On March 22, 2011 12:21 , Tushar Chavan
<tushar.chavan@xxxxxxxxxxx> wrote:
The answer to these questions is determined by which Multi-Processing Module (MPM) you choose to use with Apache HTTP Server. The prefork MPM is available for Unix-like systems and creates a process to handle each request. The StartServer, MinSpareServer, MaxSpareServer, and MaxClient directives control how many processes get started. To find out the default values of each of these directives, see the documentation: http://httpd.apache.org/docs/2.2/mod/prefork.html The worker MPM is available for Unix-like systems and uses multiple children each with multiple threads; a single child can handle multiple requests simultaneously, and each request is handled by a separate thread. The StartServers, MinSpareThreads, MaxSpareThreads, MaxClients, and ThreadsPerChild directives control how many processes and threads get started. More information (including default values for these directives) is available in the documentation: http://httpd.apache.org/docs/2.2/mod/worker.html There are other MPMs, too, including mpm_winnt for Microsoft Windows servers, mpm_netware for Netware servers, and others. For a complete list and more information, see http://httpd.apache.org/docs/2.2/mpm.html
The purpose of the scoreboard is to keep track of the status of each process or thread, and allow the multiple processes and threads to communicate and coordinate with each other. There happen to be fields in the scoreboard data structure for the maximum number of processes and maximum number of threads, but, in general, all configuration directives -- including all of the ones mentioned above -- are stored in Apache HTTP Server's various configuration data structures. The details of how these are stored are usually only important if you're writing your own third-party modules, or contributing to the Apache HTTP Server code. -- Mark Montague mark@xxxxxxxxxxx |
|