#!/bin/bashdeclare -A num wordword=([a]='index_a'[b]='index_b'[c]='index_c')num=([a]=1[b]=2[c]=3)declare -n var="$1"printf '%s\n' "${var[@]}"
bash array_call_self.sh word
Systemd is actually a fairly recent project, having been started in ~2010. Many others have been used before it: sysvinit, upstart, init-ng, s6, launchd (I think it had a Linux version once). Several Linux distributions still do not use systemd for various reasons.
Out of all those, systemd is the closest one that you even could call a "process manager". Linux doesn't actually rely on a process manager at all; any process can directly make (fork) a new child process whenever it decides to, uncontrolled. (Indeed so uncontrolled that when systemd started codifying things like "logging out should kill your leftover processes", wars were fought over it.)
The real primary job of systemd and other such programs is only to be a service manager: to automatically start certain specific processes, such as the daemon responsible for displaying the login screen (GDM, agetty) and – sometimes – to restart them when they crash. But once you get to the login screen, everything else can happen without the involvement of systemd/init.
Indeed with sysvinit, the actual init process did incredibly little: about the only service it oversaw was agetty. Literally everything else was done through ordinary shell scripts that never touched init proper. The command to start a service? Shell script. The whole initial boot process? Shell script that runs more shell scripts.
At a high level, Systemd typically serves several functions:
- Init system: Performs boot-time system initialization (e.g. creates or cleans temp files, etc.)
- Run services at boot (e.g. cron, sshd, mariadb, etc.)
- PID 1 process: The master process under which all other processes live
- Process supervisor (a.k.a. process manager): Starts, monitors, restarts additional services
- Additional services (Systemd does a lot, e.g. logging, mounts, and much more)
Most of these functions can be handled separately, but Systemd "does it all".
NAME
pam_unix - Module for traditional password authentication
SYNOPSIS
pam_unix.so [...]
DESCRIPTION
This is the standard Unix authentication module. It uses standard calls
from the system's libraries to retrieve and set account information as
well as authentication. Usually this is obtained from the /etc/passwd
and the /etc/shadow file as well if shadow is enabled.nullok
The default action of this module is to not permit the user access to a service if their official password is blank. The nullok argument overrides this default.
(if it's enabled, and the password is empty, the module doesn't even ask for the password but accepts the login directly.)
First, you don't log in to a shell, you log in to a (virtual) terminal.In current Ubuntu systems, getty is managed as a systemd service (ex. getty@tty1.service for the virtual terminal tty1), and login uses the Pluggable Authentication Modules (PAM) subsystem for authentication - in particular, the pam_unix module is "for traditional password authentication".
- a getty program opens a tty port, prompts for a login name and invokes a login program (/bin/login by default - although Linux's agetty may be configured to use a non-standard login program)
- the login program performs authentication and session setup, including initializing the environment and invoking the user's login shell.
Here are all the options for checking access/modify time:-amin n - the file was last accessed within n minutes ago.-atime n - the file was last accessed within n days ago.-cmin n - the file status changed within n minutes ago.-ctime n - the file status changed within n days ago.-mmin n - the file was last modified within n minutes ago.-mtime n - the file was last modified within n days ago.-anewer file - the searched file was modified more recently than [FILE]. If using -L or -H, and [FILE] is a symbolic link, then the file it points to is used.-cnewer file - the search for file status changed more recently than [FILE]. If using -L or -H, and [FILE] is a symbolic link, then the file it points to is used. If you add -daystart to the end of any of these, the measurement will be taken from the start of the current day, rather than the current time,
Finally, + and - modifications are also available. For instance, -amin -5 will find anything modified in the last 5 minutes, but -amin +5 will find anything modified more than 5 minutes ago.
-mmin on the other hand can count in minutes. So, if strict accuracy is vital, then -mmin +1440 ( 1440 minutes = 1 day ) could be used instead of -mtime +1
find . -name "*.txt" -mmin 5 -daystart
In other words, -mtime can count only in units of 24 hours or one day each so as far as -mtime +1 goes, this means exactly more than one day by at least one day ( ie. two days+ )Örnek
find "$DESDIR" -mtime +6 -type f -delete