Documentation< Previous | Contents | Next >
How A Process Works
When a system starts up, the kernel initiates a few of its own activities as processes and launches a program called init. init, in turn, runs a series of shell scripts (located in
/etc) called init scripts, which start all the system services. Many of these services are implemented as daemon programs, programs that just sit in the background and do their thing without having any user interface. So even if we are not logged in, the system is at least a little busy performing routine stuff.
The fact that a program can launch other programs is expressed in the process scheme as a parent process producing a child process.
The kernel maintains information about each process to help keep things organized. For example, each process is assigned a number called a process ID or PID. PIDs are as- signed in ascending order, with init always getting PID 1. The kernel also keeps track of the memory assigned to each process, as well as the processes' readiness to resume ex- ecution. Like files, processes also have owners and user IDs, effective user IDs, etc.