How it works

A single control process launches the number of child processes indicated by the NumServers directive at server startup. Each child process creates threads as specified in the StartThreads directive. The individual threads then listen for connections and serve them when they arrive.

Apache always tries to maintain a pool of spare or idle server threads, which stand ready to serve incoming requests. In this way, clients do not need to wait for new threads to be created. For each child process, Apache assesses the number of idle threads and creates or destroys threads to keep this number within the boundaries specified by MinSpareThreads and MaxSpareThreads. Since this process is very self-regulating, it is rarely necessary to modify these directives from their default values. The maximum number of clients that may be served simultaneously is determined by multiplying the number of server processes that will be created (NumServers) by the maximum number of threads created in each process (MaxThreadsPerChild).

While the parent process is usually started as root under Unix in order to bind to port 80, the child processes and threads are launched by Apache as a less-privileged user. The User and Group directives are used to set the privileges of the Apache child processes. The child processes must be able to read all the content that will be served, but should have as few privileges beyond that as possible. In addition, unless suexec is used, these directives also set the privileges which will be inherited by CGI scripts.

MaxRequestsPerChild controls how frequently the server recycles processes by killing old ones and launching new ones.

Working with different user-IDs

The perchild MPM adds the extra ability to specify that particular processes should serve requests under different user-IDs. These user-IDs can then be associated with specific virtual hosts. You have to use one ChildPerUserID directive for every user/group combination you want to be run. Then you can tie particular virtual hosts to that user and group IDs.

The following example runs 7 child processes. Two of them are run under user1/group1. The next four are run under user2/group2 and the remaining process uses the User and Group of the main server:

Using unbalanced numbers of processes as above is useful, if the particular virtual hosts produce different load. The assignment to the virtual hosts is easily done as in the example below. In conclusion with the example above the following assumes, that server2 has to serve about twice of the hits of server1.