to compile ##

make

##  How to run ##

./webserver 

Files to be served must be put in:

getcwd()/public/

## Design overview ##

When the server is started it creates a thread pool allowing 50

simultaneous connections.  Then the server binds to the port given

and sit's in a loop waiting to accept connections.

When a new connection is made, the server creates a client object 

with the socket and puts a task containing the client and handler function 

(ie client_main) in the thread pool's queue. The queue is protected

by a mutex so only one thread can access it at a time. The thread pool has

a background thread running that waits for tasks in the queue, and if 

one is ready, it waits for a thread to be available (ie threads_available 

semaphore >0), and then it creates a thread to run the task. Each thread 

is spawned using a kernel level thread so they can be truely asyncronous. 

The thread pool's scheduler thread also checks if each thread is still

alive or not and join's any dead threads making another thread available

be used by the pool.

The client_main function reads the request from the socket and

serves the requested file, then closes the socket and frees memory

used.

## How you may have handled any ambiguities in the specification ##

## Any known bugs or problems ##

- There's a memory leak somewhere, every request uses ~24kb...