System Management on tplat

User Management

Mon, 01 Jan 0001 00:00:00 +0000

User Management

User management covers accounts, groups, passwords, and privilege elevation.

Overview

Users can be created, modified, grouped, and removed
sudo is the standard way to run a command with elevated privileges without logging in as root
Group membership is a common way to grant shared access

Reference

Command	Purpose
`sudo`	Run a command as another user, usually root
`su`	Switch user and start a shell
`useradd`	Create a user
`userdel`	Remove a user
`usermod`	Modify a user account
`addgroup`	Add a group
`delgroup`	Remove a group
`passwd`	Change a password

Mental Model

Users own access
Groups share access
sudo provides controlled privilege escalation

Package Management

Mon, 01 Jan 0001 00:00:00 +0000

Package Management

Package managers install software, resolve dependencies, and keep systems updated.

Overview

Packages bundle binaries, configuration, and dependency metadata
Package managers add install, remove, upgrade, and repository workflows
Debian-based systems commonly use dpkg and apt

Reference

Common tools

Command	Purpose
`dpkg`	Install and manage Debian package files
`apt`	High-level package manager for Debian-based systems
`aptitude`	Alternative front end for Debian package management
`snap`	Install and manage snap packages
`gem`	Manage Ruby packages
`pip`	Install Python packages
`git`	Distributed version control and source retrieval

APT

dpkg works with local .deb files
apt handles dependency resolution through repositories
/etc/apt/sources.list defines the repositories a system can query
apt-cache searches the local package cache

apt-cache search impacket

Mental Model

Use dpkg for package files and apt for dependency-aware installs
Repository metadata is as important as the package itself
Package managers trade manual effort for repeatable installation and updates

Service and Process Management

Mon, 01 Jan 0001 00:00:00 +0000

Service and Process Management

Services and processes are background workloads controlled with systemd, signals, and job control.

Overview

Services can be started, stopped, restarted, enabled, and inspected
Processes are identified by PIDs and may have parent processes
Job control and command chaining are common shell workflows

Reference

`systemctl`

start, stop, restart, and status control service state
enable makes a service start at boot
systemctl list-units --type=service shows active services

systemctl list-units --type=service

Process control

Common states: running, waiting, stopped, zombie
kill, pkill, pgrep, and killall send signals or locate processes
kill -l lists available signals

Job control

Ctrl+Z suspends the current job
jobs lists suspended or background jobs
bg resumes a job in the background
fg <id> returns a job to the foreground
& runs a command in the background immediately

Command chaining

; runs commands sequentially
&& runs the next command only on success
| sends stdout from one command to the next

Mental Model

systemd manages service lifecycle on most modern distributions
PIDs identify running processes and /proc exposes process details
Signals request state changes; jobs manage interactive shell work
Simple commands become more powerful when chained together

Task Scheduling

Mon, 01 Jan 0001 00:00:00 +0000

Task Scheduling

Task scheduling runs commands later, repeatedly, or at boot.

Overview

systemd timers are the native option on modern Linux systems
cron is a simple, time-based scheduler that remains widely used

Reference

`systemd` timers

A timer unit usually has a matching service unit
OnBootSec runs once after boot
OnUnitActiveSec repeats after the previous run
Reload unit files with systemctl daemon-reload

[Unit]
Description=My Timer

[Timer]
OnBootSec=3min
OnUnitActiveSec=1hour

[Install]
WantedBy=timers.target

[Unit]
Description=My Service

[Service]
ExecStart=/full/path/to/my/script.sh

[Install]
WantedBy=multi-user.target

Cron

Field	Values
Minute	`0-59`
Hour	`0-23`
Day of month	`1-31`
Month	`1-12`
Day of week	`0-7`

0 */6 * * * /path/to/update_software.sh
0 0 1 * * /path/to/scripts/run_scripts.sh
0 0 * * 0 /path/to/scripts/clean_database.sh
0 0 * * 7 /path/to/scripts/backup.sh

Mental Model

Use timers when you want native systemd integration
Use cron when you want compact recurring schedules
Prefer the simplest scheduler that fits the job

System Management on tplat

User Management

User Management

Overview

Reference

Mental Model

Package Management

Package Management

Overview

Reference

Common tools

APT

Mental Model

Service and Process Management

Service and Process Management

Overview

Reference

systemctl

Process control

Job control

Command chaining

Mental Model

Task Scheduling

Task Scheduling

Overview

Reference

systemd timers

Cron

Mental Model

`systemctl`

`systemd` timers