Created: 2023-07-22 19:33
Status: #concept
Subject: Programming System Administration
Tags: Linux Operating System Bash Process Management BIOS Bootloader
Kernel
It is the core of the Operating System that manages Hardware and tells it how to interact with the system.
- it initializes devices & memory, then loads the init process.
$ uname -r # shows the kernel version
$ sudo apt dist-upgrade # upgrades our kernel
It knows:
- The status of the process
- The resources the process is using and receives
- The process owner
- Signal handling (more on that later)
- And basically everything else
/proc Directory
Everything in Linux is a file, even processes, which are found in
/proc.
- we can find
PIDprocess statuses in their/proc/<pid>/statusfile.
$ cat /proc/12345/status
Process States
We use
ps aux and check the STAT column for information on a process's state.
R: Running or runnable, it is just waiting for the CPU to process it.S: Interruptible sleep, waiting for an event to complete, such as input from the terminal.D: Uninterruptible sleep, processes that cannot be killed or interrupted with a signal, usually to make them go away you have torebootor fix the issue.Z: Zombie, terminated processes that are waiting to have their statuses collected by a parent.T: Stopped, a process that has been suspended/stopped.
Process Creation
When a new process is created, an existing process clones itself using the
fork System Call which creates a child process identical to the parent, but with a new Process ID (PID). When a system boots up, a persistent process called init is created (with the PID of
1) which spawns many child processes.
Process Termination
A process can manually exit and deallocate the memory it uses using the
_exit System Call along with a termination status code, similar to C's exit(0) function.
- A parent process uses the
waitSystem Call to figure out whether a child process has been killed.
Orphan Processes
When a parent process dies before a child process, the Kernel puts the children (now known as Orphan Processes) under the care of
init.
initwill not be responsible for calling thewaitSystem Call to figure out whether the child processes_exit.
Zombie Processes
When a child terminates and a parent process didn't call
wait to listen for the _exit System Call, the Kernel turns the child processes into Zombie Processes.
- the resources of the child are freed up, however there is still an entry in the Process Table for the zombie process.
- they cannot be killed with signals because they are already dead and have no resources to free.
- eventually, in a process called "reaping", when the parent calls the
waitSystem Call on the zombie process, the zombie will disappear